From ykornblum at msn.com Sun Dec 1 19:35:07 2013 From: ykornblum at msn.com (YOEL) Date: Sun, 1 Dec 2013 19:35:07 -0600 Subject: [tclug-list] linux on a virtual box Message-ID: I am trying to install linux under virtual box (VM) I downloaded from Oracle.I got difficulties actually getting linux to that VM.I got Windows 8 on my computer and in the past I has a dual boot which was not very convenient Does anyone knows where I can download a VM with linux or any other method would be appreciated. Thansk,Yuil Kornblum -------------- next part -------------- An HTML attachment was scrubbed... URL: From tclug at freakzilla.com Sun Dec 1 19:39:16 2013 From: tclug at freakzilla.com (tclug at freakzilla.com) Date: Sun, 1 Dec 2013 19:39:16 -0600 (CST) Subject: [tclug-list] linux on a virtual box In-Reply-To: References: Message-ID: Can you tell us what problems you are actually having? Basically you have to download a linux distro ISO, set up a VM and tell it ot use that ISO as the boot CD, and just go form there. If you tell Virtualbox you're installing Linux it should come up with reasonable defaults. On Sun, 1 Dec 2013, YOEL wrote: > I am trying to install linux under virtual box (VM) I downloaded from > Oracle.I got difficulties actually getting linux to that VM. > I got Windows 8 on my computer and in the past I has a dual boot which was > not very convenient? > Does anyone knows where I can download a VM with linux or any other method > would be appreciated. > > Thansk, > Yuil Kornblum > > From chewie at wookimus.net Sun Dec 1 23:47:32 2013 From: chewie at wookimus.net (Chad Walstrom) Date: Sun, 01 Dec 2013 23:47:32 -0600 Subject: [tclug-list] linux on a virtual box In-Reply-To: References: Message-ID: <871u1vn6rv.wl%chewie@wookimus.net> At Sun, 1 Dec 2013 19:35:07 -0600, YOEL wrote: > I am trying to install linux under virtual box (VM) I downloaded > from Oracle. I got difficulties actually getting linux to that VM. > I got Windows 8 on my computer and in the past I has a dual boot > which was not very convenient Does anyone knows where I can download > a VM with linux or any other method would be appreciated. Try the vagrant build for XE. https://github.com/hilverd/vagrant-ubuntu-oracle-xe -- Chad From john.a.frisk at gmail.com Mon Dec 2 12:59:58 2013 From: john.a.frisk at gmail.com (John Frisk) Date: Mon, 2 Dec 2013 12:59:58 -0600 Subject: [tclug-list] Festive get together Message-ID: Hello everyone, I am going to suggest a get together Tuesday December 10th, 7-9M (or as long as there is good conversation) at Stouts right by TIES. Nothing formal to meet about but thought it would be nice to meet outside the normal meetings. Thoughts? -------------- next part -------------- An HTML attachment was scrubbed... URL: From ryanjcole at me.com Mon Dec 2 13:21:49 2013 From: ryanjcole at me.com (Ryan Coleman) Date: Mon, 02 Dec 2013 13:21:49 -0600 Subject: [tclug-list] Festive get together In-Reply-To: References: Message-ID: Sounds great, I?d go but that?s probably out of the realm of possibilities for me. On Dec 2, 2013, at 12:59 PM, John Frisk wrote: > Hello everyone, > I am going to suggest a get together Tuesday December 10th, 7-9M (or as long as there is good conversation) at Stouts right by TIES. Nothing formal to meet about but thought it would be nice to meet outside the normal meetings. Thoughts? > > _______________________________________________ > TCLUG Mailing List - Minneapolis/St. Paul, Minnesota > tclug-list at mn-linux.org > http://mailman.mn-linux.org/mailman/listinfo/tclug-list -------------- next part -------------- An HTML attachment was scrubbed... URL: From jeremy.mountainjohnson at gmail.com Mon Dec 2 13:39:06 2013 From: jeremy.mountainjohnson at gmail.com (Jeremy MountainJohnson) Date: Mon, 2 Dec 2013 13:39:06 -0600 Subject: [tclug-list] Festive get together In-Reply-To: References: Message-ID: I would be interested as well. Weekdays work better than the weekends for me, wish I could attend more Saturday meetings at TIES :-/ -- Jeremy MountainJohnson Jeremy.MountainJohnson at gmail.com On Mon, Dec 2, 2013 at 1:21 PM, Ryan Coleman wrote: > Sounds great, I?d go but that?s probably out of the realm of possibilities > for me. > > On Dec 2, 2013, at 12:59 PM, John Frisk wrote: > > Hello everyone, > I am going to suggest a get together Tuesday December 10th, 7-9M (or as long > as there is good conversation) at Stouts right by TIES. Nothing formal to > meet about but thought it would be nice to meet outside the normal meetings. > Thoughts? > > _______________________________________________ > TCLUG Mailing List - Minneapolis/St. Paul, Minnesota > tclug-list at mn-linux.org > http://mailman.mn-linux.org/mailman/listinfo/tclug-list > > > > _______________________________________________ > TCLUG Mailing List - Minneapolis/St. Paul, Minnesota > tclug-list at mn-linux.org > http://mailman.mn-linux.org/mailman/listinfo/tclug-list > From jmore at starmind.org Mon Dec 2 13:43:22 2013 From: jmore at starmind.org (Josh More) Date: Mon, 2 Dec 2013 13:43:22 -0600 Subject: [tclug-list] Festive get together In-Reply-To: References: Message-ID: I would also be interested. I've tentatively blocked off the evening. Once it's official, let me know what the times are so I can narrow things down. -Josh On Mon, Dec 2, 2013 at 1:39 PM, Jeremy MountainJohnson < jeremy.mountainjohnson at gmail.com> wrote: > I would be interested as well. Weekdays work better than the weekends > for me, wish I could attend more Saturday meetings at TIES :-/ > -- > Jeremy MountainJohnson > Jeremy.MountainJohnson at gmail.com > > > On Mon, Dec 2, 2013 at 1:21 PM, Ryan Coleman wrote: > > Sounds great, I?d go but that?s probably out of the realm of > possibilities > > for me. > > > > On Dec 2, 2013, at 12:59 PM, John Frisk wrote: > > > > Hello everyone, > > I am going to suggest a get together Tuesday December 10th, 7-9M (or as > long > > as there is good conversation) at Stouts right by TIES. Nothing formal > to > > meet about but thought it would be nice to meet outside the normal > meetings. > > Thoughts? > > > > _______________________________________________ > > TCLUG Mailing List - Minneapolis/St. Paul, Minnesota > > tclug-list at mn-linux.org > > http://mailman.mn-linux.org/mailman/listinfo/tclug-list > > > > > > > > _______________________________________________ > > TCLUG Mailing List - Minneapolis/St. Paul, Minnesota > > tclug-list at mn-linux.org > > http://mailman.mn-linux.org/mailman/listinfo/tclug-list > > > _______________________________________________ > TCLUG Mailing List - Minneapolis/St. Paul, Minnesota > tclug-list at mn-linux.org > http://mailman.mn-linux.org/mailman/listinfo/tclug-list > -------------- next part -------------- An HTML attachment was scrubbed... URL: From admin at lctn.org Mon Dec 2 19:45:01 2013 From: admin at lctn.org (Raymond Norton) Date: Mon, 02 Dec 2013 19:45:01 -0600 Subject: [tclug-list] No TTS voices available for Chrome Speak Message-ID: <529D379D.8050608@lctn.org> Amazingly, I'm not finding info on google for this: I have Google Chrome installed on two boxes (Mint 14). On my desktop I can choose espeak voices for read back of text. Archaic, but it works. However, no voices are available for selection on my laptop. Both have espeak installed. Anyone know of a fix, or a way to add quality voices to Google Chrome? Raymond From ryanjcole at me.com Tue Dec 3 10:12:33 2013 From: ryanjcole at me.com (Ryan Coleman) Date: Tue, 03 Dec 2013 10:12:33 -0600 Subject: [tclug-list] Ubuntu box reporting free space incorrectly Message-ID: <1E767F30-54C7-4B11-BC7B-963EE3703E7C@me.com> I have a customer whose Ubuntu box (which we have limited ability to modify due to its primary function) is reporting wrong information for free disk space. I have a script that runs every hourly that clears out a specific log file but doesn?t update the available inodes in the file table. Is there a way to reset this without rebooting the server every 60 minutes? I don?t have the option to change file system types, partition sizes, I don?t have direct access to the hardware (it?s in the EU) and if I bring it offline I do not have a ready replacement ready to go (but the customer is ?working on it?). Thanks! From erikerik at gmail.com Tue Dec 3 10:20:00 2013 From: erikerik at gmail.com (Erik Anderson) Date: Tue, 3 Dec 2013 10:20:00 -0600 Subject: [tclug-list] Ubuntu box reporting free space incorrectly In-Reply-To: <1E767F30-54C7-4B11-BC7B-963EE3703E7C@me.com> References: <1E767F30-54C7-4B11-BC7B-963EE3703E7C@me.com> Message-ID: On Tue, Dec 3, 2013 at 10:12 AM, Ryan Coleman wrote: > Is there a way to reset this without rebooting the server every 60 minutes? A umount/mount cycle on that fs will typically clear out this sort of thing. *If* you keep your logs on a separate fs (which everyone *should* be doing) it's faster than a reboot, but obviously not ideal, as it may require a bit of downtime each cycle. -------------- next part -------------- An HTML attachment was scrubbed... URL: From ryanjcole at me.com Tue Dec 3 10:26:19 2013 From: ryanjcole at me.com (Ryan Coleman) Date: Tue, 03 Dec 2013 10:26:19 -0600 Subject: [tclug-list] Ubuntu box reporting free space incorrectly In-Reply-To: References: <1E767F30-54C7-4B11-BC7B-963EE3703E7C@me.com> Message-ID: <6424DC37-A2A5-4634-8600-FECA396E13D1@me.com> Sadly I don?t and I haven?t been able to figure out how to make this recognize that - but the only other partitions I have at my disposal are in use by critical tasks for this particular machine so I would have to have a series of stop/starts that would make it equivalent of rebooting anyway (the machine comes back online in 10 seconds but the services required take about a minute to load) On Dec 3, 2013, at 10:20 AM, Erik Anderson wrote: > On Tue, Dec 3, 2013 at 10:12 AM, Ryan Coleman wrote: > Is there a way to reset this without rebooting the server every 60 minutes? > > A umount/mount cycle on that fs will typically clear out this sort of thing. *If* you keep your logs on a separate fs (which everyone *should* be doing) it's faster than a reboot, but obviously not ideal, as it may require a bit of downtime each cycle. > > _______________________________________________ > TCLUG Mailing List - Minneapolis/St. Paul, Minnesota > tclug-list at mn-linux.org > http://mailman.mn-linux.org/mailman/listinfo/tclug-list -------------- next part -------------- An HTML attachment was scrubbed... URL: From sfertch at gmail.com Tue Dec 3 10:27:53 2013 From: sfertch at gmail.com (Shawn Fertch) Date: Tue, 3 Dec 2013 10:27:53 -0600 Subject: [tclug-list] Ubuntu box reporting free space incorrectly In-Reply-To: References: <1E767F30-54C7-4B11-BC7B-963EE3703E7C@me.com> Message-ID: This is more of a hack solution, and doesn't do anything to resolve the real issue. It sounds more like the file is open, and being deleted, which doesn't remove the inodes. How is the log file being cleaned out? Zeroing it, logrotate, or some other method? Does the application need to be bounced briefly, or would simply zeroing the file do the trick? On Tue, Dec 3, 2013 at 10:20 AM, Erik Anderson wrote: > On Tue, Dec 3, 2013 at 10:12 AM, Ryan Coleman wrote: > >> Is there a way to reset this without rebooting the server every 60 >> minutes? > > > A umount/mount cycle on that fs will typically clear out this sort of > thing. *If* you keep your logs on a separate fs (which everyone *should* be > doing) it's faster than a reboot, but obviously not ideal, as it may > require a bit of downtime each cycle. > > > _______________________________________________ > TCLUG Mailing List - Minneapolis/St. Paul, Minnesota > tclug-list at mn-linux.org > http://mailman.mn-linux.org/mailman/listinfo/tclug-list > > -- -Shawn -------------- next part -------------- An HTML attachment was scrubbed... URL: From ryanjcole at me.com Tue Dec 3 10:29:53 2013 From: ryanjcole at me.com (Ryan Coleman) Date: Tue, 03 Dec 2013 10:29:53 -0600 Subject: [tclug-list] Ubuntu box reporting free space incorrectly In-Reply-To: References: <1E767F30-54C7-4B11-BC7B-963EE3703E7C@me.com> Message-ID: <380826A6-FADE-4080-B91D-17F91F4A148F@me.com> It was a hack attempt in the first place - a simple delete? but I could add restart of apache into the mix? maybe that will resolve it? On Dec 3, 2013, at 10:27 AM, Shawn Fertch wrote: > This is more of a hack solution, and doesn't do anything to resolve the real issue. It sounds more like the file is open, and being deleted, which doesn't remove the inodes. > > How is the log file being cleaned out? Zeroing it, logrotate, or some other method? Does the application need to be bounced briefly, or would simply zeroing the file do the trick? > > > On Tue, Dec 3, 2013 at 10:20 AM, Erik Anderson wrote: > On Tue, Dec 3, 2013 at 10:12 AM, Ryan Coleman wrote: > Is there a way to reset this without rebooting the server every 60 minutes? > > A umount/mount cycle on that fs will typically clear out this sort of thing. *If* you keep your logs on a separate fs (which everyone *should* be doing) it's faster than a reboot, but obviously not ideal, as it may require a bit of downtime each cycle. > > > _______________________________________________ > TCLUG Mailing List - Minneapolis/St. Paul, Minnesota > tclug-list at mn-linux.org > http://mailman.mn-linux.org/mailman/listinfo/tclug-list > > > > > -- > -Shawn > > > _______________________________________________ > TCLUG Mailing List - Minneapolis/St. Paul, Minnesota > tclug-list at mn-linux.org > http://mailman.mn-linux.org/mailman/listinfo/tclug-list -------------- next part -------------- An HTML attachment was scrubbed... URL: From erikerik at gmail.com Tue Dec 3 10:32:00 2013 From: erikerik at gmail.com (Erik Anderson) Date: Tue, 3 Dec 2013 10:32:00 -0600 Subject: [tclug-list] Ubuntu box reporting free space incorrectly In-Reply-To: <380826A6-FADE-4080-B91D-17F91F4A148F@me.com> References: <1E767F30-54C7-4B11-BC7B-963EE3703E7C@me.com> <380826A6-FADE-4080-B91D-17F91F4A148F@me.com> Message-ID: On Tue, Dec 3, 2013 at 10:29 AM, Ryan Coleman wrote: > It was a hack attempt in the first place - a simple delete? but I could > add restart of apache into the mix? maybe that will resolve it? Is this something that logrotate could handle for you? -------------- next part -------------- An HTML attachment was scrubbed... URL: From sfertch at gmail.com Tue Dec 3 10:32:25 2013 From: sfertch at gmail.com (Shawn Fertch) Date: Tue, 3 Dec 2013 10:32:25 -0600 Subject: [tclug-list] Ubuntu box reporting free space incorrectly In-Reply-To: <380826A6-FADE-4080-B91D-17F91F4A148F@me.com> References: <1E767F30-54C7-4B11-BC7B-963EE3703E7C@me.com> <380826A6-FADE-4080-B91D-17F91F4A148F@me.com> Message-ID: You can't delete an open file and expect the size to be freed up. You either need to simply zero the file by catting /dev/null into it (not recommended for historical purposes), or rotate the logs. Look into logrotate, it'll resolve your problems. On Tue, Dec 3, 2013 at 10:29 AM, Ryan Coleman wrote: > It was a hack attempt in the first place - a simple delete? but I could > add restart of apache into the mix? maybe that will resolve it? > > On Dec 3, 2013, at 10:27 AM, Shawn Fertch wrote: > > This is more of a hack solution, and doesn't do anything to resolve the > real issue. It sounds more like the file is open, and being deleted, which > doesn't remove the inodes. > > How is the log file being cleaned out? Zeroing it, logrotate, or some > other method? Does the application need to be bounced briefly, or would > simply zeroing the file do the trick? > > > On Tue, Dec 3, 2013 at 10:20 AM, Erik Anderson wrote: > >> On Tue, Dec 3, 2013 at 10:12 AM, Ryan Coleman wrote: >> >>> Is there a way to reset this without rebooting the server every 60 >>> minutes? >> >> >> A umount/mount cycle on that fs will typically clear out this sort of >> thing. *If* you keep your logs on a separate fs (which everyone *should* be >> doing) it's faster than a reboot, but obviously not ideal, as it may >> require a bit of downtime each cycle. >> >> >> _______________________________________________ >> TCLUG Mailing List - Minneapolis/St. Paul, Minnesota >> tclug-list at mn-linux.org >> http://mailman.mn-linux.org/mailman/listinfo/tclug-list >> >> > > > -- > -Shawn > > > _______________________________________________ > TCLUG Mailing List - Minneapolis/St. Paul, Minnesota > tclug-list at mn-linux.org > http://mailman.mn-linux.org/mailman/listinfo/tclug-list > > > > _______________________________________________ > TCLUG Mailing List - Minneapolis/St. Paul, Minnesota > tclug-list at mn-linux.org > http://mailman.mn-linux.org/mailman/listinfo/tclug-list > > -- -Shawn -------------- next part -------------- An HTML attachment was scrubbed... URL: From jake.vath at gmail.com Tue Dec 3 10:33:20 2013 From: jake.vath at gmail.com (Jake Vath) Date: Tue, 3 Dec 2013 10:33:20 -0600 Subject: [tclug-list] Ubuntu box reporting free space incorrectly In-Reply-To: <380826A6-FADE-4080-B91D-17F91F4A148F@me.com> References: <1E767F30-54C7-4B11-BC7B-963EE3703E7C@me.com> <380826A6-FADE-4080-B91D-17F91F4A148F@me.com> Message-ID: So, it's an Apache log file? -> Jake On Tue, Dec 3, 2013 at 10:29 AM, Ryan Coleman wrote: > It was a hack attempt in the first place - a simple delete? but I could > add restart of apache into the mix? maybe that will resolve it? > > On Dec 3, 2013, at 10:27 AM, Shawn Fertch wrote: > > This is more of a hack solution, and doesn't do anything to resolve the > real issue. It sounds