HP 8000M crash - ran out of memory

[Neil_77]

Hello,

a few of the HP 8000M switches in our network, running C.09.16 reset themselves recently. The error given was:

"Software exception at svc_misc.c:378 -- in 'mftTask', task ID = 0x88dee0
-> malloc2() ran out of memory"

Can anyone shed any light on what this means and how to prevent it from happening in future?

A number of the switches in the network were showing "high collision or drop rate" alarms at the time if the resets so I'm guessing that there was maybe abnormally high network traffic around this time.

However, I haven't been able to identify a cause of the high traffic volume. Can anyone suggest anything to help troubleshoot this to prevent it happening again in future?

Thanks.

[OLARU Dan]

Update to C.09.22

ftp://ftp.hp.com/pub/networking/software/j4110922.zip

[Neil_77]

Thanks. Purely by coincidence, we are actually planning on upgrading to C.09.22 over the next few days in any case.

But I just wondered if anyone could shed any light on this problem. Or is this a known issue with C.09.16 that the new OS resolves?

[Ron Kinner]

A malloc error is usually a software bug. If you look at the release notes for C.09.22 you will see that C.09.19 corrected at least one malloc error (There was another one back in C.9.10). However, the error you have is not exactly the same as the one they have so you should probably report it to HP just in case.

http://www.hp.com/rnd/software/j4110922.htm

Ron

[Neil_77]

Hi Ron,

thanks for the advice - I'm new to HP switches (my previous experience has been almost all cisco) so its useful for me to get an idea of some of the issues and processes etc.

Neil