Re: Reopen logfile on SIGHUP
Kyotaro HORIGUCHI <horiguchi.kyotaro@lab.ntt.co.jp>
From: Kyotaro HORIGUCHI <horiguchi.kyotaro@lab.ntt.co.jp>
To: a.kuzmenkov@postgrespro.ru
Cc: sk@zsrv.org, tgl@sss.pgh.pa.us, robertmhaas@gmail.com, andres@anarazel.de,
stark@mit.edu, a.lubennikova@postgrespro.ru,
pgsql-hackers@postgresql.org, g.smolkin@postgrespro.ru
Date: 2018-08-09T07:33:15Z
Lists: pgsql-hackers
Attachments
- v5-0001-New-pg_ctl-subcommand-logrotate.patch (text/x-patch)
Hello. At Tue, 7 Aug 2018 16:36:34 +0300, Alexander Kuzmenkov <a.kuzmenkov@postgrespro.ru> wrote in <a9d2ad92-b4f9-42fb-77eb-f3085b964705@postgrespro.ru> > I think the latest v4 patch addresses the concerns raised > upthread. I'm marking the commitfest entry as RFC. Thank you, but it is forgetting pg_ctl --help output. I added it and moved logrotate entry in docs just above "kill" in app-pg-ctl.html. Also added "-s" option. Furthermore, I did the following things. - Since I'm not sure unlink is signal-handler safe on all supported platforms, I moved unlink() call out of checkLogrotateSignal() to SysLoggerMain, just before rotating log file. - Refactored moving the main stuff to syslogger.c. - The diff is splitted into two files and renamed. (but version number continues to v5). I'm not certain whether to allow the signal file alone cause rotation, but this patch doesn't. Directly placed signal file is to be removed without causing rotation. Please find the attached. regards. -- Kyotaro Horiguchi NTT Open Source Software Center
Commits
-
Implement "pg_ctl logrotate" command
- ec7436993168 12.0 landed