more like the file is open, and being deleted, which > doesn't remove the inodes. > > How is the log file being cleaned out? Zeroing it, logrotate, or some > other method? Does the application need to be bounced briefly, or would > simply zeroing the file do the trick? > > > On Tue, Dec 3, 2013 at 10:20 AM, Erik Anderson wrote: > >> On Tue, Dec 3, 2013 at 10:12 AM, Ryan Coleman wrote: >> >>> Is there a way to reset this without rebooting the server every 60 >>> minutes? >> >> >> A umount/mount cycle on that fs will typically clear out this sort of >> thing. *If* you keep your logs on a separate fs (which everyone *should* be >> doing) it's faster than a reboot, but obviously not ideal, as it may >> require a bit of downtime each cycle. >> >> >> _______________________________________________ >> TCLUG Mailing List - Minneapolis/St. Paul, Minnesota >> tclug-list at mn-linux.org >> http://mailman.mn-linux.org/mailman/listinfo/tclug-list >> >> > > > -- > -Shawn > > > _______________________________________________ > TCLUG Mailing List - Minneapolis/St. Paul, Minnesota > tclug-list at mn-linux.org > http://mailman.mn-linux.org/mailman/listinfo/tclug-list > > > > _______________________________________________ > TCLUG Mailing List - Minneapolis/St. Paul, Minnesota > tclug-list at mn-linux.org > http://mailman.mn-linux.org/mailman/listinfo/tclug-list > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From jeremy.mountainjohnson at gmail.com Tue Dec 3 10:36:00 2013 From: jeremy.mountainjohnson at gmail.com (Jeremy MountainJohnson) Date: Tue, 3 Dec 2013 10:36:00 -0600 Subject: [tclug-list] Ubuntu box reporting free space incorrectly In-Reply-To: References: <1E767F30-54C7-4B11-BC7B-963EE3703E7C@me.com> <380826A6-FADE-4080-B91D-17F91F4A148F@me.com> Message-ID: wipe and srm are possibilities too; although I would think there would be a better solution to this. -- Jeremy MountainJohnson Jeremy.MountainJohnson at gmail.com On Tue, Dec 3, 2013 at 10:32 AM, Shawn Fertch wrote: > You can't delete an open file and expect the size to be freed up. You > either need to simply zero the file by catting /dev/null into it (not > recommended for historical purposes), or rotate the logs. Look into > logrotate, it'll resolve your problems. > > > On Tue, Dec 3, 2013 at 10:29 AM, Ryan Coleman wrote: >> >> It was a hack attempt in the first place - a simple delete? but I could >> add restart of apache into the mix? maybe that will resolve it? >> >> On Dec 3, 2013, at 10:27 AM, Shawn Fertch wrote: >> >> This is more of a hack solution, and doesn't do anything to resolve the >> real issue. It sounds more like the file is open, and being deleted, which >> doesn't remove the inodes. >> >> How is the log file being cleaned out? Zeroing it, logrotate, or some >> other method? Does the application need to be bounced briefly, or would >> simply zeroing the file do the trick? >> >> >> On Tue, Dec 3, 2013 at 10:20 AM, Erik Anderson wrote: >>> >>> On Tue, Dec 3, 2013 at 10:12 AM, Ryan Coleman wrote: >>>> >>>> Is there a way to reset this without rebooting the server every 60 >>>> minutes? >>> >>> >>> A umount/mount cycle on that fs will typically clear out this sort of >>> thing. *If* you keep your logs on a separate fs (which everyone *should* be >>> doing) it's faster than a reboot, but obviously not ideal, as it may require >>> a bit of downtime each cycle. >>> >>> >>> _______________________________________________ >>> TCLUG Mailing List - Minneapolis/St. Paul, Minnesota >>> tclug-list at mn-linux.org >>> http://mailman.mn-linux.org/mailman/listinfo/tclug-list >>> >> >> >> >> -- >> -Shawn >> >> >> _______________________________________________ >> TCLUG Mailing List - Minneapolis/St. Paul, Minnesota >> tclug-list at mn-linux.org >> http://mailman.mn-linux.org/mailman/listinfo/tclug-list >> >> >> >> _______________________________________________ >> TCLUG Mailing List - Minneapolis/St. Paul, Minnesota >> tclug-list at mn-linux.org >> http://mailman.mn-linux.org/mailman/listinfo/tclug-list >> > > > > -- > -Shawn > > > > _______________________________________________ > TCLUG Mailing List - Minneapolis/St. Paul, Minnesota > tclug-list at mn-linux.org > http://mailman.mn-linux.org/mailman/listinfo/tclug-list > From ryanjcole at me.com Tue Dec 3 10:44:46 2013 From: ryanjcole at me.com (Ryan Coleman) Date: Tue, 03 Dec 2013 10:44:46 -0600 Subject: [tclug-list] Ubuntu box reporting free space incorrectly In-Reply-To: References: <1E767F30-54C7-4B11-BC7B-963EE3703E7C@me.com> <380826A6-FADE-4080-B91D-17F91F4A148F@me.com> Message-ID: <6864D609-4BE4-4A97-9BC8-B2A258B7FB3F@me.com> Yes, being filled with a whole bunch of LDAP debug information. I don?t have the time in the project to go debug it for a temporary deployment so wiping and moving forward is the easiest to proceed. On Dec 3, 2013, at 10:33 AM, Jake Vath wrote: > So, it's an Apache log file? > > -> Jake > > > On Tue, Dec 3, 2013 at 10:29 AM, Ryan Coleman wrote: > It was a hack attempt in the first place - a simple delete? but I could add restart of apache into the mix? maybe that will resolve it? > > On Dec 3, 2013, at 10:27 AM, Shawn Fertch wrote: > >> This is more of a hack solution, and doesn't do anything to resolve the real issue. It sounds more like the file is open, and being deleted, which doesn't remove the inodes. >> >> How is the log file being cleaned out? Zeroing it, logrotate, or some other method? Does the application need to be bounced briefly, or would simply zeroing the file do the trick? >> >> >> On Tue, Dec 3, 2013 at 10:20 AM, Erik Anderson wrote: >> On Tue, Dec 3, 2013 at 10:12 AM, Ryan Coleman wrote: >> Is there a way to reset this without rebooting the server every 60 minutes? >> >> A umount/mount cycle on that fs will typically clear out this sort of thing. *If* you keep your logs on a separate fs (which everyone *should* be doing) it's faster than a reboot, but obviously not ideal, as it may require a bit of downtime each cycle. >> >> >> _______________________________________________ >> TCLUG Mailing List - Minneapolis/St. Paul, Minnesota >> tclug-list at mn-linux.org >> http://mailman.mn-linux.org/mailman/listinfo/tclug-list >> >> >> >> >> -- >> -Shawn >> >> >> _______________________________________________ >> TCLUG Mailing List - Minneapolis/St. Paul, Minnesota >> tclug-list at mn-linux.org >> http://mailman.mn-linux.org/mailman/listinfo/tclug-list > > > _______________________________________________ > TCLUG Mailing List - Minneapolis/St. Paul, Minnesota > tclug-list at mn-linux.org > http://mailman.mn-linux.org/mailman/listinfo/tclug-list > > > _______________________________________________ > TCLUG Mailing List - Minneapolis/St. Paul, Minnesota > tclug-list at mn-linux.org > http://mailman.mn-linux.org/mailman/listinfo/tclug-list -------------- next part -------------- An HTML attachment was scrubbed... URL: From ryanjcole at me.com Tue Dec 3 10:44:59 2013 From: ryanjcole at me.com (Ryan Coleman) Date: Tue, 03 Dec 2013 10:44:59 -0600 Subject: [tclug-list] Ubuntu box reporting free space incorrectly In-Reply-To: References: <1E767F30-54C7-4B11-BC7B-963EE3703E7C@me.com> <380826A6-FADE-4080-B91D-17F91F4A148F@me.com> Message-ID: <3B79AD18-E97E-4E3C-A342-61368EFEA922@me.com> I?ll take a look, thanks! On Dec 3, 2013, at 10:36 AM, Jeremy MountainJohnson wrote: > wipe and srm are possibilities too; although I would think there would > be a better solution to this. From jake.vath at gmail.com Tue Dec 3 10:51:06 2013 From: jake.vath at gmail.com (Jake Vath) Date: Tue, 3 Dec 2013 10:51:06 -0600 Subject: [tclug-list] Ubuntu box reporting free space incorrectly In-Reply-To: <3B79AD18-E97E-4E3C-A342-61368EFEA922@me.com> References: <1E767F30-54C7-4B11-BC7B-963EE3703E7C@me.com> <380826A6-FADE-4080-B91D-17F91F4A148F@me.com> <3B79AD18-E97E-4E3C-A342-61368EFEA922@me.com> Message-ID: Could you use *graceful-restart *or *rotatelogs* to take care of this? I'm not sure that either of those would update any inodes though. Alternately, would you be able to edit the *LogFormat *to not include that information, in the log file, for the time being? -> Jake On Tue, Dec 3, 2013 at 10:44 AM, Ryan Coleman wrote: > I?ll take a look, thanks! > On Dec 3, 2013, at 10:36 AM, Jeremy MountainJohnson < > jeremy.mountainjohnson at gmail.com> wrote: > > > wipe and srm are possibilities too; although I would think there would > > be a better solution to this. > > _______________________________________________ > TCLUG Mailing List - Minneapolis/St. Paul, Minnesota > tclug-list at mn-linux.org > http://mailman.mn-linux.org/mailman/listinfo/tclug-list > -------------- next part -------------- An HTML attachment was scrubbed... URL: From jake.vath at gmail.com Tue Dec 3 11:32:29 2013 From: jake.vath at gmail.com (Jake Vath) Date: Tue, 3 Dec 2013 11:32:29 -0600 Subject: [tclug-list] Ubuntu box reporting free space incorrectly In-Reply-To: References: <1E767F30-54C7-4B11-BC7B-963EE3703E7C@me.com> <380826A6-FADE-4080-B91D-17F91F4A148F@me.com> <3B79AD18-E97E-4E3C-A342-61368EFEA922@me.com> Message-ID: I guess, what I was saying might not have been clear. If you used *rotatelogs* (or something to rotate the logs), wouldn't that move *some_log_file.log* to *some_log_file.log.1* and recreate *some_log_file.log?* Then your scripts could remove *some_log_file.log.1*, which would create a new file descriptor that references the newly created inode for *some_log_file.log.* When you delete *some_log_file.log.1* wouldn't that free up that file descriptor and the inode which was referenced? Wouldn't this force an update of the available inodes in the file table? Using *rotatelogs or *Apache's *graceful* restart option should be different than just running *$ :> some_log_file.log *or* $ cat /dev/null > some_log_file.log*, as both of those options retain the current inode in use. However, I'm not entirely certain how either *rotatelogs or *Apache's *graceful* restart option handles file descriptors/inodes. I should probably go read some man pages... Anyone? Am I *way *off in my thinking? -> Jake On Tue, Dec 3, 2013 at 10:51 AM, Jake Vath wrote: > Could you use *graceful-restart *or *rotatelogs* to take care of this? > I'm not sure that either of those would update any inodes though. > > Alternately, would you be able to edit the *LogFormat *to not include > that information, in the log file, for the time being? > > -> Jake > > > On Tue, Dec 3, 2013 at 10:44 AM, Ryan Coleman wrote: > >> I?ll take a look, thanks! >> On Dec 3, 2013, at 10:36 AM, Jeremy MountainJohnson < >> jeremy.mountainjohnson at gmail.com> wrote: >> >> > wipe and srm are possibilities too; although I would think there would >> > be a better solution to this. >> >> _______________________________________________ >> TCLUG Mailing List - Minneapolis/St. Paul, Minnesota >> tclug-list at mn-linux.org >> http://mailman.mn-linux.org/mailman/listinfo/tclug-list >> > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From ryanjcole at me.com Tue Dec 3 11:41:21 2013 From: ryanjcole at me.com (Ryan Coleman) Date: Tue, 03 Dec 2013 11:41:21 -0600 Subject: [tclug-list] Ubuntu box reporting free space incorrectly In-Reply-To: References: <1E767F30-54C7-4B11-BC7B-963EE3703E7C@me.com> <380826A6-FADE-4080-B91D-17F91F4A148F@me.com> <3B79AD18-E97E-4E3C-A342-61368EFEA922@me.com> Message-ID: <30AC72DE-F74B-4B9A-AC18-4291DF711640@me.com> On Dec 3, 2013, at 11:32 AM, Jake Vath wrote: > I guess, what I was saying might not have been clear. > > If you used rotatelogs (or something to rotate the logs), wouldn't that move some_log_file.log to some_log_file.log.1 and recreate some_log_file.log? > Then your scripts could remove some_log_file.log.1, which would create a new file descriptor that references the newly created inode for some_log_file.log. > When you delete some_log_file.log.1 wouldn't that free up that file descriptor and the inode which was referenced? > Wouldn't this force an update of the available inodes in the file table? Yes, I believe so. But I haven?t been using rotatelogs at all so? I wouldn?t know (even on my production servers - I have massive storage for log files I?ve always compressed them once a year and burned them to multiple media types for archiving). As for logformat if I knew where this was coming from I wouldn?t have to rotate logs - I?d just kill the logging. LDAP is only used in my application (within PHP) not on the server itself (which is using it as a different login with lighttpd). > Using rotatelogs or Apache's graceful restart option should be different than just running $ :> some_log_file.log or $ cat /dev/null > some_log_file.log, > as both of those options retain the current inode in use. > However, I'm not entirely certain how either rotatelogs or Apache's graceful restart option handles file descriptors/inodes. > I should probably go read some man pages... > > Anyone? Am I way off in my thinking? > > -> Jake > > > On Tue, Dec 3, 2013 at 10:51 AM, Jake Vath wrote: > Could you use graceful-restart or rotatelogs to take care of this? > I'm not sure that either of those would update any inodes though. > > Alternately, would you be able to edit the LogFormat to not include that information, in the log file, for the time being? > > -> Jake > > > On Tue, Dec 3, 2013 at 10:44 AM, Ryan Coleman wrote: > I?ll take a look, thanks! > On Dec 3, 2013, at 10:36 AM, Jeremy MountainJohnson wrote: > > > wipe and srm are possibilities too; although I would think there would > > be a better solution to this. > > _______________________________________________ > TCLUG Mailing List - Minneapolis/St. Paul, Minnesota > tclug-list at mn-linux.org > http://mailman.mn-linux.org/mailman/listinfo/tclug-list > > > _______________________________________________ > TCLUG Mailing List - Minneapolis/St. Paul, Minnesota > tclug-list at mn-linux.org > http://mailman.mn-linux.org/mailman/listinfo/tclug-list -------------- next part -------------- An HTML attachment was scrubbed... URL: From jake.vath at gmail.com Tue Dec 3 11:47:02 2013 From: jake.vath at gmail.com (Jake Vath) Date: Tue, 3 Dec 2013 11:47:02 -0600 Subject: [tclug-list] Ubuntu box reporting free space incorrectly In-Reply-To: <30AC72DE-F74B-4B9A-AC18-4291DF711640@me.com> References: <1E767F30-54C7-4B11-BC7B-963EE3703E7C@me.com> <380826A6-FADE-4080-B91D-17F91F4A148F@me.com> <3B79AD18-E97E-4E3C-A342-61368EFEA922@me.com> <30AC72DE-F74B-4B9A-AC18-4291DF711640@me.com> Message-ID: > > As for logformat if I knew where this was coming from I wouldn?t have to > rotate logs - I?d just kill the logging. LDAP is only used in my > application (within PHP) not on the server itself (which is using it as a > different login with lighttpd). Ahh, that makes sense. -> Jake On Tue, Dec 3, 2013 at 11:41 AM, Ryan Coleman wrote: > > On Dec 3, 2013, at 11:32 AM, Jake Vath wrote: > > I guess, what I was saying might not have been clear. > > If you used *rotatelogs* (or something to rotate the logs), wouldn't that > move *some_log_file.log* to *some_log_file.log.1* and recreate > *some_log_file.log?* > Then your scripts could remove *some_log_file.log.1*, which would create > a new file descriptor that references the newly created inode for > *some_log_file.log.