It has been running at least since 2018.
I got that from the tail
=> /sys/kernel/mm/ksm/run <==
0
==> /sys/kernel/mm/transparent_hugepage/defrag <==
[always] madvise never
==> /sys/kernel/mm/transparent_hugepage/enabled <==
always madvise [never]
==> /sys/kernel/mm/transparent_hugepage/khugepa
Thanks, I will check it out.
On Thu, Jul 21, 2022 at 4:21 PM Justin Pryzby wrote:
> On Thu, Jul 21, 2022 at 03:59:30PM -0400, bruno da silva wrote:
> > OS/version: CentOS release 6.9 (Final)
>
> How are these set ?
>
> tail /sys/kernel/mm/ksm/run
> /sys/kernel/mm/transparent_hugepage/{defrag,e
On Thu, Jul 21, 2022 at 03:59:30PM -0400, bruno da silva wrote:
> OS/version: CentOS release 6.9 (Final)
How are these set ?
tail /sys/kernel/mm/ksm/run
/sys/kernel/mm/transparent_hugepage/{defrag,enabled,khugepaged/defrag}
/proc/sys/vm/zone_reclaim_mode
I suspect you may be suffering from i
On 2022-07-21 Th 14:37, bruno da silva wrote:
> Hello.
>
> I'm investigating an issue on a PostgresSql 9.5.21 installation that
> becomes unusable in an intermittent way. Simple queries like "select
> now();" could take 20s. commits take 2s. and all gets fixed after an
> engine restart.
>
> I loo
The issue started a month ago.
On Thu, Jul 21, 2022 at 3:59 PM bruno da silva wrote:
> Thanks for the quick response.
>
> OS/version: CentOS release 6.9 (Final)
>
> Hardware(non dedicated to the db, other services and app run the same
> server):
>
> Xeon(R) CPU E5-2690 v4 @ 2.60GHz - 56
Thanks for the quick response.
OS/version: CentOS release 6.9 (Final)
Hardware(non dedicated to the db, other services and app run the same
server):
Xeon(R) CPU E5-2690 v4 @ 2.60GHz - 56 cores - 504 GB RAM
logicaldrive 1 (1.5 TB, RAID 1, OK)
physicaldrive 1I:3:1 (port 1I:box 3:bay 1, S
On Thu, Jul 21, 2022 at 02:37:35PM -0400, bruno da silva wrote:
> I'm investigating an issue on a PostgresSql 9.5.21 installation that
> becomes unusable in an intermittent way. Simple queries like "select
> now();" could take 20s. commits take 2s. and all gets fixed after an engine
> restart.
>
>
Hello.
I'm investigating an issue on a PostgresSql 9.5.21 installation that
becomes unusable in an intermittent way. Simple queries like "select
now();" could take 20s. commits take 2s. and all gets fixed after an engine
restart.
I look into the pg logs and no signs of errors. and checkpoints are
