Ascend Archive
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
(ASCEND) Re: Pipeline 130 slowness, packet loss under load
Adding to my previous post here:
>We have a client with a Pipeline 130, 5.1A, connected via a 128k
>point-to-point T1 to a TNT. They do sporadic large transfers, and their
>connection typically goes from idle to nailed, with throughput exceeding
>150K/sec.
>
>During heavy transfers, the Pipeline will sporadically fail to respond to
>queries from our SNMP monitor, and ping times will jump from the usual
>20-30 ms to 800-1900 ms with up to 4% packet loss, as measured from the TNT.
>
>Is this normal/typical for a Pipeline 130?
I have heard from a user with two Pipeline 130s experiencing the same
problem that lowering the MRU from 1524 to 1500 fixed the slow pings times,
but not their packet loss, which in their case is extreme.
Ascend's response was that this is not normal for a 130, and to check the
line from the TNT's point of view.
I cleared the t1-stats, and in twenty minutes of sustained full throughput
from the 130 got:
CRC Errors: 0
Frame Slips: 26
Framing Bit Errors: 0
Out of Frame Events: 0
Line Code Violations: 0
So I appear to be getting frame slips every 45 seconds or so. I have no
idea how to troubleshoot this, or even if it's excessive, so pointers are
most welcome.
The other thing is that the other DLCI on the Frame Relay (not
point-to-point, as I misstate above) T1 goes to a 4000, which takes the
remainder (almost all) of the bandwidth. I don't _appear_ to be having
troubles with that link, although it's never even close to pegged.
Into the guts of Frame...
Peter Lalor
Infoasis
plalor@infoasis.com
http://www.infoasis.com/
415-459-7991 x102
415-459-7992 fax
++ Ascend Users Mailing List ++
To unsubscribe: send unsubscribe to ascend-users-request@bungi.com
To get FAQ'd: <http://www.nealis.net/ascend/faq>