Re: .ready and .done files considered harmful
От | Kyotaro Horiguchi |
---|---|
Тема | Re: .ready and .done files considered harmful |
Дата | |
Msg-id | 20210824.113655.27283254519813419.horikyota.ntt@gmail.com обсуждение исходный текст |
Ответ на | Re: .ready and .done files considered harmful (Kyotaro Horiguchi <horikyota.ntt@gmail.com>) |
Ответы |
Re: .ready and .done files considered harmful
|
Список | pgsql-hackers |
(sigh..) At Tue, 24 Aug 2021 11:35:06 +0900 (JST), Kyotaro Horiguchi <horikyota.ntt@gmail.com> wrote in > > IIUC partial WAL files are handled because the next file in the > > sequence with the given TimeLineID won't be there, so we will fall > > back to a directory scan and pick it up. Timeline history files are > > handled by forcing a directory scan, which should work because they > > always have the highest priority. Backup history files, however, do > > not seem to be handled. I think one approach to fixing that is to > > also treat backup history files similarly to timeline history files. > > If one is created, we force a directory scan, and the directory scan > > logic will consider backup history files as higher priority than > > everything but timeline history files. > > Backup history files are (currently) just informational and they are > finally processed at the end of a bulk-archiving performed by the fast > path. However, I feel that it is cleaner to trigger a directory scan > every time we add an other-than-a-regular-WAL-file, as base-backup or - promotion are not supposed happen so infrequently. + promotion are not supposed happen so frequently. -- Kyotaro Horiguchi NTT Open Source Software Center
В списке pgsql-hackers по дате отправления: