[c-nsp] Cisco 3640 Memory leak problem

[robbie]

Also, I have used 'show mem dead' in the past to identify which 
processes (in my case, it was SNMP & TFTP on a 3640) are allocating 
memory and not doing a very good job of cleaning up after themselves. 
This list may give you some idea of what process(es) to start looking at.

cheers,
robbie

Oliver Boehmer (oboehmer) wrote:
> Jordan,
> 
> 
>>I have a memory problem on Cisco 3640. Recently we configured mrtg
>>graphs for used/free memory and realised that the Processor Used
>>memory is constantly increasing. In about a month it causes the
>>router to crash and after restart it takes 4-5 MB less memory.
>>
>>The router is running 12.2(15)T9 IOS. It has 2 (out of 4) E1 PRI
>>modules enabled, and H.323 gatekeeper running.
>>
>>Any suggestions would be appreciated.
> 
> 
> periodically (like once a day, depending how fast the leak is) poll
> "show proc mem" and check out which process' memory consumption
> increases over time ("show proc mem sorted" sorts the output by
> allocated mem).. If you have identified the process which leaks, gather
> "show proc mem <pid>" periodically (where <pid> is the process id shown
> in the first column of show proc mem) and give this info to TAC for
> further analysis.. Alternative ways of troubleshooting this can be found
> at http://www.cisco.com/warp/public/63/mallocfail.shtml
> 
> 	oli 
> 
> _______________________________________________
> cisco-nsp mailing list  cisco-nsp at puck.nether.net
> https://puck.nether.net/mailman/listinfo/cisco-nsp
> archive at http://puck.nether.net/pipermail/cisco-nsp/
> 
>