Re: [HACKERS] Reporting xmin from VACUUMs
От | Jim Nasby |
---|---|
Тема | Re: [HACKERS] Reporting xmin from VACUUMs |
Дата | |
Msg-id | 7f7a2d88-4aeb-b17c-932d-a3fe499812fa@BlueTreble.com обсуждение исходный текст |
Ответ на | [HACKERS] Reporting xmin from VACUUMs (Simon Riggs <simon@2ndquadrant.com>) |
Ответы |
Re: [HACKERS] Reporting xmin from VACUUMs
|
Список | pgsql-hackers |
Accidentally sent o--list. On 2/14/17 12:57 PM, Jim Nasby wrote: > On 2/13/17 12:36 AM, Masahiko Sawada wrote: >>> It seems sensible to me to do something like this. We already report >>> a lot of other fine details, so what's one more? And it could be >>> useful. >> >> Also, I've been proposing to report the number of skipped the frozen >> pages even in manual vacuum verbose log. > > Yes, please. We now do that for pages skipped because we couldn't get > the cleanup lock and skipping due to all-frozen is certainly more > user-visible than that. > > ISTR previous discussion of allowing more stats files; if that happened > I think having stats that were dedicated to (auto)vacuum would be very > useful. That's clearly a lot more work though. -- Jim Nasby, Data Architect, Blue Treble Consulting, Austin TX Experts in Analytics, Data Architecture and PostgreSQL Data in Trouble? Get it in Treble! http://BlueTreble.com 855-TREBLE2 (855-873-2532)
В списке pgsql-hackers по дате отправления: