|
I’ve been going over the log of the system at the time of the crash, but I’m not seeing something that stands out as telling
me anymore about the reason of either the crash or the failure to start the replication again. I’m attaching a part of the log file to this mail. Correct me if I’m wrong but the “select * from pg_stat_archiver” is linked with WAL archiving, correct ? Currently this
system has archiving switch off. For backup purposes we are running scheduled pg_dumps of each database.
So it didn’t give me any output. Which I think is normal since we switched it off. To setup the replication we used the pg_basebackup with the --xlog-method=stream option this way pg_basebackup will not only
copy the data as it is, but also stream the XLOG being created during the base backup to our destination server.
Not sure what the problem is since it appears to being able to perform these actions without any reported error. Cheers, Kristof From: Payal Singh [mailto:payals1@xxxxxxxx]
On Wed, Dec 6, 2017 at 8:07 AM, Bruyninckx Kristof <Kristof.Bruyninckx@xxxxxxxxxx> wrote:
Might want to check the master postgres logs during and after crash as well. Also, check for wal file progress on master (select * from pg_stat_archiver). -- Payal Singh |
Attachment:
db.log
Description: db.log
