CUCM not reachable

[hjacquemin]

Hi Guys,

We actually have a problem with one of our call manager. Sometimes (2nd time in 2 weeks) the server is unreachable on the network. When we connect in the console we have the following message:

"Memory not reachable"

It's probably a problem on the memory, only a reboot is enough. There is some diagnosys command to identify the root cause?

Thanks,

Hervé Jacquemin

[Jaime Valencia]

Version and platform?

[hjacquemin]

it's CUCM 7.1 on platform MCS7825I4-K9-CMC2

[btmulgrew]

Hi Herve - can you provide the full system version as there are quite a number of memory leak bugs releated to 7.1.x

Also - can you advise roughly how large your cluster is (number of devices / nodes)

thanks

Brian

[hjacquemin]

the version is:

7.1.5.10000-12

and only 1 pub and 1 sub in the cluster

and on the logs messages I have message like this

Sep 28 10:28:26 pdisguicm01 kern 6 kernel: Memory: 2071644k/2095488k available (1925k kernel code, 22824k reserved, 787k data, 192k init, 1177984k highmem)

and some raid problem (date Sep 27 10 am) see attachment

[btmulgrew]

Thanks - the most common memory bug is tomcat related, and your system version doesnt include the patch as below:

https://www.cisco.com/cisco/psn/bssprt/bss?searchType=bugIdfilterData&bugId=CSCtj50884&filterData=EMPTY&QueryString=7.1.5%20memory%20leak%20cucm%3Fst=kw&highLight=true&searchCriteria='7.1.5%2520memory%2520leak%2520cucm'&isadvancedpage=false&SavedSear...

Sorry - i dont have access to view your attachment, but it also may be worth taking the server down and doing a full hardware diags on it.