BTW: is there a framework to test recovery related features?
The only idea I could take from the top of my head is to comment out all the
page writes and leave only WAL logging. Then crash database at random and
verify if the index still performs as expected.
Regards,
Vladimir Sitnikov
the operation fetched 116
pages into shared buffers and 108 of those 116 were buffer hits.
Sincerely yours,
Vladimir Sitnikov
buffer_stats_in_explain_analyze.patch
Description: Binary data
--
Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)
To make changes to your subscription:
http
of that statistics could be server health monitoring:
provided there is an interface for ongoing integral statistics, one could
create a job that takes snapshots, computes the difference and plots it on a
graph.
Sincerely,
Vladimir Sitnikov
)+('dense_rank')::int4)::text'dense_rank')::int4 as position,
runnerid, time
from big_marathon
order by time
) results
where position=*2*
Best regards,
Vladimir Sitnikov
order by date
) as t
order by used asc nulls last limit *1*
Best regards,
Vladimir Sitnikov
minutes. It does not look too
bad to get the statistics of last 15 minutes lost in case of database crash
either.
Regards,
Vladimir Sitnikov
to store the SID
into some file in working directory, then read it and connect to the shared
memory.
If that makes sense, I could implement it that way and test it under Vista.
Sincerely yours,
Vladimir Sitnikov
the collected data.
I think think this change is going to split page access statistics by each
and every execution node (much like 'actual time' is split now).
Will this patch ever be accepted?
Regards,
Vladimir Sitnikov
or
log_statement_stats is enabled). I am not sure if it could interfere with
instrumentation.
Regards,
Vladimir Sitnikov
101 - 109 of 109 matches
Mail list logo
