The backup used was from well after the failover time which is why I was concerned. Interestingly enough the logs are still all prefixed with 00000004... That just makes this problem extremely bizarre.

Current log listing:
[postgres@db01pri pgwalarchives]# ls -ltr
total 82028
-rw------- 1 postgres postgres 250 Jun 15 02:05 0000000400000745000000DB.00F290D8.backup
-rw------- 1 postgres postgres 250 Jun 15 13:28 000000040000074700000065.00D48908.backup
-rw------- 1 postgres postgres 16777216 Jun 15 15:58 0000000400000747000000E4
-rw------- 1 postgres postgres 16777216 Jun 15 15:59 0000000400000747000000E5
-rw------- 1 postgres postgres 16777216 Jun 15 16:02 0000000400000747000000E6
-rw------- 1 postgres postgres 16777216 Jun 15 16:04 0000000400000747000000E7
-rw------- 1 postgres postgres 16777216 Jun 15 16:07 0000000400000747000000E8
[postgres@db01pri pgwalarchives]#

Keith

Tom Lane wrote:

Keith Pierno <kpierno@lulu.com> writes:

I was able to find the file, which was from the date of the failure
(Tuesday June 9th). My question is why the backup from this morning
would all of a sudden require a file from the 9th even though 5 hours
of logs were able to be applied?


It was apparently busy applying the logs that were from the old primary.
When it got up to the point where the failover occurred, kaboom :-(.
It would have had trouble in any case there because the subsequent log
files are presumably numbered 00000006... instead of 00000004...
Were you getting those shipped over properly from the new primary?

			regards, tom lane