Thread: Recovery problem with PostgreSQL
While running the recovery process I am getting the following errors for cannot stat:
cp: cannot stat `/data1_node5/arch/00000001.history': No such file or
directory
cp: cannot stat
`/data1_node5/arch/000000010000000000000007.02C4C530.backup': No such
file or directory
cp: cannot stat `/data1_node5/arch/000000010000000000000007': No such
file or directory
2006-05-10 17:07:35 PDT LOG: checkpoint record is at 0/1EC4C558
2006-05-10 17:07:35 PDT LOG: redo record is at 0/1EC4C530; undo record
is at 0/0; shutdown FALSE
2006-05-10 17:07:35 PDT LOG: next transaction ID: 1227111; next OID:
35682
2006-05-10 17:07:35 PDT LOG: next MultiXactId: 168; next
MultiXactOffset: 338
2006-05-10 17:07:35 PDT LOG: automatic recovery in progress
2006-05-10 17:07:35 PDT LOG: redo starts at 0/1EC4C530
2006-05-10 17:07:35 PDT LOG: record with zero length at 0/1EF5C91C
2006-05-10 17:07:35 PDT LOG: redo done at 0/1EF5C8F4
cp: cannot stat `/data1_node5/arch/000000010000000000000007': No such
file or directory
2006-05-10 17:07:35 PDT LOG: archive recovery complete
2006-05-10 17:07:35 PDT LOG: database system is ready
2006-05-10 17:07:35 PDT LOG: transaction ID wrap limit is 2147484188,
Any idea what can be causing these?
Regards,
Shoaib
cp: cannot stat `/data1_node5/arch/00000001.history': No such file or
directory
cp: cannot stat
`/data1_node5/arch/000000010000000000000007.02C4C530.backup': No such
file or directory
cp: cannot stat `/data1_node5/arch/000000010000000000000007': No such
file or directory
2006-05-10 17:07:35 PDT LOG: checkpoint record is at 0/1EC4C558
2006-05-10 17:07:35 PDT LOG: redo record is at 0/1EC4C530; undo record
is at 0/0; shutdown FALSE
2006-05-10 17:07:35 PDT LOG: next transaction ID: 1227111; next OID:
35682
2006-05-10 17:07:35 PDT LOG: next MultiXactId: 168; next
MultiXactOffset: 338
2006-05-10 17:07:35 PDT LOG: automatic recovery in progress
2006-05-10 17:07:35 PDT LOG: redo starts at 0/1EC4C530
2006-05-10 17:07:35 PDT LOG: record with zero length at 0/1EF5C91C
2006-05-10 17:07:35 PDT LOG: redo done at 0/1EF5C8F4
cp: cannot stat `/data1_node5/arch/000000010000000000000007': No such
file or directory
2006-05-10 17:07:35 PDT LOG: archive recovery complete
2006-05-10 17:07:35 PDT LOG: database system is ready
2006-05-10 17:07:35 PDT LOG: transaction ID wrap limit is 2147484188,
Any idea what can be causing these?
Regards,
Shoaib
"Shoaib Mir" <shoaibmir@gmail.com> writes: > While running the recovery process I am getting the following errors for > cannot stat: > cp: cannot stat `/data1_node5/arch/00000001.history': No such file or > directory What makes you think this is an error? As pointed out in the docs, the archive_recovery command *will* be asked for files that don't exist. You might want to tweak your script to not report an error in this case. regards, tom lane
That should be true about `/data1_node5/arch/00000001.history' since there may not be any timelines created by previous archiver recoveries
But the message
cp: cannot stat `/data1_node5/arch/000000010000000000000007''
suggests that it was not able to restore one of the log files while doing the recovery So it can be treated as an error I guess due to missing log file in the archive folder. That might be due to the archive command not setup correctly and as a result it was not able to copy files from pg_xlog to archive folder.
/Shoaib
But the message
cp: cannot stat `/data1_node5/arch/000000010000000000000007''
suggests that it was not able to restore one of the log files while doing the recovery So it can be treated as an error I guess due to missing log file in the archive folder. That might be due to the archive command not setup correctly and as a result it was not able to copy files from pg_xlog to archive folder.
/Shoaib
On 5/11/06, Tom Lane <tgl@sss.pgh.pa.us> wrote:
"Shoaib Mir" <shoaibmir@gmail.com> writes:
> While running the recovery process I am getting the following errors for
> cannot stat:
> cp: cannot stat `/data1_node5/arch/00000001.history': No such file or
> directory
What makes you think this is an error? As pointed out in the docs,
the archive_recovery command *will* be asked for files that don't
exist. You might want to tweak your script to not report an error
in this case.
regards, tom lane