Re: when the startup process doesn't (logging startup delays)
От | Robert Haas |
---|---|
Тема | Re: when the startup process doesn't (logging startup delays) |
Дата | |
Msg-id | CA+TgmoZtbqxaOLdpNkBcDbz=41tWALA8kpH4M=RWtPYHC7-KNg@mail.gmail.com обсуждение исходный текст |
Ответ на | Re: when the startup process doesn't (logging startup delays) (Justin Pryzby <pryzby@telsasoft.com>) |
Ответы |
Re: when the startup process doesn't (logging startup delays)
|
Список | pgsql-hackers |
On Sat, Aug 14, 2021 at 5:47 PM Justin Pryzby <pryzby@telsasoft.com> wrote: > Should this feature distinguish between crash recovery and archive recovery on > a hot standby ? Otherwise the standby will display this all the time. > > 2021-08-14 16:13:33.139 CDT startup[11741] LOG: redo in progress, elapsed time: 124.42 s, current LSN: 0/EEE2100 > > If so, I think maybe you'd check !InArchiveRecovery (but until Robert finishes > cleanup of xlog.c variables, I can't say that with much confidence). Hmm. My inclination is to think that on an actual standby, you wouldn't want to get messages like this, but if you were doing a point-in-time-recovery, then you would. So I think maybe InArchiveRecovery is not the right thing. Perhaps StandbyMode? -- Robert Haas EDB: http://www.enterprisedb.com
В списке pgsql-hackers по дате отправления: