Hi Matheus de Oliveira,
It is the same case with me too. The archive files are being generated to a
vast extent of 2 GB per hour. And the wal_keep_segments parameter is set 0.
And we are not following any replication practices. The size of the database
is 24GB. Daily 48GB of archive logs are being generated and this abnormal
behaviour started from December 28'th 2017.
The set of DML's which were ran before and after December 28'th 2017 are the
same. So why these many archive logs are being generated? Those are causing
a severe disk space problem.
And FYI autovacuum was set to on. But when I tried doing vacuumdb for this
database almost 1GB of archive logs are getting generated per 1 sec.
It seems to be very strange, any idea of why it is generating?(to prevent
transaction Id wraparound failures? If yes how long it will generate?)
I'm facing this issue from December 28'th 2017. Please suggest how to cope
with it. Thanks in Advance.
Regards,
Pavan
--
Sent from: http://www.postgresql-archive.org/PostgreSQL-admin-f2076596.html