Thread: Can we limit the total size of WAL?
Hi all, I hope this email finds you well. I'm new to PostgreSQL, please correct me if I'm mislead/incorrect. I have setup streaming replication using WAL The archive_command that was generated by Postgres-XL in postgresql.conf was: archive_command = 'rsync %p postgres@node03:/home/postgres/PGXL/nodes/datanodearchlog/02/%f' If I run check the total size of the folder in datanode master, it contains only 14GB: # du -sh /home/postgres/PGXL/nodes/datanodemaster/03/ 14G /home/postgres/PGXL/nodes/datanodemaster/03/ But if I check the total size of the archive folder in datanode slave, it contains 90GB, each file has 16MB size, with total files 5749: # du -sh /home/postgres/PGXL/nodes/datanodearchlog/03/ 90G /home/postgres/PGXL/nodes/datanodearchlog/03/ # ls -al /home/postgres/PGXL/nodes/datanodearchlog/03/ | grep 16777216 | wc -l 5749 How to limit the amount of files generated by WAL archiving? Or maybe if we can, how to recycle the files generated by WAL archiving? Thank you, FattahRozzaq
On Wed, Sep 9, 2015 at 3:41 PM, FattahRozzaq <ssoorruu@gmail.com> wrote: > The archive_command that was generated by Postgres-XL in postgresql.conf was: When using Postgres-XL, I guess that asking directly the XL guys is more appropriate than here, but well your question does apply to Postgres as well... > But if I check the total size of the archive folder in datanode slave, > it contains 90GB, each file has 16MB size, with total files 5749: > # du -sh /home/postgres/PGXL/nodes/datanodearchlog/03/ > 90G /home/postgres/PGXL/nodes/datanodearchlog/03/ > # ls -al /home/postgres/PGXL/nodes/datanodearchlog/03/ | grep 16777216 | wc -l > 5749 Normally standbys don't do archiving (Except in 9.5~ but XL has just synced up with it). So I am not really following. Aren't those files generated and archived by a master datanode? > How to limit the amount of files generated by WAL archiving? > Or maybe if we can, how to recycle the files generated by WAL archiving? pg_archivecleanup [1] is one method by setting up archive_cleanup_command in the recovery.conf file of the involved standby. Now be careful of two things: 1) Is another standby node in need of those files? Be careful to not cleanup an archive folder when multiple nodes fetch from it. 2) Do you expect a set of base backups to need those files or not? [1]: http://www.postgresql.org/docs/devel/static/pgarchivecleanup.html Regards, -- Michael