[Pacemaker] Segmentation fault stopping Heartbeat

Andrew Beekhof beekhof at gmail.com
Thu Oct 30 12:20:39 UTC 2008


On Oct 30, 2008, at 12:50 PM, Adrian Chapela wrote:

> Andrew Beekhof escribió:

>> Its not a file.  It will be the output from:
>>  gdb /usr/lib/heartbeat/crmd /var/run/heartbeat/cores/hacluster/ 
>> core.12345
>>  (now type 'where')
>>
>> hb_report normally does this automatically
> JEJE is not a problem o hb_report, core file doesn't exist, could be  
> this the reason ?
>
> logd[7417]: 2008/10/30_11:50:02 WARN: Core dumps could be lost if  
> multiple dumps occur.
> logd[7417]: 2008/10/30_11:50:02 WARN: Consider setting non-default  
> value in /proc/sys/kernel/core_pattern (or equivalent) for maximum  
> supportability
> logd[7417]: 2008/10/30_11:50:02 WARN: Consider setting /proc/sys/ 
> kernel/core_uses_pid (or equivalent) to 1 for maximum supportability

definitely.
perhaps try:
   gdb /usr/lib/heartbeat/crmd /var/run/heartbeat/cores/hacluster/core

but the advice in those logs is worth following for next time





More information about the Pacemaker mailing list