|
On 11/13/2016 11:52 PM, Yogesh Sharma
wrote:
you can look those relation numbers up in the pg_catalog to see what they are. you can see what the processes are in pg_stat_activity.Currently machine is not available. Please suggest if any other approach to identify the same. without access to the system thats generating the error, there's nothing you can learn. THis has nothing to do with the growing WAL logs... something is blocking checkpoints if a single WAL file keeps >growing, are you using some form of wal archiving, is that working correctly ? could something be preventing checkpoints? what are the related checkpoint and WAL settings?Many WAL files are generated in pg_xlog directory. I don’t know how I can check checkpoint and WAL settings? By default setting of postgresql.conf is used. I just noticed, you're talking about postgresql 8.1.18? thats an unsupported and obsolete version from 2009. A standalone postgres installation that hasn't been configured for WAL archiving will normally only generate up to 2*CHECKPOINT_SEGMENTS + 1 wal files of 16MB each (assuming all defaults, thats 7 files). The obsolete 8.1 documentation on this is here, https://www.postgresql.org/docs/8.1/static/wal-configuration.html If you've configured wal archiving, AND this isn't working correctly, WAL files can stack up indefinitely. See https://www.postgresql.org/docs/8.1/static/backup-online.html
-- john r pierce, recycling bits in santa cruz |
