Herold ,
A process tried to get a buffer with a size of 54361
bytes , which is bigger than every mtu in your router
.
Makes little sence does it ?
Usually indicate a memory corruption of some sort .
The trace-back messages is good info for TAC to trace
what process did the attempt .
If it happens again I would open a TAC case .
While in rommon> get a stack dump , it is very usefull
info to attach to the problem ticket .
( if they not ask to upgrade to a more recent
IOS first :) )
flem
--- Herold Heiko <[EMAIL PROTECTED]> wrote:
> Today after lunch I found my primary 3640 dead (at a
> momento of minimum
> load I must say), connecting to the console <enter>
> showed the rommonitor
> prompt. Since there was some urgent need and I had
> some users on my back
> already a quick powercycle "solved" the thing,
> however I'd like to find
> the reason. Logs didn't show much:
>
> May 22 13:33:13 ciscor1 87115: May 22 13:33:12.260
> met-dst:
> %SYS-2-GETBUF: Bad getbuffer, bytes= 54361
> May 22 13:33:13 ciscor1 87116: -Process= "<interrupt
> level>", ipl= 1
> May 22 13:33:13 ciscor1 87117: -Traceback= 6018A1E4
> 60094498 600948D0
> 60094D40 600244E0
>
> May 22 13:33:59 ciscor2 36580: May 22 13:33:58.746
> met-dst:
> %DUAL-5-NBRCHANGE: IP-EIGRP 213: Neighbor x.y.z.11
> (Ethernet0/0) is down:
> holding time expired
>
>
> Search on CCO found unrelated items or "getbuf might
> appear" in 3640
> release notes without further info.
> Any idea where to look ? Should I query TAC ?
> Oh BTW this is a 3640 with 48MB ram, usually 25MB
> free,
> C3640-IS-MZ_112-12_P.BIN, uptime was 140 days (since
> 2 Jan, after all the
> y2k tests I added a couple of WIC-1T since
> everything else was down
> anyway <grin>). Lots of smallish connections, all of
> which low bandwidth
> (x25 on a 128kb serial, a 64k framerelay, some isdn
> dialers) except one
> 2MB leased line going around 60% usage and a bridge
> to token ring usually
> at load 2/255. A bit of NAT (one on a int eth
> connected directly to
> another cisco with a 64kb CIR link only)
> Overall I'd say it's doing several things but
> shouldn't really sweat,
> also CPU is about 15% with some (few) spikes at max
> 30%.
>
> Any ideas ?
>
> Heiko
>
> -- PREVINET S.p.A.
> [EMAIL PROTECTED]
> -- Via Marocchesa, 14 ph x39-041-5494228
> -- I-31021 Mogliano V.to (TV) fax x39-041-5492263
> -- ITALY
>
>
> ___________________________________
> UPDATED Posting Guidelines:
> http://www.groupstudy.com/list/guide.html
> FAQ, list archives, and subscription info:
> http://www.groupstudy.com
> Report misconduct and Nondisclosure violations to
[EMAIL PROTECTED]
__________________________________________________
Do You Yahoo!?
Send instant messages & get email alerts with Yahoo! Messenger.
http://im.yahoo.com/
___________________________________
UPDATED Posting Guidelines: http://www.groupstudy.com/list/guide.html
FAQ, list archives, and subscription info: http://www.groupstudy.com
Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
