Re: (ASCEND) csmx woes (again)

["Jason Nealis" <nealis@rcn.com>]

The reason you are seeing these messages is because for some reason or
another the firmware on that individual DSP has become corrupted. After
each disconnect the integrity of the firmware on each DSP is
CRC checked and if that check if failed the DSP is bounced in order to
force it to re-load it's firmware.


Jason Nealis
Director of Internet Operations
RCN
----- Original Message -----
From: Tom Gwilt <tgwilt@suite224.net>
To: <ascend-users@max.bungi.com>
Sent: Thursday, April 08, 1999 2:04 PM
Subject: Re: (ASCEND) csmx woes (again)


> Kevin Kim wrote:
> >
> > How often does this happen ?
> >
> > Thanks,
> > Kevin
>
> cambot: {20} % grep "reboot requested" ctnt.log | awk -F" " '{print
> $1,$2,$3,$8,$9,$10,$11}'
> Apr 7 18:08:35 chips 0x14 reboot requested
> Apr 7 18:51:27 chips 0x100 reboot requested
> Apr 7 19:56:02 chips 0x1000 reboot requested
> Apr 7 20:14:10 chips 0x1 reboot requested
> Apr 8 06:15:23 chips 0x1 reboot requested
> Apr 8 11:50:13 chips 0x2000 reboot requested
> Apr 8 12:56:51 chips 0x8000 reboot requested
>
> Output of the logfile indicates varying times, possibly dependent on
> load. Notice that the reboots stopped at 8:14 pm on April 7, resumed at
> 6:15 am on April 8. Originally I thought I had seen a pattern of about
> an hour, but that doesn't seem to be the case.
>
> I received the following message from Paul Rolland:
>
> ===================================================
> Hello,
>
> Do not worry about this one, this is code to fix some defect that
> would cause some resources to hang...
>
> It should be fixed in May by Rockwell...
>
> This is all I got from Ascend :
>
> here is an extract of 7.0.4 release notes that explains that those
> messages
>  are normal warnings.
>
>
>
> N/A Graceful reboot of Series56-II digital modems under abnormal
> conditions.
> ==================================================================
>
> Procedures were implemented to make all modem channels within a physical
> device inactive
> before rebooting when abnormal conditions are detected. The graceful
> reboot
>  procedure waits
> up to 60 minutes for these conditions to occur, and then resets the
> modem
> anyway.
> As part of the graceful reboot procedure, the classification of the
> following log messages was
> changed from Error to Warning:
> modemdrv.c: xxxx - modem x bad last x calls
> csmxdrv.c: xxxx - outbound mbox message queue full, chip x
> csmxdrv.c: xxxx - chips xxx reboot requested
> modemdrv.c: xxxx - channel xxx failed to come out of reset
> The following message was changed from Info to Warning:
> csmxdrv.c: xxxx - chips xxx reboot done
> And, the following Warning message was added, to indicate that the
> specified channel is
> inactive and ready for restart:
> Modem channel # xxx is ready for restart
> In addition, the following new Disconnect codes are now supported:
> Disconnect code Meaning
> 15 Modem disconnected because of a missing OK response from a
> CSM/3 modem channel.
> 16 Modem disconnected because of a stuck/full mailbox message
> queue of a CSM/3 modem chip.
> 17 Modem disconnected because of a dead CSM/3 modem channel.
> 18 Timeout for graceful reboot forced a modem channel to
> disconnect.
>
> Regards,
> Paul
> ++ Ascend Users Mailing List ++
> To unsubscribe: send unsubscribe to ascend-users-request@bungi.com
> To get FAQ'd: <http://www.nealis.net/ascend/faq>
>

++ Ascend Users Mailing List ++
To unsubscribe:	send unsubscribe to ascend-users-request@bungi.com
To get FAQ'd:	<http://www.nealis.net/ascend/faq>