On 8/4/15 2:47 AM, Jan Keirse wrote:
CAST (age(relfrozenxid) AS real) / CAST(trunc(((2^32)/2)-1-1000000) AS real)
>AS perc_until_wraparound_server_freeze
>
>
>(Note that we do this at the table level rather than the database level like
>you did, though, so that we have the information we need to tune the
>settings for individual tables.)
Thanks for the correction regarding the 1.000.000 safety margin! I
chose to monitor only the total value value to limit the amount of
extra metrics in the monitoring database. In case the value increased
we'll execute the queries to find out what table(s) is/are causing the
problem interactively.
Something to be aware of; with certain workloads you can actually run
out of MXIDs faster than XIDs, and I don't think there's any easy way to
monitor MXID consumption.
BTW, you might find my talk from pgCon or the related video useful:
http://www.pgcon.org/2015/schedule/events/829.en.html
https://www.youtube.com/watch?v=b1fcvkl0ffQ
--
Jim Nasby, Data Architect, Blue Treble Consulting, Austin TX
Data in Trouble? Get it in Treble! http://BlueTreble.com
--
Sent via pgsql-general mailing list (pgsql-general@xxxxxxxxxxxxxx)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-general
