wambacher wrote
> hi,
>
> waiting for the index (104/121GB), i read his document
> http://www.postgresql.org/docs/9.4/static/kernel-resources.html#LINUX-MEMORY-OVERCOMMIT
> and will do some changes before the next analyze:
>
> some comments:
>
> - the OOM did not kill the Postmaster but the Analyze-Job.
> - started with 24GB real and 18GB Swap - that must be enought! --> killed
>
> - will reduce shared_mem from 1GB to 512 MB
> - will reduce max_connections to 100 (but no user was actice, the
> applications are down)
> - will do "sysctl -w vm.overcommit_memory=2" (just did it)
> - may do "echo -1000 > /proc/self/oom_score_adj" but only if the other
> actions fail
> - the last steps for older kernels are not relevant, i'm running ubuntu
> 14.04 LTS, which is 3.13
>
> i'll keep you informed.
>
> regards
> walter
very strange, after clustering, recreating the gis-index and changing
nothing else no poblems at all.
may be the index was corrupt? or there was so much reorganization needed
that analyze run in trouble?
i'll do most of the changes now, reboot and bring my system live again.
Let's see whats happening the next days.
Thanks to all
walter
--
View this message in context:
http://postgresql.nabble.com/autovacuum-worker-running-amok-and-me-too-tp5840299p5840786.html
Sent from the PostgreSQL - general mailing list archive at Nabble.com.