On Thu, 2021-04-29 at 11:11 -0500, Don Seiler wrote: > On Thu, Apr 29, 2021 at 1:38 AM Laurenz Albe <laurenz.albe@xxxxxxxxxxx> wrote: > > My gues is that you have too many active client connections, and you are suffering > > from contention between the many backends that all want to write WAL. > > > > In that case, use a connection pool to limit the number of active connections. > > We do have pgbouncer in place already. What is the maximum pool size? How many connections in state "active" do you have? > What I really want to know in this case is if there is some other PG operation that > accounts for a WALWriteLock wait, or is it always an I/O (write) to the WAL file > storage, and we can focus our investigation there? WALWriteLock: must be held to write WAL buffers to disk (XLogWrite or XLogFlush). It might be I/O overload, but it could also be internal contention for the lock. Check the "%iowait" column in "sar -u 1". Yours, Laurenz Albe -- Cybertec | https://www.cybertec-postgresql.com