* > When you delete *some_log_file.log.1* wouldn't that free up that file > descriptor and the inode which was referenced? > Wouldn't this force an update of the available inodes in the file table? > > > Yes, I believe so. But I haven?t been using rotatelogs at all so? I > wouldn?t know (even on my production servers - I have massive storage for > log files I?ve always compressed them once a year and burned them to > multiple media types for archiving). > > As for logformat if I knew where this was coming from I wouldn?t have to > rotate logs - I?d just kill the logging. LDAP is only used in my > application (within PHP) not on the server itself (which is using it as a > different login with lighttpd). > > > Using *rotatelogs or *Apache's *graceful* restart option should be > different than just running *$ :> some_log_file.log *or* $ cat /dev/null > > some_log_file.log*, > as both of those options retain the current inode in use. > However, I'm not entirely certain how either *rotatelogs or *Apache's > *graceful* restart option handles file descriptors/inodes. > I should probably go read some man pages... > > Anyone? Am I *way *off in my thinking? > > -> Jake > > > On Tue, Dec 3, 2013 at 10:51 AM, Jake Vath wrote: > >> Could you use *graceful-restart *or *rotatelogs* to take care of this? >> I'm not sure that either of those would update any inodes though. >> >> Alternately, would you be able to edit the *LogFormat *to not include >> that information, in the log file, for the time being? >> >> -> Jake >> >> >> On Tue, Dec 3, 2013 at 10:44 AM, Ryan Coleman wrote: >> >>> I?ll take a look, thanks! >>> On Dec 3, 2013, at 10:36 AM, Jeremy MountainJohnson < >>> jeremy.mountainjohnson at gmail.com> wrote: >>> >>> > wipe and srm are possibilities too; although I would think there would >>> > be a better solution to this. >>> >>> _______________________________________________ >>> TCLUG Mailing List - Minneapolis/St. Paul, Minnesota >>> tclug-list at mn-linux.org >>> http://mailman.mn-linux.org/mailman/listinfo/tclug-list >>> >> >> > _______________________________________________ > TCLUG Mailing List - Minneapolis/St. Paul, Minnesota > tclug-list at mn-linux.org > http://mailman.mn-linux.org/mailman/listinfo/tclug-list > > > > _______________________________________________ > TCLUG Mailing List - Minneapolis/St. Paul, Minnesota > tclug-list at mn-linux.org > http://mailman.mn-linux.org/mailman/listinfo/tclug-list > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From john.a.frisk at gmail.com Wed Dec 4 15:37:57 2013 From: john.a.frisk at gmail.com (John Frisk) Date: Wed, 4 Dec 2013 15:37:57 -0600 Subject: [tclug-list] Penguins unbound informal get together Message-ID: Come and join the penguins for an informal get together Tuesday December 10th, 7-9PM (or whenever) at Stouts across the street from TIES. 1611 Larpenteur Ave W, Falcon Heights, MN 55113 http://www.stoutspub.com We can chat about any topics and I'll probably mention next years line up of meeting thoughts. -------------- next part -------------- An HTML attachment was scrubbed... URL: From canito at dalan.us Fri Dec 6 14:09:36 2013 From: canito at dalan.us (canito at dalan.us) Date: Fri, 06 Dec 2013 14:09:36 -0600 Subject: [tclug-list] Count Number of Matched Patterns Message-ID: <20131206140936.Horde.XxBkDjEchZmc-Dos6FqGcw5@mail.dalan.us> In the process of writing a script which I would like to count the number matched patterns (command line parameters). The issue I am running into using grep is that the string has a dash in front, and it throws an error. E.g: ./script -test PATTERN=$1 egrep -ic $PATTERN --> egrep: invalid option -- 't' awk and egrep work using a file, but not on a variable: EXAMPLE=`awk '/test/ { nlines++ } {print nlines}' $PATTERN` awk: cmd. line:1: fatal: cannot open file `-test' for reading (No such file or directory) I know using and if command works, so am I just over doing it? What am I doing wrong? if [[ $PATTERN == "-test" ]; then Thanks in advanced! Saul David Alanis From jake.vath at gmail.com Fri Dec 6 14:12:21 2013 From: jake.vath at gmail.com (Jake Vath) Date: Fri, 6 Dec 2013 14:12:21 -0600 Subject: [tclug-list] Count Number of Matched Patterns In-Reply-To: <20131206140936.Horde.XxBkDjEchZmc-Dos6FqGcw5@mail.dalan.us> References: <20131206140936.Horde.XxBkDjEchZmc-Dos6FqGcw5@mail.dalan.us> Message-ID: Did you copy-past that if-statement wrong? You're missing a closing ] -> Jake On Fri, Dec 6, 2013 at 2:09 PM, wrote: > In the process of writing a script which I would like to count the number > matched patterns (command line parameters). > > The issue I am running into using grep is that the string has a dash in > front, and it throws an error. > > E.g: > > ./script -test > > PATTERN=$1 > > egrep -ic $PATTERN --> egrep: invalid option -- 't' > > awk and egrep work using a file, but not on a variable: > > EXAMPLE=`awk '/test/ { nlines++ } {print nlines}' $PATTERN` > > awk: cmd. line:1: fatal: cannot open file `-test' for reading (No such > file or directory) > > I know using and if command works, so am I just over doing it? What am I > doing wrong? > > if [[ $PATTERN == "-test" ]; then > > Thanks in advanced! > > Saul David Alanis > > > > > _______________________________________________ > TCLUG Mailing List - Minneapolis/St. Paul, Minnesota > tclug-list at mn-linux.org > http://mailman.mn-linux.org/mailman/listinfo/tclug-list > -------------- next part -------------- An HTML attachment was scrubbed... URL: From canito at dalan.us Fri Dec 6 14:23:41 2013 From: canito at dalan.us (canito at dalan.us) Date: Fri, 06 Dec 2013 14:23:41 -0600 Subject: [tclug-list] Count Number of Matched Patterns In-Reply-To: References: <20131206140936.Horde.XxBkDjEchZmc-Dos6FqGcw5@mail.dalan.us> Message-ID: <20131206142341.Horde.zOoS-aGEF41iwj08u1r3HA1@mail.dalan.us> Quoting Jake Vath : > Did you copy-past that if-statement wrong? > You're missing a closing ] > > -> Jake > > > On Fri, Dec 6, 2013 at 2:09 PM, wrote: > >> In the process of writing a script which I would like to count the number >> matched patterns (command line parameters). >> >> The issue I am running into using grep is that the string has a dash in >> front, and it throws an error. >> >> E.g: >> >> ./script -test >> >> PATTERN=$1 >> >> egrep -ic $PATTERN --> egrep: invalid option -- 't' >> >> awk and egrep work using a file, but not on a variable: >> >> EXAMPLE=`awk '/test/ { nlines++ } {print nlines}' $PATTERN` >> >> awk: cmd. line:1: fatal: cannot open file `-test' for reading (No such >> file or directory) >> >> I know using and if command works, so am I just over doing it? What am I >> doing wrong? >> >> if [[ $PATTERN == "-test" ]; then >> >> Thanks in advanced! >> >> Saul David Alanis >> >> >> >> >> _______________________________________________ >> TCLUG Mailing List - Minneapolis/St. Paul, Minnesota >> tclug-list at mn-linux.org >> http://mailman.mn-linux.org/mailman/listinfo/tclug-list >> Hi Jake- Thanks, just a typo. :) I would like to know if I am just missing something using egrep / awk? SDA From jus at krytosvirus.com Fri Dec 6 14:24:10 2013 From: jus at krytosvirus.com (Justin Krejci) Date: Fri, 06 Dec 2013 14:24:10 -0600 Subject: [tclug-list] Count Number of Matched Patterns Message-ID: <7iia4jqeavwqye75lf9wxy0p.1386361450630@email.android.com> I think most cli utilities support using -- (two dashes) to signify the end of the commands options.? Example? $ grep -- -t file.txt? -------- Original message -------- From: canito at dalan.us Date: 12/06/2013 2:09 PM (GMT-06:00) To: tclug-list at mn-linux.org Subject: [tclug-list] Count Number of Matched Patterns In the process of writing a script which I would like to count the? number matched patterns (command line parameters). The issue I am running into using grep is that the string has a dash? in front, and it throws an error. E.g: ./script -test PATTERN=$1 egrep -ic $PATTERN --> egrep: invalid option -- 't' awk and egrep work using a file, but not on a variable: EXAMPLE=`awk '/test/ { nlines++ } {print nlines}' $PATTERN` awk: cmd. line:1: fatal: cannot open file `-test' for reading (No such? file or directory) I know using and if command works, so am I just over doing it? What am? I doing wrong? if [[ $PATTERN == "-test" ]; then Thanks in advanced! Saul David Alanis _______________________________________________ TCLUG Mailing List - Minneapolis/St. Paul, Minnesota tclug-list at mn-linux.org http://mailman.mn-linux.org/mailman/listinfo/tclug-list -------------- next part -------------- An HTML attachment was scrubbed... URL: From canito at dalan.us Fri Dec 6 14:44:15 2013 From: canito at dalan.us (canito at dalan.us) Date: Fri, 06 Dec 2013 14:44:15 -0600 Subject: [tclug-list] Count Number of Matched Patterns In-Reply-To: <7iia4jqeavwqye75lf9wxy0p.1386361450630@email.android.com> References: <7iia4jqeavwqye75lf9wxy0p.1386361450630@email.android.com> Message-ID: <20131206144415.Horde.axGoelRcKa8gdofl8Gk1KQ1@mail.dalan.us> Quoting Justin Krejci : > I think most cli utilities support using -- (two dashes) to signify > the end of the commands options.? > > Example? > > $ grep -- -t file.txt? > > > Thank you Justin- I gave that a try since the script accepts single and double dash, but the problem remains using egrep. Looking at the man pages I read the following, which to me means that the pattern can have a dash in front? Matching Control -e PATTERN, --regexp=PATTERN Use PATTERN as the pattern. This can be used to specify multiple search patterns, or to protect a pattern beginning with a hyphen (-). (-e is specified by POSIX.) Anyway, I tried with -- and added the -e switch to egrep ending with the same results. I may just have to stick with an if statement. :( SDA > -------- Original message -------- > From: canito at dalan.us > Date: 12/06/2013 2:09 PM (GMT-06:00) > To: tclug-list at mn-linux.org > Subject: [tclug-list] Count Number of Matched Patterns > > In the process of writing a script which I would like to count the? > number matched patterns (command line parameters). > > The issue I am running into using grep is that the string has a dash? > in front, and it throws an error. > > E.g: > > ./script -test > > PATTERN=$1 > > egrep -ic $PATTERN --> egrep: invalid option -- 't' > > awk and egrep work using a file, but not on a variable: > > EXAMPLE=`awk '/test/ { nlines++ } {print nlines}' $PATTERN` > > awk: cmd. line:1: fatal: cannot open file `-test' for reading (No such? > file or directory) > > I know using and if command works, so am I just over doing it? What am? > I doing wrong? > > if [[ $PATTERN == "-test" ]; then > > Thanks in advanced! > > Saul David Alanis > > > > > _______________________________________________ > TCLUG Mailing List - Minneapolis/St. Paul, Minnesota > tclug-list at mn-linux.org > http://mailman.mn-linux.org/mailman/listinfo/tclug-list From jake.vath at gmail.com Fri Dec 6 14:56:27 2013 From: jake.vath at gmail.com (Jake Vath) Date: Fri, 6 Dec 2013 14:56:27 -0600 Subject: [tclug-list] Count Number of Matched Patterns In-Reply-To: <20131206142341.Horde.zOoS-aGEF41iwj08u1r3HA1@mail.dalan.us> References: <20131206140936.Horde.XxBkDjEchZmc-Dos6FqGcw5@mail.dalan.us> <20131206142341.Horde.zOoS-aGEF41iwj08u1r3HA1@mail.dalan.us> Message-ID: I'm apologize, I didn't read your email in its entirety. -> Jake On Fri, Dec 6, 2013 at 2:23 PM, wrote: > > Quoting Jake Vath : > > Did you copy-past that if-statement wrong? >> You're missing a closing ] >> >> -> Jake >> >> >> On Fri, Dec 6, 2013 at 2:09 PM, wrote: >> >> In the process of writing a script which I would like to count the number >>> matched patterns (command line parameters). >>> >>> The issue I am running into using grep is that the string has a dash in >>> front, and it throws an error. >>> >>> E.g: >>> >>> ./script -test >>> >>> PATTERN=$1 >>> >>> egrep -ic $PATTERN --> egrep: invalid option -- 't' >>> >>> awk and egrep work using a file, but not on a variable: >>> >>> EXAMPLE=`awk '/test/ { nlines++ } {print nlines}' $PATTERN` >>> >>> awk: cmd. line:1: fatal: cannot open file `-test' for reading (No such >>> file or directory) >>> >>> I know using and if command works, so am I just over doing it? What am I >>> doing wrong? >>> >>> if [[ $PATTERN == "-test" ]; then >>> >>> Thanks in advanced! >>> >>> Saul David Alanis >>> >>> >>> >>> >>> _______________________________________________ >>> TCLUG Mailing List - Minneapolis/St. Paul, Minnesota >>> tclug-list at mn-linux.org >>> http://mailman.mn-linux.org/mailman/listinfo/tclug-list >>> >>> > Hi Jake- > > Thanks, just a typo. :) > > I would like to know if I am just missing something using egrep / awk? > > SDA > > > _______________________________________________ > TCLUG Mailing List - Minneapolis/St. Paul, Minnesota > tclug-list at mn-linux.org > http://mailman.mn-linux.org/mailman/listinfo/tclug-list > -------------- next part -------------- An HTML attachment was scrubbed... URL: From mkorangestripe at gmail.com Fri Dec 6 15:13:56 2013 From: mkorangestripe at gmail.com (Gavin Purcell) Date: Fri, 6 Dec 2013 15:13:56 -0600 Subject: [tclug-list] Count Number of Matched Patterns In-Reply-To: References: <20131206140936.Horde.XxBkDjEchZmc-Dos6FqGcw5@mail.dalan.us> <20131206142341.Horde.zOoS-aGEF41iwj08u1r3HA1@mail.dalan.us> Message-ID: Part of this reminds me of when Bash evaluates an expression (e.g. *ap*les ? aptles*) when a file named aptles exists in the working directory, but the intended search pattern is apples, appples, etc. On Fri, Dec 6, 2013 at 2:56 PM, Jake Vath wrote: > I'm apologize, I didn't read your email in its entirety. > > -> Jake > > > On Fri, Dec 6, 2013 at 2:23 PM, wrote: > >> >> Quoting Jake Vath : >> >> Did you copy-past that if-statement wrong? >>> You're missing a closing ] >>> >>> -> Jake >>> >>> >>> On Fri, Dec 6, 2013 at 2:09 PM, wrote: >>> >>> In the process of writing a script which I would like to count the >>>> number >>>> matched patterns (command line parameters). >>>> >>>> The issue I am running into using grep is that the string has a dash in >>>> front, and it throws an error. >>>> >>>> E.g: >>>> >>>> ./script -test >>>> >>>> PATTERN=$1 >>>> >>>> egrep -ic $PATTERN --> egrep: invalid option -- 't' >>>> >>>> awk and egrep work using a file, but not on a variable: >>>> >>>> EXAMPLE=`awk '/test/ { nlines++ } {print nlines}' $PATTERN` >>>> >>>> awk: cmd. line:1: fatal: cannot open file `-test' for reading (No such >>>> file or directory) >>>> >>>> I know using and if command works, so am I just over doing it? What am I >>>> doing wrong? >>>> >>>> if [[ $PATTERN == "-test" ]; then >>>> >>>> Thanks in advanced! >>>> >>>> Saul David Alanis >>>> >>>> >>>> >>>> >>>> _______________________________________________ >>>> TCLUG Mailing List - Minneapolis/St. Paul, Minnesota >>>> tclug-list at mn-linux.org >>>> http://mailman.mn-linux.org/mailman/listinfo/tclug-list >>>> >>>> >> Hi Jake- >> >> Thanks, just a typo. :) >> >> I would like to know if I am just missing something using egrep / awk? >> >> SDA >> >> >> _______________________________________________ >> TCLUG Mailing List - Minneapolis/St. Paul, Minnesota >> tclug-list at mn-linux.org >> http://mailman.mn-linux.org/mailman/listinfo/tclug-list >> > > > _______________________________________________ > TCLUG Mailing List - Minneapolis/St. Paul, Minnesota > tclug-list at mn-linux.org > http://mailman.mn-linux.org/mailman/listinfo/tclug-list > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From zedlan at invec.net Fri Dec 6 15:16:34 2013 From: zedlan at invec.net (zedlan at invec.net) Date: Fri, 06 Dec 2013 21:16:34 +0000 Subject: [tclug-list] Count Number of Matched Patterns Message-ID: Saul, You should add END between the scopes, such as: awk '/pattern/ { nlines++ } END {print nlines}' file ... works with gnu AWK Jack Pope -----Original Message----- From: canito at dalan.us [mailto:canito at dalan.us] Sent: Friday, December 6, 2013 03:09 PM To: tclug-list at mn-linux.org Subject: [tclug-list] Count Number of Matched Patterns In the process of writing a script which I would like to count the number matched patterns (command line parameters).The issue I am running into using grep is that the string has a dash in front, and it throws an error.E.g:./script -testPATTERN=$1egrep -ic $PATTERN --> egrep: invalid option -- 't'awk and egrep work using a file, but not on a variable:EXAMPLE=`awk '/test/ { nlines++ } {print nlines}' $PATTERN`awk: cmd. line:1: fatal: cannot open file `-test' for reading (No such file or directory)I know using and if command works, so am I just over doing it? What am I doing wrong?if [[ $PATTERN == "-test" ]; thenThanks in advanced!Saul David Alanis_______________________________________________TCLUG Mailing List - Minneapolis/St. Paul, Minnesotatclug-list at mn-linux.orghttp://mailman.mn-linux.org/mailman/listinfo/tclug-list -------------- next part -------------- An HTML attachment was scrubbed... URL: From tclug1 at whitleymott.net Fri Dec 6 16:41:08 2013 From: tclug1 at whitleymott.net (gregrwm) Date: Fri, 6 Dec 2013 16:41:08 -0600 Subject: [tclug-list] reattach stranded screen and/or prod less to write buffer Message-ID: i logged in while numerous fscks (several on same disc) (including /) were running simultaneously (a mountall bug), ran ps|less (inside screen) to capture the evidence, but lost that login before telling less to write it out. i logged in again, but in the meantime the fscks finished. now, the prior screen, which is still running, is not listed in /run/screen (i'm guessing because it predated the initialization of /run), so even tho i specify SCREEN -D -RR, i just get a new screen session, not the old one. how can i somehow either attach to the original screen which is still running, or prod that running copy of less to write out it's buffer? i tried copying /run/screen/S-root/4951.pts-10.jocur (the more recent screen session) to /run/screen/870.pts-13.jocur, also tried values of pts-0 through pts-9 but none of them got me into screen. no doubt more than just the name of that pipe is relevant. how might i craft the right pipe to get me into that screen, or do you have a better idea? #00us#Dec06Fri16:20#Greg at jocur:/run/screen/S-root# pg SCR 870 4951 USER SESS TPGID PGRP PPID PID STARTED NI WCHAN STAT COMMAND root 4951 -1 4951 4950 4951 16:20:06 0 poll_s Ss SCREEN -O -d -RR -U -e^Bb root 870 -1 870 1 870 13:38:51 0 poll_s Ss SCREEN -O -d -RR -U -e^Bb #00us#Dec06Fri16:20#Greg at jocur:/run/screen/S-root# l. prwx------ 1 root greg 0 2013-12-06 Fri 16:20:06 4951.pts-10.jocur| prwx------ 1 root greg 0 2013-12-06 Fri 16:00:32 870.pts-13.jocur| #00us#Dec06Fri16:20#Greg at jocur:/run/screen/S-root# lt /dev/pts/* c--------- 1 root root 5, 2 2013-12-06 Fri 13:37:34 /dev/pts/ptmx crw--w---- 1 root tty 136, 6 2013-12-06 Fri 13:39:31 /dev/pts/6 crw--w---- 1 root tty 136, 26 2013-12-06 Fri 13:39:36 /dev/pts/26 crw------- 1 root tty 136, 3 2013-12-06 Fri 13:43:20 /dev/pts/3 crw------- 1 root tty 136, 2 2013-12-06 Fri 13:46:08 /dev/pts/2 crw------- 1 root tty 136, 7 2013-12-06 Fri 14:17:12 /dev/pts/7 crw--w---- 1 g tty 136, 0 2013-12-06 Fri 15:41:20 /dev/pts/0 crw--w---- 1 root tty 136, 13 2013-12-06 Fri 16:04:40 /dev/pts/13 crw--w---- 1 g tty 136, 10 2013-12-06 Fri 16:20:16 /dev/pts/10 crw------- 1 root tty 136, 5 2013-12-06 Fri 16:20:16 /dev/pts/5 crw------- 1 root tty 136, 8 2013-12-06 Fri 16:20:16 /dev/pts/8 crw--w---- 1 root tty 136, 1 2013-12-06 Fri 19:37:43 /dev/pts/1 crw--w---- 1 root tty 136, 4 2013-12-06 Fri 19:37:44 /dev/pts/4 -------------- next part -------------- An HTML attachment was scrubbed... URL: From jason.lindberg at gmail.com Fri Dec 6 18:57:05 2013 From: jason.lindberg at gmail.com (Jason Lindberg) Date: Fri, 6 Dec 2013 18:57:05 -0600 Subject: [tclug-list] reattach stranded screen and/or prod