From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
---|---|
To: | "Mikheev, Vadim" <vmikheev(at)SECTORBASE(dot)COM> |
Cc: | Hiroshi Inoue <Inoue(at)tpf(dot)co(dot)jp>, pgsql-hackers(at)postgresql(dot)org |
Subject: | Re: WAL does not recover gracefully from out-of-disk-sp ace |
Date: | 2001-03-09 20:30:45 |
Message-ID: | 8571.984169845@sss.pgh.pa.us |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
"Mikheev, Vadim" <vmikheev(at)SECTORBASE(dot)COM> writes:
>> True. But at least the write is (hopefully) being done at a
>> non-performance-critical time.
> There is no such hope: XLogWrite may be called from XLogFlush
> (at commit time and from bufmgr on replacements) *and* from XLogInsert
> - ie new log file may be required at any time.
Sure, but if we have create-ahead enabled then there's a good chance of
the log files being made by the checkpoint process, rather than by
working backends. In that case the prefill is not time critical.
In any case, my tests so far show that prefilling and then writing with
O_SYNC or better O_DSYNC is in fact faster than not prefilling; this
matches pretty well the handwaving argument I gave Andreas this morning.
(With fsync() or fdatasync() it seems we're at the mercy of inefficient
kernel algorithms, a factor I didn't consider before.)
regards, tom lane
From | Date | Subject | |
---|---|---|---|
Next Message | Peter Eisentraut | 2001-03-09 20:45:14 | Re: Internationalized error messages |
Previous Message | Tom Lane | 2001-03-09 20:21:22 | Re: porting question: funky uid names? |