From: | Paul van der Linden <paul(dot)vanderlinden(at)mapcreator(dot)eu> |
---|---|
To: | "pgsql-bugs(at)lists(dot)postgresql(dot)org" <pgsql-bugs(at)lists(dot)postgresql(dot)org> |
Subject: | Possible regression from PG10 to PG11 on windows |
Date: | 2019-02-13 18:44:22 |
Message-ID: | AM0PR0402MB3425DD01DC0DF74C770E3A5DEC660@AM0PR0402MB3425.eurprd04.prod.outlook.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-bugs |
Hi,
I have a suspicion that since the upgrade of Postgres to 11 processes are not fully closed after their job is done.
The symptom I noticed IS that the process-ids are steadily increasing for new proccesses, they used to be in the 4-digit numbers with older Postgres versions, and now they are in the 6-digit range and still increasing.
It doesn't seem to have any other effects than that, but in theory it could run out of resources of some kind given enough time.
A possible cause of this could be that a handle to a finished backend-process isn't closed, and thus still hold a reference to that process (and prevent reuse of it's process-id).
Didn't investigate deeper yet, but perhaps this triggers someone's mind enough to know the cause/solution?
Paul
From | Date | Subject | |
---|---|---|---|
Next Message | Michael Paquier | 2019-02-14 01:30:06 | Re: BUG #15631: Generated as identity field in a temporary table with on commit drop corrupts system catalogs |
Previous Message | Andrew Gierth | 2019-02-13 14:52:06 | Re: BUG #15634: extension missing |