less to write buffer In-Reply-To: References: Message-ID: Try: screen -r 4951 *Jason Lindberg* Systems Engineer - UNIX/Linux (763) 355-2473 On Fri, Dec 6, 2013 at 4:41 PM, gregrwm wrote: > i logged in while numerous fscks (several on same disc) (including /) were > running simultaneously (a mountall bug), ran ps|less (inside screen) to > capture the evidence, but lost that login before telling less to write it > out. i logged in again, but in the meantime the fscks finished. now, the > prior screen, which is still running, is not listed in /run/screen (i'm > guessing because it predated the initialization of /run), so even tho i > specify SCREEN -D -RR, i just get a new screen session, not the old one. > > how can i somehow either attach to the original screen which is still > running, or prod that running copy of less to write out it's buffer? > > i tried copying /run/screen/S-root/4951.pts-10.jocur (the more recent > screen session) to /run/screen/870.pts-13.jocur, also tried values of pts-0 > through pts-9 but none of them got me into screen. no doubt more than just > the name of that pipe is relevant. how might i craft the right pipe to get > me into that screen, or do you have a better idea? > > #00us#Dec06Fri16:20#Greg at jocur:/run/screen/S-root# pg SCR > 870 4951 > USER SESS TPGID PGRP PPID PID STARTED NI WCHAN STAT COMMAND > root 4951 -1 4951 4950 4951 16:20:06 0 poll_s Ss SCREEN -O > -d -RR -U -e^Bb > root 870 -1 870 1 870 13:38:51 0 poll_s Ss SCREEN -O > -d -RR -U -e^Bb > #00us#Dec06Fri16:20#Greg at jocur:/run/screen/S-root# l. > prwx------ 1 root greg 0 2013-12-06 Fri 16:20:06 4951.pts-10.jocur| > prwx------ 1 root greg 0 2013-12-06 Fri 16:00:32 870.pts-13.jocur| > #00us#Dec06Fri16:20#Greg at jocur:/run/screen/S-root# lt /dev/pts/* > c--------- 1 root root 5, 2 2013-12-06 Fri 13:37:34 /dev/pts/ptmx > crw--w---- 1 root tty 136, 6 2013-12-06 Fri 13:39:31 /dev/pts/6 > crw--w---- 1 root tty 136, 26 2013-12-06 Fri 13:39:36 /dev/pts/26 > crw------- 1 root tty 136, 3 2013-12-06 Fri 13:43:20 /dev/pts/3 > crw------- 1 root tty 136, 2 2013-12-06 Fri 13:46:08 /dev/pts/2 > crw------- 1 root tty 136, 7 2013-12-06 Fri 14:17:12 /dev/pts/7 > crw--w---- 1 g tty 136, 0 2013-12-06 Fri 15:41:20 /dev/pts/0 > crw--w---- 1 root tty 136, 13 2013-12-06 Fri 16:04:40 /dev/pts/13 > crw--w---- 1 g tty 136, 10 2013-12-06 Fri 16:20:16 /dev/pts/10 > crw------- 1 root tty 136, 5 2013-12-06 Fri 16:20:16 /dev/pts/5 > crw------- 1 root tty 136, 8 2013-12-06 Fri 16:20:16 /dev/pts/8 > crw--w---- 1 root tty 136, 1 2013-12-06 Fri 19:37:43 /dev/pts/1 > crw--w---- 1 root tty 136, 4 2013-12-06 Fri 19:37:44 /dev/pts/4 > > _______________________________________________ > TCLUG Mailing List - Minneapolis/St. Paul, Minnesota > tclug-list at mn-linux.org > http://mailman.mn-linux.org/mailman/listinfo/tclug-list > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From tclug1 at whitleymott.net Sat Dec 7 00:37:05 2013 From: tclug1 at whitleymott.net (gregrwm) Date: Sat, 7 Dec 2013 00:37:05 -0600 Subject: [tclug-list] reattach stranded screen and/or prod less to write buffer Message-ID: > > screen -r 4951 > only works if the /run/screen entry hasn't been deleted: #0us#Dec06Fri22:50#Greg at jocur:/proc/6319/fd# rm /run/screen/S-root/6319.pts-10.jocur #0us#Dec06Fri22:51#Greg at jocur:/proc/6319/fd# screen -r 6319.pts-10.jocur There is no screen to be resumed matching 6319.pts-10.jocur. reptyr looks good, thank you. unfortunately in my attempts to hack into that screen, it died: #0us#Dec06Fri22:38#Greg at jocur:/run/screen/S-root# link 870.pts-0.jocur /proc/870/fd/4 `870.pts-0.jocur' -> `/proc/870/fd/4' #0us#Dec06Fri22:38#Greg at jocur:/run/screen/S-root# screen -D -RR Cannot contact screen again. Sigh. (and it was gone) -------------- next part -------------- An HTML attachment was scrubbed... URL: From mbmiller+l at gmail.com Mon Dec 9 01:19:55 2013 From: mbmiller+l at gmail.com (Mike Miller) Date: Mon, 9 Dec 2013 01:19:55 -0600 (CST) Subject: [tclug-list] Count Number of Matched Patterns In-Reply-To: <20131206140936.Horde.XxBkDjEchZmc-Dos6FqGcw5@mail.dalan.us> References: <20131206140936.Horde.XxBkDjEchZmc-Dos6FqGcw5@mail.dalan.us> Message-ID: I'm not sure of exactly what your issue is. I'm thinking you need the "--", so I made a script with that, exactly like this: -------begin script on next line--------- #!/bin/bash PATTERN=$1 grep -Eic -- "$PATTERN" -------end script on previous line--------- And I ran it: $ echo -e "-testa\n-testb\nblah\nfoo\nbar\ntest\na-test" | ./script -test 3 That's the right answer. Note that if you don't have a filename, you have to grep from stdin, which is what I did. Mike On Fri, 6 Dec 2013, canito at dalan.us wrote: > In the process of writing a script which I would like to count the number > matched patterns (command line parameters). > > The issue I am running into using grep is that the string has a dash in > front, and it throws an error. > > E.g: > > ./script -test > > PATTERN=$1 > > egrep -ic $PATTERN --> egrep: invalid option -- 't' > > awk and egrep work using a file, but not on a variable: > > EXAMPLE=`awk '/test/ { nlines++ } {print nlines}' $PATTERN` > > awk: cmd. line:1: fatal: cannot open file `-test' for reading (No such file > or directory) > > I know using and if command works, so am I just over doing it? What am I > doing wrong? > > if [[ $PATTERN == "-test" ]; then > > Thanks in advanced! > > Saul David Alanis From tclug1 at whitleymott.net Mon Dec 9 01:57:18 2013 From: tclug1 at whitleymott.net (gregrwm) Date: Mon, 9 Dec 2013 01:57:18 -0600 Subject: [tclug-list] reattach stranded screen and/or prod less to write buffer In-Reply-To: References: Message-ID: yes, on startup. i'm launching sshd early, and logging in while fsck is still running. what i'd like to know now is, what startup script is responsible for clearing /run/screen/*/*? /etc/init/screen-cleanup.conf doesn't look like it does it. perhaps something clears all of /run? what script? On Dec 8, 2013 9:09 PM, "gsker" wrote: > /run is dynamic, so if it truly was on startup, then it would go away. On Dec 8, 2013 2:52 PM, "gregrwm" wrote: > here root launched screen while fsck of / was still running, and g > launched it later. root hasn't detached. g detached and reattached: > > # pg SC > 1003 4276 > USER SESS TPGID PGRP PPID PID STARTED NI WCHAN STAT COMMAND > root 1003 -1 1003 1002 1003 Dec 06 0 poll_s Ss SCREEN -O > -d -RR -U -e^Aa > g 4276 -1 4276 1 4276 07:03:29 0 poll_s Ss SCREEN -O > -d -RR -U -e^Aa > # lu /proc/{1003,4276}/fd/4 /run/sc*/*/* > lr-x------ 1 root root 64 2013-12-08 Sun 14:25:50 /proc/1003/fd/4 -> > /run/screen/S-root/1003.tty5.nana (deleted) > lr-x------ 1 root root 64 2013-12-08 Sun 14:25:50 /proc/4276/fd/4 -> > /dev/pts/6 > prwx------ 1 g greg 0 2013-12-08 Sun 14:30:56 > /run/screen/S-g/4276.pts-14.nana| > > root's /run/screen entry says "(deleted)", which suggests it was created, > and something subsequently deletes /run/screen/*/*. can anyone point me > into the startup morass to where that happens? > -------------- next part -------------- An HTML attachment was scrubbed... URL: From canito at dalan.us Mon Dec 9 10:44:26 2013 From: canito at dalan.us (canito at dalan.us) Date: Mon, 09 Dec 2013 10:44:26 -0600 Subject: [tclug-list] Count Number of Matched Patterns In-Reply-To: References: <20131206140936.Horde.XxBkDjEchZmc-Dos6FqGcw5@mail.dalan.us> Message-ID: <20131209104426.Horde.5eEJCuCn3VLsUTM4-1QfOA1@mail.dalan.us> Quoting Mike Miller : > I'm not sure of exactly what your issue is. I'm thinking you need > the "--", so I made a script with that, exactly like this: > > > -------begin script on next line--------- > #!/bin/bash > > PATTERN=$1 > > grep -Eic -- "$PATTERN" > -------end script on previous line--------- > > > And I ran it: > > > $ echo -e "-testa\n-testb\nblah\nfoo\nbar\ntest\na-test" | ./script -test > 3 > > > That's the right answer. Note that if you don't have a filename, > you have to grep from stdin, which is what I did. > > Mike > Thank you Mike- Not sure what the problem is either. I've had a couple of issues with egrep/grep on Red Hat. Again, I do appreciate the follow up. SDA > > On Fri, 6 Dec 2013, canito at dalan.us wrote: > >> In the process of writing a script which I would like to count the >> number matched patterns (command line parameters). >> >> The issue I am running into using grep is that the string has a >> dash in front, and it throws an error. >> >> E.g: >> >> ./script -test >> >> PATTERN=$1 >> >> egrep -ic $PATTERN --> egrep: invalid option -- 't' >> >> awk and egrep work using a file, but not on a variable: >> >> EXAMPLE=`awk '/test/ { nlines++ } {print nlines}' $PATTERN` >> >> awk: cmd. line:1: fatal: cannot open file `-test' for reading (No >> such file or directory) >> >> I know using and if command works, so am I just over doing it? What >> am I doing wrong? >> >> if [[ $PATTERN == "-test" ]; then >> >> Thanks in advanced! >> >> Saul David Alanis > _______________________________________________ > TCLUG Mailing List - Minneapolis/St. Paul, Minnesota > tclug-list at mn-linux.org > http://mailman.mn-linux.org/mailman/listinfo/tclug-list From mbmiller+l at gmail.com Mon Dec 9 15:40:39 2013 From: mbmiller+l at gmail.com (Mike Miller) Date: Mon, 9 Dec 2013 15:40:39 -0600 (CST) Subject: [tclug-list] Count Number of Matched Patterns In-Reply-To: <20131209104426.Horde.5eEJCuCn3VLsUTM4-1QfOA1@mail.dalan.us> References: <20131206140936.Horde.XxBkDjEchZmc-Dos6FqGcw5@mail.dalan.us> <20131209104426.Horde.5eEJCuCn3VLsUTM4-1QfOA1@mail.dalan.us> Message-ID: On Mon, 9 Dec 2013, canito at dalan.us wrote: > Quoting Mike Miller : > >> I'm not sure of exactly what your issue is. I'm thinking you need the >> "--", so I made a script with that, exactly like this: >> >> >> -------begin script on next line--------- >> #!/bin/bash >> >> PATTERN=$1 >> >> grep -Eic -- "$PATTERN" >> -------end script on previous line--------- >> >> >> And I ran it: >> >> >> $ echo -e "-testa\n-testb\nblah\nfoo\nbar\ntest\na-test" | ./script -test >> 3 >> >> >> That's the right answer. Note that if you don't have a filename, you have >> to grep from stdin, which is what I did. >> >> Mike > > Thank you Mike- > > Not sure what the problem is either. I've had a couple of issues with > egrep/grep on Red Hat. > > Again, I do appreciate the follow up. What version of grep are you using? This is mine: $ grep --version GNU grep 2.5.4 $ egrep --version GNU grep 2.5.4 I read somewhere a few years ago that egrep was deprecated and that we were supposed to use "grep -E" instead. I doubt that's the problem (at least on my system, I get the same result with egrep as with "grep -E"). I know that earlier versions of the perl grep, grep -P, were pretty buggy but I've had better luck recently. Mike From canito at dalan.us Mon Dec 9 16:21:51 2013 From: canito at dalan.us (canito at dalan.us) Date: Mon, 09 Dec 2013 16:21:51 -0600 Subject: [tclug-list] Count Number of Matched Patterns In-Reply-To: References: <20131206140936.Horde.XxBkDjEchZmc-Dos6FqGcw5@mail.dalan.us> <20131209104426.Horde.5eEJCuCn3VLsUTM4-1QfOA1@mail.dalan.us> Message-ID: <20131209162151.Horde.z2e0fk5kaSnh7wUIrjObYg1@mail.dalan.us> Quoting Mike Miller : > On Mon, 9 Dec 2013, canito at dalan.us wrote: > >> Quoting Mike Miller : >> >>> I'm not sure of exactly what your issue is. I'm thinking you need >>> the "--", so I made a script with that, exactly like this: >>> >>> >>> -------begin script on next line--------- >>> #!/bin/bash >>> >>> PATTERN=$1 >>> >>> grep -Eic -- "$PATTERN" >>> -------end script on previous line--------- >>> >>> >>> And I ran it: >>> >>> >>> $ echo -e "-testa\n-testb\nblah\nfoo\nbar\ntest\na-test" | ./script -test >>> 3 >>> >>> >>> That's the right answer. Note that if you don't have a filename, >>> you have to grep from stdin, which is what I did. >>> >>> Mike >> >> Thank you Mike- >> >> Not sure what the problem is either. I've had a couple of issues >> with egrep/grep on Red Hat. >> >> Again, I do appreciate the follow up. > > > What version of grep are you using? This is mine: > > $ grep --version > GNU grep 2.5.4 > > $ egrep --version > GNU grep 2.5.4 > > I read somewhere a few years ago that egrep was deprecated and that > we were supposed to use "grep -E" instead. I doubt that's the > problem (at least on my system, I get the same result with egrep as > with "grep -E"). > > I know that earlier versions of the perl grep, grep -P, were pretty > buggy but I've had better luck recently. > > Mike grep --version GNU grep 2.6.3 Pretty weird. I'll just assume user error. cat param.sh #!/bin/bash set -x echo $1 FUNC=$1 grep -Eic -- "$FUNC" ./param.sh --test + echo --test --test + FUNC=--test + grep -Eic -- --test ^C It just hangs, which is weird. I was under the impression that one could grep variables. SDA > _______________________________________________ > TCLUG Mailing List - Minneapolis/St. Paul, Minnesota > tclug-list at mn-linux.org > http://mailman.mn-linux.org/mailman/listinfo/tclug-list From chewie at wookimus.net Mon Dec 9 16:35:06 2013 From: chewie at wookimus.net (Chad Walstrom) Date: Mon, 09 Dec 2013 16:35:06 -0600 Subject: [tclug-list] Count Number of Matched Patterns In-Reply-To: <20131206140936.Horde.XxBkDjEchZmc-Dos6FqGcw5@mail.dalan.us> References: <20131206140936.Horde.XxBkDjEchZmc-Dos6FqGcw5@mail.dalan.us> Message-ID: <16537.1386628506@ydalir.wookimus> canito at dalan.us wrote: > In the process of writing a script which I would like to count the > number matched patterns (command line parameters). I think your approach can be refined. The `grep' family of commands would certainly work here, but not necessarily in the way you're envisioning. There are two basic forms of using grep grep -c PATTERN FILE COMMAND_OUTPUT | grep -c PATTERN The `-c' option gives you count, but only in terms of lines that match, not necessarily in terms of every matching instance. cat << EOMSG | grep -c ONE ONE, TWO, THREE ONE, ONE ONE TWO, ONE, ONE EOMSG will result in an answer of 4 rather than 6. Remember, counting lines rather than instances. If instances on a single line are important, such as with argument strings, you could do some pre-formatting tricks to still use grep. echo "ONE TWO ONE THREE ONE FOUR" | sed -e 's/[[:space:]]/\n/g' | \ grep -c ONE The `sed' command will replace all white space with newlines, allowing you to count lines to find matches in your arguments. There are other ways to skin this cat (pun intended) with shell-fu, but these are simple and easy to grep... (I can't stop.) Chad -- Chad Walstrom http://runswithd6s.blogspot.net/ From chewie at wookimus.net Mon Dec 9 19:17:23 2013 From: chewie at wookimus.net (Chad Walstrom) Date: Mon, 09 Dec 2013 19:17:23 -0600 Subject: [tclug-list] reattach stranded screen and/or prod less to write buffer In-Reply-To: References: Message-ID: <21612.1386638243@ydalir.wookimus> /run is on tmpfs for most systems these days. It's ethereal and disappears whenever you reboot. The folders and files thereunder are recreated by the services that need them. $ mount | grep "/run" tmpfs on /run type tmpfs (rw,nosuid,noexec,relatime,size=1639308k,mode=755) tmpfs on /run/lock type tmpfs (rw,nosuid,nodev,noexec,relatime,size=5120k) tmpfs on /run/shm type tmpfs (rw,nosuid,nodev,noexec,relatime,size=3278600k) Chad From mbmiller+l at gmail.com Mon Dec 9 19:22:44 2013 From: mbmiller+l at gmail.com (Mike Miller) Date: Mon, 9 Dec 2013 19:22:44 -0600 (CST) Subject: [tclug-list] Count Number of Matched Patterns In-Reply-To: <20131209162151.Horde.z2e0fk5kaSnh7wUIrjObYg1@mail.dalan.us> References: <20131206140936.Horde.XxBkDjEchZmc-Dos6FqGcw5@mail.dalan.us> <20131209104426.Horde.5eEJCuCn3VLsUTM4-1QfOA1@mail.dalan.us> <20131209162151.Horde.z2e0fk5kaSnh7wUIrjObYg1@mail.dalan.us> Message-ID: On Mon, 9 Dec 2013, canito at dalan.us wrote: > grep --version > GNU grep 2.6.3 > > Pretty weird. I'll just assume user error. > > cat param.sh > #!