Обсуждение: WAL file question
listing of wal file time stamps for one of our production databases: Nov 17 06:22 000000010000006100000013 Nov 17 06:42 000000010000006100000014 Nov 17 07:02 000000010000006100000015 Nov 17 07:22 000000010000006100000016 Nov 17 07:42 000000010000006100000017 Nov 17 08:02 000000010000006100000018 Nov 17 08:22 000000010000006100000019 Nov 17 08:34 000000010000006100000012 I would expect that these things are sequential, yet the file that I would think would be the oldest (000000010000006100000012) has the latest time stamp. What am I missing? -- Until later, Geoffrey Those who would give up essential Liberty, to purchase a little temporary Safety, deserve neither Liberty nor Safety. - Benjamin Franklin
Geoffrey <lists@serioustechnology.com> writes:
> listing of wal file time stamps for one of our production databases:
> Nov 17 06:22 000000010000006100000013
> Nov 17 06:42 000000010000006100000014
> Nov 17 07:02 000000010000006100000015
> Nov 17 07:22 000000010000006100000016
> Nov 17 07:42 000000010000006100000017
> Nov 17 08:02 000000010000006100000018
> Nov 17 08:22 000000010000006100000019
> Nov 17 08:34 000000010000006100000012
> I would expect that these things are sequential, yet the file that I
> would think would be the oldest (000000010000006100000012) has the
> latest time stamp.
> What am I missing?
Most of those are probably old files that have been renamed into place
for future use --- and the renamer doesn't really worry about reusing
old files in order. I suspect ...12 is the only one that's live.
A look at pg_controldata output would help you check that.
regards, tom lane
Tom Lane wrote: > Geoffrey <lists@serioustechnology.com> writes: >> listing of wal file time stamps for one of our production databases: >> Nov 17 06:22 000000010000006100000013 >> Nov 17 06:42 000000010000006100000014 >> Nov 17 07:02 000000010000006100000015 >> Nov 17 07:22 000000010000006100000016 >> Nov 17 07:42 000000010000006100000017 >> Nov 17 08:02 000000010000006100000018 >> Nov 17 08:22 000000010000006100000019 >> Nov 17 08:34 000000010000006100000012 > >> I would expect that these things are sequential, yet the file that I >> would think would be the oldest (000000010000006100000012) has the >> latest time stamp. > >> What am I missing? > > Most of those are probably old files that have been renamed into place > for future use --- and the renamer doesn't really worry about reusing > old files in order. I suspect ...12 is the only one that's live. > A look at pg_controldata output would help you check that. > > regards, tom lane Thanks Tom. -- Until later, Geoffrey Those who would give up essential Liberty, to purchase a little temporary Safety, deserve neither Liberty nor Safety. - Benjamin Franklin