| From: | Jayadevan M <maymala(dot)jayadevan(at)gmail(dot)com> |
|---|---|
| To: | Binand Sethumadhavan <binand(at)gmx(dot)net> |
| Cc: | "pgsql-novice(at)postgresql(dot)org" <pgsql-novice(at)postgresql(dot)org> |
| Subject: | Re: Autovacuum Woes |
| Date: | 2016-08-09 06:29:05 |
| Message-ID: | CAFS1N4gbbwZuwXVVVe8aP9xh7oNgB2jGaHC6pc3=-gtGez931w@mail.gmail.com |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-novice |
>
>
>
> This worked fine for many months, but of late a new problem has
> started. At the time of disabling and enabling, we are seeing large
> performance degradation. Several hundred connections like this:
>
> 19090 ? Ss 0:00 postgres: user dbname 10.13.36.19(42782)
> PARSE waiting
> 19091 ? Ss 0:00 postgres: user dbname 10.13.36.19(42783)
> PARSE waiting
> 19092 ? Ss 0:00 postgres: user dbname 10.13.36.19(42784)
> PARSE waiting
> 19093 ? Ss 0:00 postgres: user dbname 10.13.36.19(42785)
> PARSE waiting
> 19095 ? Ss 0:00 postgres: user dbname 10.13.36.19(42786)
> PARSE waiting
>
> So obviously, disabling/enabling autovacuum has side-effects.
>
>
> Use the queries in this link to find out what is blocking these
transactions.
https://wiki.postgresql.org/wiki/Lock_Monitoring
<https://wiki.postgresql.org/wiki/Lock_Monitoring>
It is highly likely there are autovacuum processes to take care of
transactionid wraparound. (See point no 4 under Vacuuming basics)
https://www.postgresql.org/docs/current/static/routine-vacuuming.html
Regards,
Jayadevan
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Kevin Grittner | 2016-08-09 14:13:05 | Re: Autovacuum Woes |
| Previous Message | Binand Sethumadhavan | 2016-08-09 06:02:11 | Autovacuum Woes |