/bin/bash > set -x > > echo $1 > FUNC=$1 > > grep -Eic -- "$FUNC" > > ./param.sh --test > > + echo --test > --test > > + FUNC=--test > > + grep -Eic -- --test > ^C > > It just hangs, which is weird. I was under the impression that one could > grep variables. It hangs because grep has no input. You need to pump some data into it, or add another argument for a filename. Try this: echo -e "--test\na\nb\n--test\n--tested" | ./param.sh --test That should return "3" (without the quotes!). Mike From tclug1 at whitleymott.net Mon Dec 9 22:13:32 2013 From: tclug1 at whitleymott.net (gregrwm) Date: Mon, 9 Dec 2013 22:13:32 -0600 Subject: [tclug-list] reattach stranded screen and/or prod less to write buffer In-Reply-To: References: Message-ID: > But for your purposes, nothing should be clearing it. >... > But I don't know what distro you are using. Just some ideas. well, with screen launched before fsck finishes, when fsck finishes, the /run/screen/*/* entry gets deleted, by something. ubuntu saucy. if we can't find what's wiping /run/screen, perhaps i can hardlink to the pipe just after launching screen, if i get lucky and find a corner of /run that doesn't get wiped. or perhaps set the immutable bit. or maybe supply a softlink for /run/screen out of /run. -------------- next part -------------- An HTML attachment was scrubbed... URL: From tompoe at meltel.net Tue Dec 10 22:47:04 2013 From: tompoe at meltel.net (Tom Poe) Date: Tue, 10 Dec 2013 22:47:04 -0600 Subject: [tclug-list] which dvd holds ubuntu 12.10? Message-ID: <52A7EE48.8070604@meltel.net> Hello: Have an old desktop with ubuntu 12.04 currently. Want to figure out what my dvd device is, and how big a dvd I need to burn ubuntu 12.10 to replace 12.04. Any advice appreciated. Tom From tclug at freakzilla.com Tue Dec 10 23:08:07 2013 From: tclug at freakzilla.com (tclug at freakzilla.com) Date: Tue, 10 Dec 2013 23:08:07 -0600 (CST) Subject: [tclug-list] which dvd holds ubuntu 12.10? In-Reply-To: <52A7EE48.8070604@meltel.net> References: <52A7EE48.8070604@meltel.net> Message-ID: The device is probably /dev/dvd or /dev/sr0. You can tell what DVD size you need by looking at the file size for that Ubuntu dist, but you can probably install from a USB drive. Saves you a DVD. On Tue, 10 Dec 2013, Tom Poe wrote: > Hello: Have an old desktop with ubuntu 12.04 currently. Want to figure out > what my dvd device is, and how big a dvd I need to burn ubuntu 12.10 to > replace 12.04. Any advice appreciated. > Tom > > _______________________________________________ > TCLUG Mailing List - Minneapolis/St. Paul, Minnesota > tclug-list at mn-linux.org > http://mailman.mn-linux.org/mailman/listinfo/tclug-list > From ryanjcole at me.com Tue Dec 10 23:12:50 2013 From: ryanjcole at me.com (Ryan Coleman) Date: Tue, 10 Dec 2013 23:12:50 -0600 Subject: [tclug-list] which dvd holds ubuntu 12.10? In-Reply-To: References: <52A7EE48.8070604@meltel.net> Message-ID: <6B14D10B-5BB7-4696-BC25-A77E30D00ADD@me.com> Can?t speak for Ubuntu but usually if there?s a .img file it?s a DD you can write to a USB On Dec 10, 2013, at 11:08 PM, tclug at freakzilla.com wrote: > You can tell what DVD size you need by looking at the file size for that Ubuntu dist, but you can probably install from a USB drive. Saves you a DVD. -------------- next part -------------- An HTML attachment was scrubbed... URL: From tompoe at meltel.net Tue Dec 10 23:45:02 2013 From: tompoe at meltel.net (Tom Poe) Date: Tue, 10 Dec 2013 23:45:02 -0600 Subject: [tclug-list] which dvd holds ubuntu 12.10? In-Reply-To: References: <52A7EE48.8070604@meltel.net> Message-ID: <52A7FBDE.30906@meltel.net> On 12/10/2013 11:08 PM, tclug at freakzilla.com wrote: > > The device is probably /dev/dvd or /dev/sr0. > > You can tell what DVD size you need by looking at the file size for > that Ubuntu dist, but you can probably install from a USB drive. Saves > you a DVD. > > On Tue, 10 Dec 2013, Tom Poe wrote: > >> Hello: Have an old desktop with ubuntu 12.04 currently. Want to >> figure out what my dvd device is, and how big a dvd I need to burn >> ubuntu 12.10 to replace 12.04. Any advice appreciated. >> Tom >> >> _______________________________________________ >> TCLUG Mailing List - Minneapolis/St. Paul, Minnesota >> tclug-list at mn-linux.org >> http://mailman.mn-linux.org/mailman/listinfo/tclug-list >> > _______________________________________________ > TCLUG Mailing List - Minneapolis/St. Paul, Minnesota > tclug-list at mn-linux.org > http://mailman.mn-linux.org/mailman/listinfo/tclug-list > > I get this line when I run: sudo lshw capabilities: removable audio cd-r cd-rw dvd I would like to figure out what dvd to buy, e.g., dvd +R or some other ? I plan to download .iso image from ubuntu web site, then burn that to a dvd, then insert in dvd drive, and start up with cdrom option. Does that make sense? Tom From tclug at freakzilla.com Tue Dec 10 23:49:54 2013 From: tclug at freakzilla.com (tclug at freakzilla.com) Date: Tue, 10 Dec 2013 23:49:54 -0600 (CST) Subject: [tclug-list] which dvd holds ubuntu 12.10? In-Reply-To: <52A7FBDE.30906@meltel.net> References: <52A7EE48.8070604@meltel.net> <52A7FBDE.30906@meltel.net> Message-ID: On Tue, 10 Dec 2013, Tom Poe wrote: > I get this line when I run: sudo lshw > capabilities: removable audio cd-r cd-rw dvd > > I would like to figure out what dvd to buy, e.g., dvd +R or some other ? > > I plan to download .iso image from ubuntu web site, then burn that to a dvd, > then insert in dvd drive, and start up with cdrom option. Does that make > sense? I don't know that your drive can burn DVDs. Can you boot from USB? You might want to just do that. From tompoe at meltel.net Wed Dec 11 00:23:54 2013 From: tompoe at meltel.net (Tom Poe) Date: Wed, 11 Dec 2013 00:23:54 -0600 Subject: [tclug-list] which dvd holds ubuntu 12.10? In-Reply-To: References: <52A7EE48.8070604@meltel.net> <52A7FBDE.30906@meltel.net> Message-ID: <52A804FA.7020905@meltel.net> On 12/10/2013 11:49 PM, tclug at freakzilla.com wrote: > > I don't know that your drive can burn DVDs. > > Can you boot from USB? You might want to just do that. > _______________________________________________ > TCLUG Mailing List - Minneapolis/St. Paul, Minnesota > tclug-list at mn-linux.org > http://mailman.mn-linux.org/mailman/listinfo/tclug-list > > Aha! Google returned a result for an app called, UNetbootin ( http://unetbootin.sourceforge.net/ ) I'll buy a USB drive, and give it a go. Thanks, Tom From john.a.frisk at gmail.com Wed Dec 11 00:28:33 2013 From: john.a.frisk at gmail.com (John Frisk) Date: Wed, 11 Dec 2013 00:28:33 -0600 Subject: [tclug-list] 2014 Penguins Unbound Schedule Message-ID: Hello everyone, We had a small group with great conversation for the informal get together of Penguins Unbound tonight. Thanks to all who could make it. Please take a look at the 2014 schedule from the meetings tab. We have a few scheduled talks already and plan to fill out the agenda for the rest of the year in upcoming meetings. I plan on having more focused meetings in 2014. One track will be more for the hobbyist and one for IT professionals. Keep in mind everyone is welcome to all meetings but I want to help members of Penguins Unbound who may or may not have interest in a subject a better opportunity to decide which talks they want to hear about. This will also help our members that come from the greater MN area decide if they wish to commute. With that spirit in mind, the first two months will be (by chance) focused on IT professionals with ZFS in January and Chef in February. March we'll have an update with Arduino followed by Open Q&A in April. May will bring the Install for Ubuntu 14.04 LTS release. The rest of the year is open at this point and there is plenty of items on the Future Ideas page so we'll be prioritizing in future meetings. I hope everyone safely enjoys the holidays this end of year and look forward to 2014! John -------------- next part -------------- An HTML attachment was scrubbed... URL: From tclug at freakzilla.com Wed Dec 11 00:36:19 2013 From: tclug at freakzilla.com (tclug at freakzilla.com) Date: Wed, 11 Dec 2013 00:36:19 -0600 (CST) Subject: [tclug-list] which dvd holds ubuntu 12.10? In-Reply-To: <52A804FA.7020905@meltel.net> References: <52A7EE48.8070604@meltel.net> <52A7FBDE.30906@meltel.net> <52A804FA.7020905@meltel.net> Message-ID: I was going to suggest unetbootin, once you said you could boot from USB. Get an 8gb usb drive just in case, that should be plenty. Download the Ubuntu ISO, and make sure to mount the USB stick before you run unetbootin! On Wed, 11 Dec 2013, Tom Poe wrote: > On 12/10/2013 11:49 PM, tclug at freakzilla.com wrote: >> >> I don't know that your drive can burn DVDs. >> >> Can you boot from USB? You might want to just do that. >> _______________________________________________ >> TCLUG Mailing List - Minneapolis/St. Paul, Minnesota >> tclug-list at mn-linux.org >> http://mailman.mn-linux.org/mailman/listinfo/tclug-list >> >> > Aha! Google returned a result for an app called, UNetbootin ( > http://unetbootin.sourceforge.net/ ) > > I'll buy a USB drive, and give it a go. > Thanks, Tom > > > _______________________________________________ > TCLUG Mailing List - Minneapolis/St. Paul, Minnesota > tclug-list at mn-linux.org > http://mailman.mn-linux.org/mailman/listinfo/tclug-list > From tclug1 at whitleymott.net Wed Dec 11 02:43:37 2013 From: tclug1 at whitleymott.net (gregrwm) Date: Wed, 11 Dec 2013 02:43:37 -0600 Subject: [tclug-list] reattach stranded screen and/or prod less to write buffer In-Reply-To: References: Message-ID: i get that remount -rw of / has to await completion of fsck of /, but i wonder if mountall runs fsck even before the pivot_root happens. and i wonder if anything can get mounted before the remount -rw, like perhaps the /run init could be moved sooner. mounting anything before the pivot_root seems unlikely to fly, tho ya never know. early login easy, just start /etc/init/ssh.conf "on startup". and live with the caveats of early login! perhaps tmux worth another try. i'd run into something it couldn't do, tho i forget what. i'm feeling tempted to introduce a feature in screen to restore its /run/screen/*/* pipe if it disappears. -------------- next part -------------- An HTML attachment was scrubbed... URL: From blutgens at gmail.com Wed Dec 11 08:16:19 2013 From: blutgens at gmail.com (Ben) Date: Wed, 11 Dec 2013 08:16:19 -0600 Subject: [tclug-list] which dvd holds ubuntu 12.10? In-Reply-To: References: <52A7EE48.8070604@meltel.net> <52A7FBDE.30906@meltel.net> <52A804FA.7020905@meltel.net> Message-ID: Or just run "update-manager -d" i believe is the command to upgrade. On Wed, Dec 11, 2013 at 12:36 AM, wrote: > I was going to suggest unetbootin, once you said you could boot from USB. > > Get an 8gb usb drive just in case, that should be plenty. Download the > Ubuntu ISO, and make sure to mount the USB stick before you run unetbootin! > > > On Wed, 11 Dec 2013, Tom Poe wrote: > > On 12/10/2013 11:49 PM, tclug at freakzilla.com wrote: >> >>> >>> I don't know that your drive can burn DVDs. >>> >>> Can you boot from USB? You might want to just do that. >>> _______________________________________________ >>> TCLUG Mailing List - Minneapolis/St. Paul, Minnesota >>> tclug-list at mn-linux.org >>> http://mailman.mn-linux.org/mailman/listinfo/tclug-list >>> >>> >>> Aha! Google returned a result for an app called, UNetbootin ( >> http://unetbootin.sourceforge.net/ ) >> >> I'll buy a USB drive, and give it a go. >> Thanks, Tom >> >> >> _______________________________________________ >> TCLUG Mailing List - Minneapolis/St. Paul, Minnesota >> tclug-list at mn-linux.org >> http://mailman.mn-linux.org/mailman/listinfo/tclug-list >> >> _______________________________________________ > TCLUG Mailing List - Minneapolis/St. Paul, Minnesota > tclug-list at mn-linux.org > http://mailman.mn-linux.org/mailman/listinfo/tclug-list > -- Ben Lutgens Linux / Unix System Administrator Three of your friends throw up after eating chicken salad. Do you think: "I should find more robust friends" or "we should check that refrigerator"? -- Donald Becker, on vortex-bug, suspecting a network-wide problem -------------- next part -------------- An HTML attachment was scrubbed... URL: From tclug1 at whitleymott.net Wed Dec 11 09:40:22 2013 From: tclug1 at whitleymott.net (gregrwm) Date: Wed, 11 Dec 2013 09:40:22 -0600 Subject: [tclug-list] which dvd holds ubuntu 12.10? Message-ID: and there's always netboot. just download and boot the netboot linux&initrd pair. easy. the netboot installer is a tad annoying in that the questions it asks are dispersed amidst the long waits of the installation process. but it works just fine. another trick i use with netboot: install a minimal system, ie no *-desktop or other major software. keep it somewhere, tar it up if you like. later, just use it again for a new release by merely changing /etc/lsb-release (followed by apt-get update and apt-get --auto-remove dist-upgrade). this has never failed me, whereas the same trick is far more likely to fail if scads of packages are installed. (then apt-get install your favorite *-desktop or whatever) -------------- next part -------------- An HTML attachment was scrubbed... URL: From justin.kremer at gmail.com Wed Dec 11 10:48:41 2013 From: justin.kremer at gmail.com (Justin Kremer) Date: Wed, 11 Dec 2013 10:48:41 -0600 Subject: [tclug-list] which dvd holds ubuntu 12.10? In-Reply-To: References: Message-ID: If you are already running Ubuntu, and you don't like the update-manager option, Ubuntu has the capability to make USB startup disks built in. There is an application called Startup Disk Creator installed by default that you can just point at another Ubuntu .iso file and at a USB device that you don't mind wiping the filesystem on, and it makes it for you. It usually works fine with newer versions of Ubuntu than what you are currently running. I have used it many times. - Justin On Wed, Dec 11, 2013 at 9:40 AM, gregrwm wrote: > and there's always netboot. just download and boot the netboot linux&initrd > pair. easy. the netboot installer is a tad annoying in that the questions > it asks are dispersed amidst the long waits of the installation process. > but it works just fine. > > another trick i use with netboot: install a minimal system, ie no *-desktop > or other major software. keep it somewhere, tar it up if you like. later, > just use it again for a new release by merely changing /etc/lsb-release > (followed by apt-get update and apt-get --auto-remove dist-upgrade). this > has never failed me, whereas the same trick is far more likely to fail if > scads of packages are installed. (then apt-get install your favorite > *-desktop or whatever) > > _______________________________________________ > TCLUG Mailing List - Minneapolis/St. Paul, Minnesota > tclug-list at mn-linux.org > http://mailman.mn-linux.org/mailman/listinfo/tclug-list > From admin at lctn.org Wed Dec 11 11:05:05 2013 From: admin at lctn.org (Raymond Norton) Date: Wed, 11 Dec 2013 11:05:05 -0600 Subject: [tclug-list] Suexec script help Message-ID: <52A89B41.1050009@lctn.org> I have suexec setup on an internal server. I am able to launch a vlc script via php, just fine. ( ) However, I am not able to restart the vlc stream via a 2nd php page, unless I am at the counsel. I know its a permissions or environment issue. Any ideas how to fix it? I want to kill VLC (via a browser) , so I can relaunch the working script. -- Raymond Norton LCTN 952.955.7766 Sent from My Desktop From admin at lctn.org Wed Dec 11 11:18:57 2013 From: admin at lctn.org (Raymond Norton) Date: Wed, 11 Dec 2013 11:18:57 -0600 Subject: [tclug-list] Suexec script help In-Reply-To: <52A89B41.1050009@lctn.org> References: <52A89B41.1050009@lctn.org> Message-ID: <52A89E81.6080307@lctn.org> Fixed it by adding www-data to the SuexecUserGroup line. On 12/11/2013 11:05 AM, Raymond Norton wrote: > I have suexec setup on an internal server. I am able to launch a vlc > script via php, just fine. > > ( exec("./vlc-stream"); > ?> > ) > > However, I am not able to restart the vlc stream via a 2nd php page, > unless I am at the counsel. I know its a permissions or environment > issue. Any ideas how to fix it? > > I want to kill VLC (via a browser) , so I can relaunch the working > script. > > -- Raymond Norton LCTN 952.955.7766 Sent from My Desktop From stuporglue at gmail.com Thu Dec 12 10:04:02 2013 From: stuporglue at gmail.com (Michael Moore) Date: Thu, 12 Dec 2013 10:04:02 -0600 Subject: [tclug-list] Best way to convert a folder of png files into a pdf? Message-ID: I've got a folder of png files that I want to turn into a multi-page pdf with letter-sized pages. I'm currently trying to use imagemagick, but the resulting PDF is smaller than it should be. I've posted a message on ImageMagick's forums,[1] but I'm wondering if I'm not just using the wrong tool in the first place. Does anyone have a better tool to be using for this? Thanks, Michael Moore [1] http://www.imagemagick.org/discourse-server/viewtopic.php?f=1&t=24633 -------------- next part -------------- An HTML attachment was scrubbed... URL: From jmore at starmind.org Thu Dec 12 10:08:49 2013 From: jmore at starmind.org (Josh More) Date: Thu, 12 Dec 2013 10:08:49 -0600 Subject: [tclug-list] Best way to convert a folder of png files into a pdf? In-Reply-To: References: Message-ID: You need to do it in two steps: convert *.png test.mng convert test.mng test.pdf This is how I did my security comic book. The only gotcha is to check the page order with an "ls *.png" first. I had to preface each file with the pagenumber (00 - 24) to get them in the right order. -Josh On Thu, Dec 12, 2013 at 10:04 AM, Michael Moore wrote: > I've got a folder of png files that I want to turn into a multi-page pdf > with letter-sized pages. > > I'm currently trying to use imagemagick, but the resulting PDF is smaller > than it should be. I've posted a message on ImageMagick's forums,[1] but > I'm wondering if I'm not just using the wrong tool in the first place. > > Does anyone have a better tool to be using for this? > > Thanks, > Michael Moore > > > [1] http://www.imagemagick.org/discourse-server/viewtopic.php?f=1&t=24633 > > _______________________________________________ > TCLUG Mailing List - Minneapolis/St. Paul, Minnesota > tclug-list at mn-linux.org > http://mailman.mn-linux.org/mailman/listinfo/tclug-list > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From mbmiller+l at gmail.com Thu Dec 12 12:11:51 2013 From: mbmiller+l at gmail.com (Mike Miller) Date: Thu, 12 Dec 2013 12:11:51 -0600 (CST) Subject: [tclug-list] Best way to convert a folder of png files into a pdf? In-Reply-To: References: Message-ID: On Thu, 12 Dec 2013, Josh More wrote: > You need to do it in two steps: > > convert *.png test.mng > convert test.mng test.pdf > > This is how I did my security comic book. The only gotcha is to check > the page order with an "ls *.png" first. I had to preface each file > with the pagenumber (00 - 24) to get them in the right order. I'm