Tags
Its a known issue, and its applicable to Solaris environment. I came across this issue recently, this one I handled multiple times. The actual cause is the process keeps on spinning and eventually times out. The LMON process issues the kill instance due to time outs.
Solaris: Process spins/ASM and DB Crash if RAC Instance Is Up For > 248 Days by LMHB with ORA-29770 (Doc ID 2159643.1) — It’s available in metalink.
The affected databases need a patch for bug 18740837 and bug 22901797.
In particular, the killed process has the below stack trace:
<- __sighndlr() <- call_user_handler() <- __pollsys() <- _pollsys()
The call pollsys() passed the handler and times out eventually, in this case the process didn’t finish the ack due to spinning. It’s a known issue on solaris platform, as per the documentation it’s issue with C compiler.
It may also affect non-RAC asm instances. But very likely on RAC environments.