Not sure if this is really the right board for this, but here goes. In a
performance critical, highly available application, when a misbehaving process
core dumps the creation of the corefile (gigabytes in size) puts a lot of
pressure on the abililty to restart and recover in a timely fashion.
From the outside looking in, by using top(1), it appears that one aspect is
that one cpu is being hogged at system scheduling priority of 60 during the
entire course of creating the core file.
So the question would be, is there some means to reduce this scheduling
priority such that the dumping operation could take place rather lazily, only
using otherwise-idle cycles? Or is that even really the problem?
I've tried to dig into corectl, etc but cant find the actual dumping code
anywhere so thats part of the reason for my ignorance.
Thanks
-d
This message posted from opensolaris.org
_______________________________________________
perf-discuss mailing list
perf-discuss@opensolaris.org