not 100% sure that it would work for you, but here's a trick I sometimes use in this kind of situation (in Bash): convert $(\ls -1v *.png) test.mng The backslash turns of aliasing (which might be adding color to the text). The -v option uses "version" ordering of filenames. To see how this works, run these commands: mkdir foo cd foo touch {1..100}.txt ls ls -v cd .. rm -rf foo If there are spaces in filenames, I'm not sure that it works and you might have to do this first: IFS=" " Mike From stuporglue at gmail.com Thu Dec 12 12:30:31 2013 From: stuporglue at gmail.com (Michael Moore) Date: Thu, 12 Dec 2013 12:30:31 -0600 Subject: [tclug-list] Best way to convert a folder of png files into a pdf? In-Reply-To: References: Message-ID: On Thu, Dec 12, 2013 at 12:11 PM, Mike Miller wrote: > On Thu, 12 Dec 2013, Josh More wrote: > > You need to do it in two steps: >> >> convert *.png test.mng >> convert test.mng test.pdf >> >> This is how I did my security comic book. The only gotcha is to check >> the page order with an "ls *.png" first. I had to preface each file with >> the pagenumber (00 - 24) to get them in the right order. >> > I ended up getting the same results with both convert output/*.png output.pdf as with the two-step process. > I'm not 100% sure that it would work for you, but here's a trick I > sometimes use in this kind of situation (in Bash): > > convert $(\ls -1v *.png) test.mng > > The backslash turns of aliasing (which might be adding color to the text). > The -v option uses "version" ordering of filenames. > I'll have to remember that for the future. I had already sorted and named my pages. In the end I was able to work around the imagemagick page size issue I was having by doing an extra padding step to get all the images centered and the right size before converting to pdf. Thanks, Michael Moore -------------- next part -------------- An HTML attachment was scrubbed... URL: From nakorsakov at gmail.com Thu Dec 12 14:26:28 2013 From: nakorsakov at gmail.com (N K) Date: Thu, 12 Dec 2013 14:26:28 -0600 Subject: [tclug-list] Converting a folder of png files into a pdf. In-Reply-To: References: Message-ID: <52AA1BF4.1060009@gmail.com> If the other method didn't work, I think that I once used GIMP to convert a PNG to a PDF. It's worth a try. Even so, assuming that you can convert each PNG to a PDF, I can recommend this command line program (PDF tool kit....aka pdftk) for gluing together many individual PDFs into one PDF. I was using it on a Windows machine, but it seems like it comes also with Mac installer, and possibly Linux may have a package. If you are going to use it, read their example page, and possibly their man page, first. Not that hard to set up and get going. http://www.pdflabs.com/tools/pdftk-server/ If there is better free alternative out there for Linux (or cross-platform), please let me know. :-) -- Nicholas On 12/12/13 12:00 PM, tclug-list-request at mn-linux.org wrote: > Send tclug-list mailing list submissions to > tclug-list at mn-linux.org > > To subscribe or unsubscribe via the World Wide Web, visit > http://mailman.mn-linux.org/mailman/listinfo/tclug-list > or, via email, send a message with subject or body 'help' to > tclug-list-request at mn-linux.org > > You can reach the person managing the list at > tclug-list-owner at mn-linux.org > > When replying, please edit your Subject line so it is more specific > than "Re: Contents of tclug-list digest..." > > > Today's Topics: > > 1. Best way to convert a folder of png files into a pdf? > (Michael Moore) > 2. Re: Best way to convert a folder of png files into a pdf? > (Josh More) > > > ---------------------------------------------------------------------- > > Message: 1 > Date: Thu, 12 Dec 2013 10:04:02 -0600 > From: Michael Moore > To: TCLUG Mailing List > Subject: [tclug-list] Best way to convert a folder of png files into a > pdf? > Message-ID: > > Content-Type: text/plain; charset="utf-8" > > I've got a folder of png files that I want to turn into a multi-page pdf > with letter-sized pages. > > I'm currently trying to use imagemagick, but the resulting PDF is smaller > than it should be. I've posted a message on ImageMagick's forums,[1] but > I'm wondering if I'm not just using the wrong tool in the first place. > > Does anyone have a better tool to be using for this? > > Thanks, > Michael Moore > > > [1] http://www.imagemagick.org/discourse-server/viewtopic.php?f=1&t=24633 > -------------- next part -------------- > An HTML attachment was scrubbed... > URL: > > ------------------------------ > > Message: 2 > Date: Thu, 12 Dec 2013 10:08:49 -0600 > From: Josh More > To: TCLUG Mailing List > Subject: Re: [tclug-list] Best way to convert a folder of png files > into a pdf? > Message-ID: > > Content-Type: text/plain; charset="iso-8859-1" > > You need to do it in two steps: > > convert *.png test.mng > convert test.mng test.pdf > > This is how I did my security comic book. The only gotcha is to check the > page order with an "ls *.png" first. I had to preface each file with the > pagenumber (00 - 24) to get them in the right order. > > -Josh > > > > On Thu, Dec 12, 2013 at 10:04 AM, Michael Moore wrote: > >> I've got a folder of png files that I want to turn into a multi-page pdf >> with letter-sized pages. >> >> I'm currently trying to use imagemagick, but the resulting PDF is smaller >> than it should be. I've posted a message on ImageMagick's forums,[1] but >> I'm wondering if I'm not just using the wrong tool in the first place. >> >> Does anyone have a better tool to be using for this? >> >> Thanks, >> Michael Moore >> >> >> [1] http://www.imagemagick.org/discourse-server/viewtopic.php?f=1&t=24633 >> >> _______________________________________________ >> TCLUG Mailing List - Minneapolis/St. Paul, Minnesota >> tclug-list at mn-linux.org >> http://mailman.mn-linux.org/mailman/listinfo/tclug-list >> >> > -------------- next part -------------- > An HTML attachment was scrubbed... > URL: > > ------------------------------ > > _______________________________________________ > TCLUG Mailing List - Minneapolis/St. Paul, Minnesota > tclug-list at mn-linux.org > http://mailman.mn-linux.org/mailman/listinfo/tclug-list > > End of tclug-list Digest, Vol 108, Issue 14 > ******************************************* > From admin at lctn.org Thu Dec 12 14:32:44 2013 From: admin at lctn.org (Raymond Norton) Date: Thu, 12 Dec 2013 14:32:44 -0600 Subject: [tclug-list] Converting a folder of png files into a pdf. In-Reply-To: <52AA1BF4.1060009@gmail.com> References: <52AA1BF4.1060009@gmail.com> Message-ID: <52AA1D6C.7000104@lctn.org> Here is plugin for Gimp batch conversion http://registry.gimp.org/taxonomy/term/539 On 12/12/2013 02:26 PM, N K wrote: > If the other method didn't work, I think that I once used GIMP to > convert a PNG to a PDF. It's worth a try. > > Even so, assuming that you can convert each PNG to a PDF, I can > recommend this command line program (PDF tool kit....aka pdftk) for > gluing together many individual PDFs into one PDF. I was using it on a > Windows machine, but it seems like it comes also with Mac installer, and > possibly Linux may have a package. > > If you are going to use it, read their example page, and possibly their > man page, first. > Not that hard to set up and get going. > > http://www.pdflabs.com/tools/pdftk-server/ > > If there is better free alternative out there for Linux (or > cross-platform), please let me know. :-) > > -- Nicholas > > On 12/12/13 12:00 PM, tclug-list-request at mn-linux.org wrote: >> Send tclug-list mailing list submissions to >> tclug-list at mn-linux.org >> >> To subscribe or unsubscribe via the World Wide Web, visit >> http://mailman.mn-linux.org/mailman/listinfo/tclug-list >> or, via email, send a message with subject or body 'help' to >> tclug-list-request at mn-linux.org >> >> You can reach the person managing the list at >> tclug-list-owner at mn-linux.org >> >> When replying, please edit your Subject line so it is more specific >> than "Re: Contents of tclug-list digest..." >> >> >> Today's Topics: >> >> 1. Best way to convert a folder of png files into a pdf? >> (Michael Moore) >> 2. Re: Best way to convert a folder of png files into a pdf? >> (Josh More) >> >> >> ---------------------------------------------------------------------- >> >> Message: 1 >> Date: Thu, 12 Dec 2013 10:04:02 -0600 >> From: Michael Moore >> To: TCLUG Mailing List >> Subject: [tclug-list] Best way to convert a folder of png files into a >> pdf? >> Message-ID: >> >> Content-Type: text/plain; charset="utf-8" >> >> I've got a folder of png files that I want to turn into a multi-page pdf >> with letter-sized pages. >> >> I'm currently trying to use imagemagick, but the resulting PDF is smaller >> than it should be. I've posted a message on ImageMagick's forums,[1] but >> I'm wondering if I'm not just using the wrong tool in the first place. >> >> Does anyone have a better tool to be using for this? >> >> Thanks, >> Michael Moore >> >> >> [1] http://www.imagemagick.org/discourse-server/viewtopic.php?f=1&t=24633 >> -------------- next part -------------- >> An HTML attachment was scrubbed... >> URL: >> >> ------------------------------ >> >> Message: 2 >> Date: Thu, 12 Dec 2013 10:08:49 -0600 >> From: Josh More >> To: TCLUG Mailing List >> Subject: Re: [tclug-list] Best way to convert a folder of png files >> into a pdf? >> Message-ID: >> >> Content-Type: text/plain; charset="iso-8859-1" >> >> You need to do it in two steps: >> >> convert *.png test.mng >> convert test.mng test.pdf >> >> This is how I did my security comic book. The only gotcha is to check the >> page order with an "ls *.png" first. I had to preface each file with the >> pagenumber (00 - 24) to get them in the right order. >> >> -Josh >> >> >> >> On Thu, Dec 12, 2013 at 10:04 AM, Michael Moore wrote: >> >>> I've got a folder of png files that I want to turn into a multi-page pdf >>> with letter-sized pages. >>> >>> I'm currently trying to use imagemagick, but the resulting PDF is smaller >>> than it should be. I've posted a message on ImageMagick's forums,[1] but >>> I'm wondering if I'm not just using the wrong tool in the first place. >>> >>> Does anyone have a better tool to be using for this? >>> >>> Thanks, >>> Michael Moore >>> >>> >>> [1] http://www.imagemagick.org/discourse-server/viewtopic.php?f=1&t=24633 >>> >>> _______________________________________________ >>> TCLUG Mailing List - Minneapolis/St. Paul, Minnesota >>> tclug-list at mn-linux.org >>> http://mailman.mn-linux.org/mailman/listinfo/tclug-list >>> >>> >> -------------- next part -------------- >> An HTML attachment was scrubbed... >> URL: >> >> ------------------------------ >> >> _______________________________________________ >> TCLUG Mailing List - Minneapolis/St. Paul, Minnesota >> tclug-list at mn-linux.org >> http://mailman.mn-linux.org/mailman/listinfo/tclug-list >> >> End of tclug-list Digest, Vol 108, Issue 14 >> ******************************************* >> > _______________________________________________ > TCLUG Mailing List - Minneapolis/St. Paul, Minnesota > tclug-list at mn-linux.org > http://mailman.mn-linux.org/mailman/listinfo/tclug-list > -- Raymond Norton LCTN 952.955.7766 Sent from My Desktop -------------- next part -------------- An HTML attachment was scrubbed... URL: From stevetrapp at comcast.net Thu Dec 12 15:49:43 2013 From: stevetrapp at comcast.net (Steve Trapp) Date: Thu, 12 Dec 2013 15:49:43 -0600 Subject: [tclug-list] Best way to convert a folder of png files into a pdf? In-Reply-To: References: Message-ID: <20131212154943.407376fe@falcon.cavelan.local> Josh, Mike, and alls-y'all-T.C.LUG-gers-following-this-thread- There are multiple ways of converting .png-files to .pdf-files. I end up using *gimp* a lot for this. I'm sure other ways work just fine. However, when I'm done, I have one .pdf file for each .png file. To catenate the pages together into a single .pdf file, I use the tool *pdftk* as follows: pdftk .pdf .pdf ... .pdf cat output .pdf Naturally, you can use wild cards if your .pdf-pages have a sequence to them. Something like: pdftk [0-9].pdf [0-9][0-9].pdf [0-9][0-9][0-9].pdf \ cat output .pdf The above is an extreme example for the case when you don't have LEADING ZEROES. It's also a bad idea if you REALLY have more than a 100 files because there's a limit (usually, anyway) to the number of characters in a command line (historically, I believe, it was 4096). Hope this helps, -Steve P.S.- Sometimes things end up with a page type of A4 instead of letterSize (a problem in the US and Canada), or end up as letterSize instead of A4 (a problem outside of {.us, .ca}). One of {A4, letter} is longer, and the other is wider. The differences in {length, width} are, I believe, less than 1_inch (2.54_cm). On Thu, 12 Dec 2013 12:30:31 -0600, Michael Moore wrote: > On Thu, Dec 12, 2013 at 12:11 PM, Mike Miller > wrote: > > > On Thu, 12 Dec 2013, Josh More wrote: > > > > You need to do it in two steps: > >> > >> convert *.png test.mng > >> convert test.mng test.pdf > >> > >> This is how I did my security comic book. The only gotcha is to check > >> the page order with an "ls *.png" first. I had to preface each file > >> with the pagenumber (00 - 24) to get them in the right order. > >> > > > I ended up getting the same results with both > > convert output/*.png output.pdf > as with the two-step process. > > > > I'm not 100% sure that it would work for you, but here's a trick I > > sometimes use in this kind of situation (in Bash): > > > > convert $(\ls -1v *.png) test.mng > > > > The backslash turns of aliasing (which might be adding color to the > > text). The -v option uses "version" ordering of filenames. > > > > I'll have to remember that for the future. I had already sorted and named > my pages. > > In the end I was able to work around the imagemagick page size issue I was > having by doing an extra padding step to get all the images centered and > the right size before converting to pdf. > > Thanks, > Michael Moore -- Name: Steve Trapp Homepage: http://steventrapp.home.comcast.net Email: stevetrapp **AT** comcast **DOT** net Locale: en_US.UTF-8 | Location: Upper Midwest From jmore at starmind.org Thu Dec 12 15:59:20 2013 From: jmore at starmind.org (Josh More) Date: Thu, 12 Dec 2013 15:59:20 -0600 Subject: [tclug-list] Best way to convert a folder of png files into a pdf? In-Reply-To: <20131212154943.407376fe@falcon.cavelan.local> References: <20131212154943.407376fe@falcon.cavelan.local> Message-ID: Yes, that's a way to do it. I've used pdftk after I generate the PDF using ImageMagick to mess with pagination. It's a good tool too. -Josh On Thu, Dec 12, 2013 at 3:49 PM, Steve Trapp wrote: > Josh, Mike, and alls-y'all-T.C.LUG-gers-following-this-thread- > > There are multiple ways of converting .png-files to .pdf-files. I end up > using *gimp* a lot for this. I'm sure other ways work just fine. > > However, when I'm done, I have one .pdf file for each .png file. > > To catenate the pages together into a single .pdf file, I use the tool > *pdftk* as follows: > > pdftk .pdf .pdf ... .pdf cat output .pdf > > Naturally, you can use wild cards if your .pdf-pages have a sequence to > them. Something like: > > pdftk [0-9].pdf [0-9][0-9].pdf [0-9][0-9][0-9].pdf \ > cat output .pdf > > The above is an extreme example for the case when you don't have LEADING > ZEROES. It's also a bad idea if you REALLY have more than a 100 files > because there's a limit (usually, anyway) to the number of characters in a > command line (historically, I believe, it was 4096). > > Hope this helps, > -Steve > > P.S.