Re: Memory leak with XLogFileCopy since de768844 (WAL file with .partial)

From: Michael Paquier <michael(dot)paquier(at)gmail(dot)com>
To: hlinnaka(at)iki(dot)fi
Cc: Fujii Masao <masao(dot)fujii(at)gmail(dot)com>, PostgreSQL mailing lists <pgsql-hackers(at)postgresql(dot)org>
Subject: Re: Memory leak with XLogFileCopy since de768844 (WAL file with .partial)
Date: 2015-06-09 01:09:02
Message-ID: CAB7nPqT4Y32Z7AOLrvMhRpd=B4B2D++TOLji82kvNi2pmjYhpA@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

On Tue, Jun 9, 2015 at 6:25 AM, Heikki Linnakangas wrote:
> I'm still not sure if I should've just reverted that refactoring, to make
> XLogFileCopy() look the same in master and back-branches, which makes
> back-patching easier, or keep the refactoring, because it makes the code
> slightly nicer. But the current situation is the worst of both worlds: the
> interface of XLogFileCopy() is no better than it used to be, but it's
> different enough to cause merge conflicts. At this point, it's probably best
> to revert the code to look the same as in 9.4.

That's a valid concern. What about the attached then? I think that it
is still good to keep upto to copy only data up to the switch point at
recovery exit. InstallXLogFileSegment() changes a bit as well because
of its modifications of arguments.
--
Michael

Attachment Content-Type Size
20150609_xlogcopy_fixes_v4.patch text/x-diff 5.8 KB

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Michael Paquier 2015-06-09 01:33:04 Re: pg_stat_archiver issue with aborted archiver
Previous Message David Gould 2015-06-08 23:02:21 Re: [CORE] Restore-reliability mode