Thread: Clarification on CONTEXT: xlog redo xid assignment
Looking for some clarification on some log output. We are moving our
databases to
new physical machines. To do this we are taking zfs snapshots from the
slaves
copying the snapshots to the new machines and then restarting the new
machines
with a failover file.
The database is able to start and I can access it, run commands against it ,
create pg_dumps
and reload those without any errors. However I am getting the following
errors in the logs
at startup.
Dec 7 12:40:53 postgres[28199]: [ID 748848 local2.info] [1-1] LOG:
database system was interrupted while in recovery at log time 2011-12-07
00:32:49 PST
Dec 7 12:40:53 postgres[28199]: [ID 748848 local2.info] [1-2] HINT: If
this has occurred more than once some data might be corrupted and you might
need to choose an earlier recovery target.
Dec 7 12:40:53 postgres[28199]: [ID 748848 local2.info] [2-1] LOG:
entering standby mode
Dec 7 12:40:53 postgres[28199]: [ID 748848 local2.info] [3-1] LOG:
restored log file "000000010000060600000057" from archive
Dec 7 12:40:53 postgres[28199]: [ID 748848 local2.info] [4-1] LOG: redo
starts at 606/57536038
Dec 7 12:40:55 postgres[28199]: [ID 748848 local2.info] [5-1] LOG: file
"pg_subtrans/8732" doesn't exist, reading as zeroes
Dec 7 12:40:55 postgres[28199]: [ID 748848 local2.info] [5-2] CONTEXT:
xlog redo xid assignment xtop 2268215780: subxacts: 2268215781 2268215782
2268215791 2268215792 2268215793 2268215794 2268215799 2268215800 2268215818
2268215819 2268215837 2268215838 2268215841 2268215842 2268215845 2268215846
2268215849 2268215851 2268215857 2268215858 2268215863 2268215864 2268215865
2268215866 2268215867 2268215869 2268215871 2268215872 2268215875 2268215876
2268215878 2268215879 2268215880 2268215881 2268215882 2268215883 2268215886
2268215887 2268215888 2268215889 2268215890 2268215891 2268215892 2268215893
2268215894 2268215895 2268215896 2268215897 2268215898 2268215899 2268215900
2268215901 2268215902 2268215903 2268215904 2268215905 2268215906 2268215907
2268215908 2268215909 2268215910 2268215911 2268215914 2268215915
My questions are
1. What exactly is postgres doing here CONTEXT: xlog redo xid assignment
.....
and here LOG: file "pg_subtrans/8732" doesn't exist, reading as
zeroes
Is it possible we have some data corruption?
2. I am also wondering if the "database system was interrupted while in
recovery at log time 2011-12-07 00:32:49 PST"
message is due to taking the snapshots while the servers are running.
On launch
we will shut them down prior to taking the final snapshot.
Any help is appreciated.
--
View this message in context:
http://postgresql.1045698.n5.nabble.com/Clarification-on-CONTEXT-xlog-redo-xid-assignment-tp5059725p5059725.html
Sent from the PostgreSQL - general mailing list archive at Nabble.com.
On Thu, Dec 8, 2011 at 6:37 PM, kc5 <kim@colandraengineering.com> wrote: > Dec 7 12:40:55 postgres[28199]: [ID 748848 local2.info] [5-1] LOG: file > "pg_subtrans/8732" doesn't exist, reading as zeroes > Dec 7 12:40:55 postgres[28199]: [ID 748848 local2.info] [5-2] CONTEXT: > xlog redo xid assignment xtop 2268215780: subxacts: 2268215781 2268215782 Looks like the bug fixed in 9.1.2 and 9.0.6 -- Simon Riggs http://www.2ndQuadrant.com/ PostgreSQL Development, 24x7 Support, Training & Services
What were the ramifications of the bug? I don't see any details in the release notes. It is not really an option to upgrade before the move. -- View this message in context: http://postgresql.1045698.n5.nabble.com/Clarification-on-CONTEXT-xlog-redo-xid-assignment-tp5059725p5060301.html Sent from the PostgreSQL - general mailing list archive at Nabble.com.