- Sometimes things end up with a page type of A4 instead of letterSize > (a problem in the US and Canada), or end up as letterSize instead of A4 (a > problem outside of {.us, .ca}). One of {A4, letter} is longer, and the > other is wider. The differences in {length, width} are, I believe, less > than 1_inch (2.54_cm). > > On Thu, 12 Dec 2013 12:30:31 -0600, > Michael Moore wrote: > > On Thu, Dec 12, 2013 at 12:11 PM, Mike Miller > > wrote: > > > > > On Thu, 12 Dec 2013, Josh More wrote: > > > > > > You need to do it in two steps: > > >> > > >> convert *.png test.mng > > >> convert test.mng test.pdf > > >> > > >> This is how I did my security comic book. The only gotcha is to check > > >> the page order with an "ls *.png" first. I had to preface each file > > >> with the pagenumber (00 - 24) to get them in the right order. > > >> > > > > > I ended up getting the same results with both > > > > convert output/*.png output.pdf > > as with the two-step process. > > > > > > > I'm not 100% sure that it would work for you, but here's a trick I > > > sometimes use in this kind of situation (in Bash): > > > > > > convert $(\ls -1v *.png) test.mng > > > > > > The backslash turns of aliasing (which might be adding color to the > > > text). The -v option uses "version" ordering of filenames. > > > > > > > I'll have to remember that for the future. I had already sorted and named > > my pages. > > > > In the end I was able to work around the imagemagick page size issue I > was > > having by doing an extra padding step to get all the images centered and > > the right size before converting to pdf. > > > > Thanks, > > Michael Moore > > -- > Name: Steve Trapp > Homepage: http://steventrapp.home.comcast.net > Email: stevetrapp **AT** comcast **DOT** net > Locale: en_US.UTF-8 | Location: Upper Midwest > _______________________________________________ > TCLUG Mailing List - Minneapolis/St. Paul, Minnesota > tclug-list at mn-linux.org > http://mailman.mn-linux.org/mailman/listinfo/tclug-list > -------------- next part -------------- An HTML attachment was scrubbed... URL: From tclug1 at whitleymott.net Thu Dec 12 16:06:33 2013 From: tclug1 at whitleymott.net (gregrwm) Date: Thu, 12 Dec 2013 16:06:33 -0600 Subject: [tclug-list] Best way to convert a folder of png files into a pdf? Message-ID: i do prefer commandline, but convert barfed one too many times for me. i've used gimp since. then i pull each image into a libreoffice writer doc via page background, one per pagestyle. a kluge, but better than a pdf for each page, and can easily add commentary, and makes a good pdf. tho i'll be delighted to hear of&try new ways.. -------------- next part -------------- An HTML attachment was scrubbed... URL: From mbmiller+l at gmail.com Thu Dec 12 21:05:26 2013 From: mbmiller+l at gmail.com (Mike Miller) Date: Thu, 12 Dec 2013 21:05:26 -0600 (CST) Subject: [tclug-list] Best way to convert a folder of png files into a pdf? In-Reply-To: References: Message-ID: On Thu, 12 Dec 2013, Michael Moore wrote: > I've got a folder of png files that I want to turn into a multi-page pdf > with letter-sized pages. > > I'm currently trying to use imagemagick, but the resulting PDF is > smaller than it should be. I've posted a message on ImageMagick's > forums,[1] but I'm wondering if I'm not just using the wrong tool in the > first place. > > Does anyone have a better tool to be using for this? > > [1] http://www.imagemagick.org/discourse-server/viewtopic.php?f=1&t=24633 You have to know the dpi for the images: identify -format "%x x %y\n" output/*.png | less Are they all the same? Are they all in PixelsPerInch? If they are different, you might want to fix that by doing something like this: mkdir output2 cd output for file in *.png ; do convert -units PixelsPerInch -density 300 "$file" ../output2/"$file" ; done cd .. If the images were all the same size in pixels, you can choose the density to get the correct size on the page. Otherwise you might have to run them one at a time or in groups. I would stick with convert, when possible. If it sometimes gives a bad result, I think that's because there is a trick you haven't learned yet. Mike From mbmiller+l at gmail.com Thu Dec 12 21:05:26 2013 From: mbmiller+l at gmail.com (Mike Miller) Date: Thu, 12 Dec 2013 21:05:26 -0600 (CST) Subject: [tclug-list] Best way to convert a folder of png files into a pdf? In-Reply-To: References: Message-ID: On Thu, 12 Dec 2013, Michael Moore wrote: > I've got a folder of png files that I want to turn into a multi-page pdf > with letter-sized pages. > > I'm currently trying to use imagemagick, but the resulting PDF is > smaller than it should be. I've posted a message on ImageMagick's > forums,[1] but I'm wondering if I'm not just using the wrong tool in the > first place. > > Does anyone have a better tool to be using for this? > > [1] http://www.imagemagick.org/discourse-server/viewtopic.php?f=1&t=24633 You have to know the dpi for the images: identify -format "%x x %y\n" output/*.png | less Are they all the same? Are they all in PixelsPerInch? If they are different, you might want to fix that by doing something like this: mkdir output2 cd output for file in *.png ; do convert -units PixelsPerInch -density 300 "$file" ../output2/"$file" ; done cd .. If the images were all the same size in pixels, you can choose the density to get the correct size on the page. Otherwise you might have to run them one at a time or in groups. I would stick with convert, when possible. If it sometimes gives a bad result, I think that's because there is a trick you haven't learned yet. Mike From brad+tclug at mifflinet.net Sun Dec 15 22:05:19 2013 From: brad+tclug at mifflinet.net (Brad) Date: Sun, 15 Dec 2013 22:05:19 -0600 Subject: [tclug-list] reattach stranded screen and/or prod less to write buffer In-Reply-To: References: Message-ID: <20131216040519.GD25641@euler.llan.mifflinet.net> On Wed Dec 11, 2013 at 02:43:37AM -0600, gregrwm wrote: > [snip] > i'm feeling tempted to introduce a feature in screen to restore its > /run/screen/*/* pipe if it disappears. Send a CHLD to the master screen process. It should re-create the pipe. From erik.mitchell at gmail.com Mon Dec 16 13:48:11 2013 From: erik.mitchell at gmail.com (Erik Mitchell) Date: Mon, 16 Dec 2013 13:48:11 -0600 Subject: [tclug-list] Recommendations on a backup server? Message-ID: Hi everyone, I am working on some end of year projects for my business and one of those is to get a backup server going that will connect to all of my various machines and perform regular backups. The last time I've looked into this "rsnapshot" was one of the more nifty and free as in speech solutions. I'm thinking I'll build a box with three 1-2TB drives, in a RAID5 configuration, and have that connect to my various machines using a "backup" account. Each machine will get backed up to my backup server, and then in the future I'll also get an offsite machine set up to sync with my backup server. I'm not an expert in this, and I'm guessing a lot of folks on the list might have some better ideas. I'd like to stick with "free as in speech" solutions, and also do the storage on real hardware that I own -- no cloud solutions. If anyone has any suggestions please share! Thanks, Erik -- Erik K. Mitchell erik.mitchell at gmail.com From admin at lctn.org Mon Dec 16 14:09:17 2013 From: admin at lctn.org (Raymond Norton) Date: Mon, 16 Dec 2013 14:09:17 -0600 Subject: [tclug-list] Recommendations on a backup server? In-Reply-To: References: Message-ID: <52AF5DED.8020809@lctn.org> Bacula VM with NFS mounted drives. On 12/16/2013 01:48 PM, Erik Mitchell wrote: > Hi everyone, > I am working on some end of year projects for my business and one of > those is to get a backup server going that will connect to all of my > various machines and perform regular backups. > > The last time I've looked into this "rsnapshot" was one of the more > nifty and free as in speech solutions. I'm thinking I'll build a box > with three 1-2TB drives, in a RAID5 configuration, and have that > connect to my various machines using a "backup" account. Each machine > will get backed up to my backup server, and then in the future I'll > also get an offsite machine set up to sync with my backup server. > > I'm not an expert in this, and I'm guessing a lot of folks on the list > might have some better ideas. I'd like to stick with "free as in > speech" solutions, and also do the storage on real hardware that I own > -- no cloud solutions. If anyone has any suggestions please share! > > Thanks, > > Erik > -- Raymond Norton LCTN 952.955.7766 Sent from My Desktop From jjensen at apache.org Mon Dec 16 14:13:10 2013 From: jjensen at apache.org (Jeff Jensen) Date: Mon, 16 Dec 2013 14:13:10 -0600 Subject: [tclug-list] Recommendations on a backup server? In-Reply-To: References: Message-ID: I ran BackupPC for many years on an old machine. Worked great. Someone else on this list recommended it to me "way back"... When the OS drive on it had enough (failed), I did not fix/setup a new one; instead, went directly to cloud. It's really nice not having to manage another box. That's for personal. Now it's the same for our corporate too... On Mon, Dec 16, 2013 at 1:48 PM, Erik Mitchell wrote: > Hi everyone, > I am working on some end of year projects for my business and one of > those is to get a backup server going that will connect to all of my > various machines and perform regular backups. > > The last time I've looked into this "rsnapshot" was one of the more > nifty and free as in speech solutions. I'm thinking I'll build a box > with three 1-2TB drives, in a RAID5 configuration, and have that > connect to my various machines using a "backup" account. Each machine > will get backed up to my backup server, and then in the future I'll > also get an offsite machine set up to sync with my backup server. > > I'm not an expert in this, and I'm guessing a lot of folks on the list > might have some better ideas. I'd like to stick with "free as in > speech" solutions, and also do the storage on real hardware that I own > -- no cloud solutions. If anyone has any suggestions please share! > > Thanks, > > Erik > > -- > Erik K. Mitchell > erik.mitchell at gmail.com > _______________________________________________ > TCLUG Mailing List - Minneapolis/St. Paul, Minnesota > tclug-list at mn-linux.org > http://mailman.mn-linux.org/mailman/listinfo/tclug-list > -------------- next part -------------- An HTML attachment was scrubbed... URL: From tonyyarusso at gmail.com Mon Dec 16 16:14:19 2013 From: tonyyarusso at gmail.com (Tony Yarusso) Date: Mon, 16 Dec 2013 16:14:19 -0600 Subject: [tclug-list] Recommendations on a backup server? In-Reply-To: References: Message-ID: On Mon, Dec 16, 2013 at 1:48 PM, Erik Mitchell wrote: > I am working on some end of year projects for my business and one of > those is to get a backup server going that will connect to all of my > various machines and perform regular backups. How many machines are we talking about, and what OSs do they run? If it's just a couple of personal things, something like BackupPC is probably best. But, since you said business, if you have a dozen or more I'd definitely look at stuff like Bacula and Amanda. I've waffled back and forth on the feature lists of those two a bit, and currently am leaning Amanda. From erikerik at gmail.com Mon Dec 16 16:42:44 2013 From: erikerik at gmail.com (Erik Anderson) Date: Mon, 16 Dec 2013 16:42:44 -0600 Subject: [tclug-list] Recommendations on a backup server? In-Reply-To: <52AF5DED.8020809@lctn.org> References: <52AF5DED.8020809@lctn.org> Message-ID: On Mon, Dec 16, 2013 at 2:09 PM, Raymond Norton wrote: > Bacula VM with NFS mounted drives. Negative, captain. :) Bacula has gone rogue[1]. Check into Bareos, which is a 100% F/OSS fork of Bacula. That said, if I were building a backup server now, it would likely be a FreeBSD box w/ a big 'ol ZFS filesystem. I'd rsync my various servers into folders on the backup server and then take a ZFS snapshot after each rsync. This achieves something similar to what you'd get with rsnapshot, but with the advantage of being able to use the more advanced features of ZFS: end-to-end data verification and healing (bit rot detection), snapshots, zfs send/receive (quite useful for when you want to ship your backups offsite), compression, etc. -Erik [1] http://www.bareos.org/en/faq/items/why_fork.html -------------- next part -------------- An HTML attachment was scrubbed... URL: From droidjd at gmail.com Mon Dec 16 16:46:45 2013 From: droidjd at gmail.com (Andrew Dahl) Date: Mon, 16 Dec 2013 16:46:45 -0600 Subject: [tclug-list] Recommendations on a backup server? In-Reply-To: References: Message-ID: Another +1 for Bacula. When I worked on campus a few years back, I setup a configuration that used Bacula to backup a few servers to tapes. Worked really well, in my opinion. For smaller stuff, CrashPlan offers their utility to use for free (backup of unlimited computers to one computer, and then if you wanted to pay for their off-site usage, you can push all of that as well. It's what I currently use at home with my family. RAID-5 on my server and then everything's off-site as well in case of severe hardware failure or a fire) On Mon, Dec 16, 2013 at 1:48 PM, Erik Mitchell wrote: > Hi everyone, > I am working on some end of year projects for my business and one of > those is to get a backup server going that will connect to all of my > various machines and perform regular backups. > > The last time I've looked into this "rsnapshot" was one of the more > nifty and free as in speech solutions. I'm thinking I'll build a box > with three 1-2TB drives, in a RAID5 configuration, and have that > connect to my various machines using a "backup" account. Each machine > will get backed up to my backup server, and then in the future I'll > also get an offsite machine set up to sync with my backup server. > > I'm not an expert in this, and I'm guessing a lot of folks on the list > might have some better ideas. I'd like to stick with "free as in > speech" solutions, and also do the storage on real hardware that I own > -- no cloud solutions. If anyone has any suggestions please share! > > Thanks, > > Erik > > -- > Erik K. Mitchell > erik.mitchell at gmail.com > _______________________________________________ > TCLUG Mailing List - Minneapolis/St. Paul, Minnesota > tclug-list at mn-linux.org > http://mailman.mn-linux.org/mailman/listinfo/tclug-list > -------------- next part -------------- An HTML attachment was scrubbed... URL: From cschumann at twp-llc.com Mon Dec 16 17:22:15 2013 From: cschumann at twp-llc.com (Chris Schumann) Date: Mon, 16 Dec 2013 17:22:15 -0600 Subject: [tclug-list] Recommendations on a backup server? In-Reply-To: References: Message-ID: <52AF8B27.507@twp-llc.com> Andrew wrote: > CrashPlan The thing I don't get about CrashPlan is this: What are you supposed to do if your hard disk dies? How long does it take to get me working again? Chris From droidjd at gmail.com Mon Dec 16 17:37:11 2013 From: droidjd at gmail.com (Andrew Dahl) Date: Mon, 16 Dec 2013 17:37:11 -0600 Subject: [tclug-list] Recommendations on a backup server? In-Reply-To: <52AF8B27.507@twp-llc.com> References: <52AF8B27.507@twp-llc.com> Message-ID: That all depends on your Internet connection and how much data you have pushed. To restore the data, you'd put the utility on your system, login, and use the restoration feature to pull your files down. So, from their servers, it would take as long as you downloading all your files again. If you're just pulling it from your local server, then it would depend on your network and any hardware bottlenecks. (I've never tested how fast CrashPlan is at pulling the files locally, so I can't add anything useful to that.) Unrelated: I've actually used CrashPlan's restore utilities to pull down some files I needed when I didn't have access to my server. Worked pretty slick. I /think/ you can actually access the files through your mobile device and their app as well. For me, it works well and I'd recommend it. I have several terabytes of data backed up though, so for the price, I don't see an easy way to beat it as an off-site backup. That said, it took a really long time to get all of that uploaded since I opted to not send them an external drive. And come to think of it, a friend of mine set it up at a small business with 6 machines backing up to 1. He seemed to be pretty impressed at how slick it was compared to other solutions. -Andrew On Mon, Dec 16, 2013 at 5:22 PM, Chris Schumann wrote: > Andrew wrote: > > CrashPlan > > The thing I don't get about CrashPlan is this: What are you supposed to do > if your hard disk dies? How long does it take to get me working again? > > Chris > > > _______________________________________________ > TCLUG Mailing List - Minneapolis/St. Paul, Minnesota > tclug-list at mn-linux.org > http://mailman.mn-linux.org/mailman/listinfo/tclug-list > -------------- next part -------------- An HTML attachment was scrubbed... URL: From chewie at wookimus.net Mon Dec 16 19:21:11 2013 From: chewie at wookimus.net (Chad Walstrom) Date: Mon, 16 Dec 2013 19:21:11 -0600 Subject: [tclug-list] Recommendations on a backup server? In-Reply-To: References: <52AF5DED.8020809@lctn.org> Message-ID: <7466.1387243271@ydalir.wookimus> Erik Anderson wrote: >> Bacula VM with NFS mounted drives. > > Negative, captain. :) > > Bacula has gone rogue[1]. Check into Bareos, which is a 100% F/OSS > fork of Bacula. Interesting. I implemented a pair of Bacula 8U multi-terrabyte servers when working at Zayo Managed Services (was Onvoy) in 2006 to replace a Legato server and tape silo. Worked great! As a traditional style network backup solution, Bacula worked wonderfully. It still suffered from typical storage issues of traditional backup solutions that rely upon full, incremental, and differential backups. It also required you to know how to maintain a MySQL server or PostgreSQL server to store metadata about your backups. I haven't kept up to date with it, and if it has indeed gone rogue, probably not the best option. I've also tried BackupPC and liked it. Worked well on my Linux machines, but I never really managed to spend time to get it working on my wife's Mac. I currently am a paying customer of CrashPlan and like the service. It's a local company of awesome dudes just trying to write great software. Yes, it's commercial, but not everything in life has to be Free as in Libre. I'm paying them to keep an up to date copy of all of my computers on their network; I figure I'm getting my money's worth. To augment CrashPlan and to address a different issue, synchronizing files, I've been using git-annex[1], written by Joey Hess in Haskel. It uses git to store and synchronize metadata and a plugin (annex) to copy the files around (doing cool things like ensuring multiple copies exist, track off-line volumes, chose media types and or file types for different annexes, etc). It keeps my Keepassx password files, MP3, and photos synchronized between my phone, laptop, and kid's computer. I plan on using it to synchronize to my wife's computer for media sharing, etc. It takes a bit to wrap your head around it, but once you understand how the files and metadata works, you're golden. I've also been eyeballing bup[2]. Another git-based backup utility that operates similar to other backup solutions in that you don't need to great git repositories in the directories you want to track. Seems useful! Additionally, there are ways of tying bup and git-annex together - for example using bup as a backup repository for git-annex. :) Fun stuff. [1] http://git-annex.branchable.com/ [2] https://github.com/bup/bup -- Chad From tclug at freakzilla.com Mon Dec 16 19:31:13 2013 From: tclug at freakzilla.com (tclug at freakzilla.com) Date: Mon, 16 Dec 2013 19:31:13 -0600 (CST) Subject: [tclug-list] Recommendations on a backup server? In-Reply-To: <7466.1387243271@ydalir.wookimus> References: <52AF5DED.8020809@lctn.org> <7466.1387243271@ydalir.wookimus> Message-ID: On Mon, 16 Dec 2013, Chad Walstrom wrote: > I currently am a paying customer of CrashPlan and like the service. I've been wondering about CrashPlan... when they say "unlimited" cloud storage, do they mean it? I mean can I dump like 16tb on there?... -- From chewie at wookimus.net Mon Dec 16 21:36:29 2013 From: chewie at wookimus.net (Chad Walstrom) Date: Mon, 16 Dec 2013 21:36:29 -0600 Subject: [tclug-list] Recommendations on a backup server? In-Reply-To: References: <52AF5DED.8020809@lctn.org> <7466.1387243271@ydalir.wookimus> Message-ID: Indeed. Unlimited means no limit. ;) tclug at freakzilla.com wrote: >On Mon, 16 Dec 2013, Chad Walstrom wrote: > >> I currently am a paying customer of CrashPlan and like the service. > >I've been wondering about CrashPlan... when they say "unlimited" cloud >storage, do they mean it? I mean can I dump like 16tb on there?... > >-- >_______________________________________________ >TCLUG Mailing List - Minneapolis/St. Paul, Minnesota >tclug-list at mn-linux.org >http://mailman.mn-linux.org/mailman/listinfo/tclug-list -- Chad Walstrom -------------- next part -------------- An HTML attachment was scrubbed... URL: From adam at askewview.net Tue Dec 17 09:21:02 2013 From: adam at askewview.net (Adam Barthelemy) Date: Tue, 17 Dec 2013 09:21:02 -0600 Subject: [tclug-list] =?utf-8?q?Recommendations_on_a_backup_server=3F?