Received: from malur.postgresql.org ([217.196.149.56]) by arkaria.postgresql.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_CBC_SHA1:256) (Exim 4.89) (envelope-from ) id 1iQv2c-0007sl-6S for pgsql-bugs@arkaria.postgresql.org; Sat, 02 Nov 2019 15:10:14 +0000 Received: from localhost ([127.0.0.1] helo=malur.postgresql.org) by malur.postgresql.org with esmtp (Exim 4.89) (envelope-from ) id 1iQv2a-0003hs-NB for pgsql-bugs@arkaria.postgresql.org; Sat, 02 Nov 2019 15:10:12 +0000 Received: from makus.postgresql.org ([2001:4800:3e1:1::229]) by malur.postgresql.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_CBC_SHA1:256) (Exim 4.89) (envelope-from ) id 1iQv2a-0003hl-CA for pgsql-bugs@lists.postgresql.org; Sat, 02 Nov 2019 15:10:12 +0000 Received: from sss.pgh.pa.us ([66.207.139.130]) by makus.postgresql.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.92) (envelope-from ) id 1iQv2W-00027s-T4 for pgsql-bugs@lists.postgresql.org; Sat, 02 Nov 2019 15:10:11 +0000 Received: from sss1.sss.pgh.pa.us (localhost [127.0.0.1]) by sss.pgh.pa.us (8.14.4/8.14.4) with ESMTP id xA2FA6Y9030150; Sat, 2 Nov 2019 11:10:06 -0400 From: Tom Lane To: Alexey Ermakov cc: Sergei Kornilov , "pgsql-bugs@lists.postgresql.org" Subject: Re: BUG #16016: deadlock with startup process, AccessExclusiveLock on pg_statistic's toast table In-reply-to: <5DBBF7B5.6040002@dataegret.com> References: <16016-ed0345229ddc484c@postgresql.org> <158201568991600@myt5-2a1eccb4c218.qloud-c.yandex.net> <21601.1569179167@sss.pgh.pa.us> <5DBBF7B5.6040002@dataegret.com> Comments: In-reply-to Alexey Ermakov message dated "Fri, 01 Nov 2019 15:15:33 +0600" MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-ID: <30148.1572707406.1@sss.pgh.pa.us> Content-Transfer-Encoding: quoted-printable Date: Sat, 02 Nov 2019 11:10:06 -0400 Message-ID: <30149.1572707406@sss.pgh.pa.us> List-Id: List-Help: List-Subscribe: List-Post: List-Owner: List-Archive: Precedence: bulk Alexey Ermakov writes: > On 9/23/19 01:06, Tom Lane wrote: >> If you have a test case, could you collect stack traces from each >> of the stuck processes? That would eliminate a lot of hypothesizing. > I reproduced Sergei's test case on postgresql 11.5, replica hung up = > almost immediately after pgbench ran. > stack trace of startup process 9907: > #0 0x00007fdff4a025b3 in __select_nocancel () at = > ../sysdeps/unix/syscall-template.S:84 > #1 0x00005608fde0c6cd in pg_usleep (microsec=3D) at = > /build/postgresql-11-d6c2wG/postgresql-11-11.5/build/../src/port/pgsleep= .c:56 > #2 0x00005608fdc94126 in WaitExceedsMaxStandbyDelay () at = > /build/postgresql-11-d6c2wG/postgresql-11-11.5/build/../src/backend/stor= age/ipc/standby.c:201 > #3 ResolveRecoveryConflictWithVirtualXIDs (waitlist=3D0x5608fe46a450, = > reason=3Dreason@entry=3DPROCSIG_RECOVERY_CONFLICT_SNAPSHOT) at = > /build/postgresql-11-d6c2wG/postgresql-11-11.5/build/../src/backend/stor= age/ipc/standby.c:262 This does not look like a deadlock: the startup process is just biding its time until the standby delay timeout elapses, after which it's gonna kill the conflicting queries. It is, perhaps, arguable that it's a damn bad idea to allow max_standby_streaming_delay or max_standby_archive_delay to be set to "forever". regards, tom lane