= In-Reply-To: References: <52AF5DED.8020809@lctn.org> <7466.1387243271@ydalir.wookimus> Message-ID: <98d96eb9d972c5606d6ef2d0a74fd23e@askewview.net> I'm on their "Family" plan and backing up over 2TB between all of my machines. Now 16TB is going to take a long time to get there. They have an option to order a "seeded backup" but it is only for 1TB of data the rest would have to be transferred over the internet. They also offer a "restore to door" service which allows you to retrieve up to 3.5TB of data via a drive shipped to whatever desired location. On 2013-12-16 19:31, tclug at freakzilla.com wrote: > On Mon, 16 Dec 2013, Chad Walstrom wrote: > >> I currently am a paying customer of CrashPlan and like the service. > > I've been wondering about CrashPlan... when they say "unlimited" > cloud storage, do they mean it? I mean can I dump like 16tb on > there?... > > -- > _______________________________________________ > TCLUG Mailing List - Minneapolis/St. Paul, Minnesota > tclug-list at mn-linux.org > http://mailman.mn-linux.org/mailman/listinfo/tclug-list From woodbrian77 at gmail.com Tue Dec 17 14:08:25 2013 From: woodbrian77 at gmail.com (Brian Wood) Date: Tue, 17 Dec 2013 14:08:25 -0600 Subject: [tclug-list] Hope for your project Message-ID: In June I wrote: > Is your C++ project on the ropes? I'm willing to donate 15 hours > a week for up to six months on a project that uses the C++ > Middleware Writer (CMW). The CMW is an on line code > generator that writes low-level C++ marshalling code based on > high-level user input. The CMW is an increasingly robust producer > of concrete code. In addition to the above offer, I'll give $500 cash and a $1,000 investment in my company for help finding someone interested in this. I'll pay the cash part after working with the person or team for four months. The company rewards investments to 3 times their original amount, so the investment will result in between $0 and $3,000, depending on how things go for the company. -- Brian Ebenezer Enterprises - In G-d we trust. http://webEbenezer.net -------------- next part -------------- An HTML attachment was scrubbed... URL: From tclug1 at whitleymott.net Tue Dec 17 14:54:44 2013 From: tclug1 at whitleymott.net (gregrwm) Date: Tue, 17 Dec 2013 14:54:44 -0600 Subject: [tclug-list] reattach stranded screen and/or prod less to write buffer Message-ID: > > Send a CHLD to the master screen process. It should re-create the pipe. > yay! thanks. -------------- next part -------------- An HTML attachment was scrubbed... URL: From cschumann at twp-llc.com Tue Dec 17 14:57:32 2013 From: cschumann at twp-llc.com (Chris Schumann) Date: Tue, 17 Dec 2013 14:57:32 -0600 Subject: [tclug-list] tclug-list Digest, Vol 108, Issue 19 In-Reply-To: References: Message-ID: <52B0BABC.7090006@twp-llc.com> On 12/17/2013 12:00 PM, tclug-list-request at mn-linux.org wrote: > Date: Mon, 16 Dec 2013 17:37:11 -0600 > From: Andrew Dahl > > That all depends on your Internet connection and how much data you have > pushed. To restore the data, you'd put the utility on your system, login, > and use the restoration feature to pull your files down. Yes. Easy to restore data. NOT easy to restore my system including my OS, applications, and settings starting with a blank hard disk. Chris > On Mon, Dec 16, 2013 at 5:22 PM, Chris Schumann wrote: > Andrew wrote: >>> CrashPlan >> The thing I don't get about CrashPlan is this: What are you supposed to do >> if your hard disk dies? How long does it take to get me working again? >> >> Chris >> From mbmiller+l at gmail.com Tue Dec 17 17:37:49 2013 From: mbmiller+l at gmail.com (Mike Miller) Date: Tue, 17 Dec 2013 17:37:49 -0600 (CST) Subject: [tclug-list] assigning IP within script? Message-ID: I'm not sure this is possible. What I want to do is make the sort of assignment we see in /etc/hosts but have it work only within a script. Here's the problem. machineA is behind a firewall, but it is accessible via ssh from machineB, so I've written a script that lets me to ssh to machineA via ssh through machineB using port forwarding: ssh -f -L 25922:machineA:22 ${USERB}@machineB sleep 1 ; ssh -X -p 25922 ${USERA}@localhost It would be nice if I could use scp in a fairly straightforward way while connected that way. (Ignore usernames for the rest of this to keep it simple.) For example, I wish this would just work: scp machineA:file . That can't work, but can I write a script that would make it work? That script would read that command and execute this one: scp -P 25922 localhost:file . So it would be neat if I could write the script so that it could automatically convert "machineA" to localhost or 127.0.0.1. I have this line in /etc/hosts: 127.0.0.1 localhost Is there any way to basically make this assignment inside of the script (so that it doesn't change anything except within the script)?: 127.0.0.1 localhost machineA Then something like this might work in a script: scp -P 25922 $* The user could then just do something like this: scpA.bash machineA:file . Is there any hope of making that work? If not, maybe I can search for "machineA" and replace with localhost. Mike From jay at slushpupie.com Tue Dec 17 19:56:20 2013 From: jay at slushpupie.com (Jay Kline) Date: Tue, 17 Dec 2013 19:56:20 -0600 Subject: [tclug-list] assigning IP within script? In-Reply-To: References: Message-ID: On Tue, Dec 17, 2013 at 5:37 PM, Mike Miller wrote: > Here's the problem. machineA is behind a firewall, but it is accessible via > ssh from machineB, so I've written a script that lets me to ssh to machineA > via ssh through machineB using port forwarding: > A technique Ive use for this type of problem is to combine the ProxyCommand option with netcat (nc). So in your .ssh/config put this: Host machineA ProxyCommand ssh machineB /usr/bin/nc machineA 22 Using this method, all the native ssh-based tools "just work". To ease the time it takes to log on, I also make use of session multiplexing. That means it keeps the session open even after you exit, and future use of ssh/scp/etc will reuse the connection. Set that up with this entry: Host * ControlMaster auto ControlPath /tmp/.ssh_control_%r@%h:%p Note that using session multiplexing makes setting up port forwards a little more tricky, but not impossible (read the man page for ssh_config), and if your network link changes/drops you need to issue "ssh -O exit hostname" from your local host instead of using the ~. break command. Jay From mbmiller+l at gmail.com Wed Dec 18 02:37:29 2013 From: mbmiller+l at gmail.com (Mike Miller) Date: Wed, 18 Dec 2013 02:37:29 -0600 (CST) Subject: [tclug-list] assigning IP within script? In-Reply-To: References: Message-ID: Thanks for the great tips, Jay. Unfortunately, I'm trying to make a script that I can share with users who probably won't be interested in editing their config files. I might be able to use some of your ideas in some way in other work, though. Mike On Tue, 17 Dec 2013, Jay Kline wrote: > On Tue, Dec 17, 2013 at 5:37 PM, Mike Miller wrote: >> Here's the problem. machineA is behind a firewall, but it is >> accessible via ssh from machineB, so I've written a script that lets me >> to ssh to machineA via ssh through machineB using port forwarding: >> > > A technique Ive use for this type of problem is to combine the > ProxyCommand option with netcat (nc). So in your .ssh/config put this: > > Host machineA > ProxyCommand ssh machineB /usr/bin/nc machineA 22 > > Using this method, all the native ssh-based tools "just work". To ease > the time it takes to log on, I also make use of session multiplexing. > That means it keeps the session open even after you exit, and future use > of ssh/scp/etc will reuse the connection. Set that up with this entry: > > Host * > ControlMaster auto > ControlPath /tmp/.ssh_control_%r@%h:%p > > Note that using session multiplexing makes setting up port forwards a > little more tricky, but not impossible (read the man page for > ssh_config), and if your network link changes/drops you need to issue > "ssh -O exit hostname" from your local host instead of using the ~. > break command. > > Jay From jay at slushpupie.com Wed Dec 18 08:30:00 2013 From: jay at slushpupie.com (Jay Kline) Date: Wed, 18 Dec 2013 08:30:00 -0600 Subject: [tclug-list] assigning IP within script? In-Reply-To: References: Message-ID: On Wed, Dec 18, 2013 at 2:37 AM, Mike Miller wrote: > Thanks for the great tips, Jay. Unfortunately, I'm trying to make a script > that I can share with users who probably won't be interested in editing > their config files. I might be able to use some of your ideas in some way > in other work, though. > If you are scripting it and cant or dont want to edit the config file directly, remember that you can pass config file options on the commandline, with -o. So you could do something like this: scp -o 'ProxyCommand=ssh machineB /usr/bin/nc machineA 22' $* Jay From chewie at wookimus.net Wed Dec 18 08:33:40 2013 From: chewie at wookimus.net (Chad Walstrom) Date: Wed, 18 Dec 2013 08:33:40 -0600 Subject: [tclug-list] assigning IP within script? In-Reply-To: References: Message-ID: <11988.1387377220@ydalir.wookimus> Jay Kline wrote: > A technique Ive use for this type of problem is to combine the > ProxyCommand option with netcat (nc). So in your .ssh/config put this: > > Host machineA > ProxyCommand ssh machineB /usr/bin/nc machineA 22 Wow... That's just pretty; a winner for simple A->B single-host mappings. If you want a more full-blown VPN over SSH: https://wiki.archlinux.org/index.php/VPN_over_SSH I've used SSH+PPP in the past with good success way back in the day, but it does require PPPD to be installed on the remote side. The built-in layer-3 OpenSSH tunneling using the TUN device looks quite useful. > ... > Host * > ControlMaster auto > ControlPath /tmp/.ssh_control_%r@%h:%p > > ... > ssh_config), and if your network link changes/drops you need to issue > "ssh -O exit hostname" from your local host instead of using the ~. > break command. Yeah, I've been using ControlMaster for a couple of months now after a tip from a co-worker. It's epecially helpful when using TRAMP in Emacs to view remote files or initiate remote shells. Chad -- Chad Walstrom http://runswithd6s.blogspot.net/ From jay at slushpupie.com Wed Dec 18 09:27:00 2013 From: jay at slushpupie.com (Jay Kline) Date: Wed, 18 Dec 2013 09:27:00 -0600 Subject: [tclug-list] assigning IP within script? In-Reply-To: <11988.1387377220@ydalir.wookimus> References: <11988.1387377220@ydalir.wookimus> Message-ID: On Wed, Dec 18, 2013 at 8:33 AM, Chad Walstrom wrote: > Jay Kline wrote: >> A technique Ive use for this type of problem is to combine the >> ProxyCommand option with netcat (nc). So in your .ssh/config put this: >> >> Host machineA >> ProxyCommand ssh machineB /usr/bin/nc machineA 22 > > Wow... That's just pretty; a winner for simple A->B single-host > mappings. You can chain them together too, if you need to go A->B->C. Though at that point you've tunneled ssh inside itself twice, so the actual packet payload size starts getting smaller, and it can be painful on slower networks. > If you want a more full-blown VPN over SSH: > https://wiki.archlinux.org/index.php/VPN_over_SSH > > I've used SSH+PPP in the past with good success way back in the day, but > it does require PPPD to be installed on the remote side. The built-in > layer-3 OpenSSH tunneling using the TUN device looks quite useful. Ive done this in the past too, but Ive found it to be somewhat fragile for more than a couple of users (though "users" in this case typically ends up being systems, not people). It also requires more privileged access, whereas the ProxyCommand I can implement without any special permissions, even when sshd disallows port forwarding :-) >> ... >> Host * >> ControlMaster auto >> ControlPath /tmp/.ssh_control_%r@%h:%p >> >> ... >> ssh_config), and if your network link changes/drops you need to issue >> "ssh -O exit hostname" from your local host instead of using the ~. >> break command. > > Yeah, I've been using ControlMaster for a couple of months now after a > tip from a co-worker. It's epecially helpful when using TRAMP in Emacs > to view remote files or initiate remote shells. > I started using it when I had to do svn over a link that had a screwed up DNS configuration, so the initial connection took forever, but link speed after that was fine. 90% of the time I love it. But its sometimes a pain, since its harder to interact directly with the real ssh process making the connection. Jay From goeko at Goecke-Dolan.com Sat Dec 21 13:02:10 2013 From: goeko at Goecke-Dolan.com (Brian Dolan-Goecke) Date: Sat, 21 Dec 2013 13:02:10 -0600 Subject: [tclug-list] Possible meeting topic/presentation In-Reply-To: References: Message-ID: <52B5E5B2.3020105@Goecke-Dolan.com> Erik, Sorry for repling so long after you sent your message. I would love to hear what you have to say about AWS and/or Ansible. If you would have time and be willing it would be great if you could come talk to the Penguins Unbound Meeting.. Looking at the schedule http://www.penguinsunbound.com/Meetings it looks like the first meeting that is open would be April. If that doesn't work for you take a look at the calendar and let us (me or John Frisk) know what would work. I know the dates are a bit far off, but it would be great to get you on the calendar. Thanks. Again, thanks for your willingness to talk. ==>brian. On 11/27/2013 10:58 AM, Erik Anderson wrote: > I've been doing a *lot* of work with both AWS Cloudformation[1] and > Ansible[2] lately, and have found both to be delightful to work with. > > I'd be willing to put together a meeting presentation on either or both > of these at some point early next year if there's interest. > > Cloudformation may not be as applicable, as I'm not sure how many folks > on the list are actually working in AWS, but Ansible (or at least *some* > configuration management system) is something that IMHO every *nix user > ought to be using, even if it's for a single system. > > So, would anyone be interested in hearing about either of these > technologies? > > Derp, > Erik > > [1] http://aws.amazon.com/cloudformation/ > [2] https://github.com/ansible/ansible > > > _______________________________________________ > TCLUG Mailing List - Minneapolis/St. Paul, Minnesota > tclug-list at mn-linux.org > http://mailman.mn-linux.org/mailman/listinfo/tclug-list > From ryanjcole at me.com Mon Dec 30 21:15:10 2013 From: ryanjcole at me.com (Ryan Coleman) Date: Mon, 30 Dec 2013 21:15:10 -0600 Subject: [tclug-list] Monitoring a bunch of remote UPS units Message-ID: <475D507B-E5A2-4D83-93EC-2B2FA2C263A1@me.com> I have a need to monitor a lot of UPS units that are installed around the country at customer?s retail sites and I?m trying to figure out if NUT will do the job for me. The firewall I?m using won?t allow for a clean install of apcupsd (and I will admit I?d like some control over the systems, if possible). I?m trying to get NUT going but I am not sure if I am barking up the wrong tree. Ideally I?d have a ACL-protected port to call on the UPS remotely but some sites won?t allow for that so I?d need to have the ability to have it call my protected server as well. Any ideas? The firewall is pfsense (BSD, I know? but it?s serving us quite well right now at 30 locations) but the server can be anything I need it to be. Thanks! ? Ryan