diff --git a/doc/TODO.detail/win32intr b/doc/TODO.detail/win32intr new file mode 100644 index 0000000000..479658532f --- /dev/null +++ b/doc/TODO.detail/win32intr @@ -0,0 +1,535 @@ +From pgsql-bugs-owner+M14436@postgresql.org Tue Feb 21 22:21:37 2006 +Return-path: +Received: from ams.hub.org (ams.hub.org [200.46.204.13]) + by candle.pha.pa.us (8.11.6/8.11.6) with ESMTP id k1M3La514372 + for ; Tue, 21 Feb 2006 22:21:37 -0500 (EST) +Received: from postgresql.org (postgresql.org [200.46.204.71]) + by ams.hub.org (Postfix) with ESMTP id 3F65267B6A8 + for ; Tue, 21 Feb 2006 23:21:32 -0400 (AST) +X-Original-To: pgsql-bugs-postgresql.org@localhost.postgresql.org +Received: from localhost (av.hub.org [200.46.204.144]) + by postgresql.org (Postfix) with ESMTP id 39FB99DCBE3 + for ; Tue, 21 Feb 2006 23:21:08 -0400 (AST) +Received: from postgresql.org ([200.46.204.71]) + by localhost (av.hub.org [200.46.204.144]) (amavisd-new, port 10024) + with ESMTP id 11699-01 + for ; + Tue, 21 Feb 2006 23:21:12 -0400 (AST) +X-Greylist: from auto-whitelisted by SQLgrey- +Received: from floppy.pyrenet.fr (news.pyrenet.fr [194.116.145.2]) + by postgresql.org (Postfix) with ESMTP id 271CB9DCABD + for ; Tue, 21 Feb 2006 23:21:05 -0400 (AST) +Received: by floppy.pyrenet.fr (Postfix, from userid 106) + id 90AFD30BB2; Wed, 22 Feb 2006 04:21:09 +0100 (MET) +From: "Qingqing Zhou" +X-Newsgroups: pgsql.bugs +Subject: [BUGS] FATAL: semctl(1672698088, 12, SETVAL, 0) failed +Date: Wed, 22 Feb 2006 11:17:51 +0800 +Organization: Hub.Org Networking Services +Lines: 50 +Message-ID: +X-Complaints-To: usenet@news.hub.org +X-Priority: 3 +X-MSMail-Priority: Normal +X-Newsreader: Microsoft Outlook Express 6.00.2800.1506 +X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1506 +To: pgsql-bugs@postgresql.org +X-Virus-Scanned: by amavisd-new at hub.org +X-Spam-Status: No, score=0.16 required=5 tests=[AWL=0.160] +X-Spam-Score: 0.16 +X-Mailing-List: pgsql-bugs +List-Archive: +List-Help: +List-Id: +List-Owner: +List-Post: +List-Subscribe: +List-Unsubscribe: +Precedence: bulk +Sender: pgsql-bugs-owner@postgresql.org +Status: ORr + +I encountered an error when I fast shutdown 8.1.1 on Win2k: + + FATAL: semctl(1672698088, 12, SETVAL, 0) failed: A blocking operation + was interrupted by a call to WSACancelBlockingCall. + +A similar error on 8.1/win2003 was reported on pgsql-general (sorry, I can't +dig out the +original post from our web archives): + + From: Niederland + Date: Tues, Dec 13 2005 9:49 am + + 2005-12-12 20:30:00 FATAL: semctl(50884184, 15, SETVAL, 0) failed: A + non-blocking socket operation could not be completed immediately. + +--- + +There are two problems here: + +(1) Why a socket error? +In port/win32.h, we have + +#undef EAGAIN +#undef EINTR +#define EINTR WSAEINTR +#define EAGAIN WSAEWOULDBLOCK + +What's the rationale of doing so? + +(2) What's happened here? +It may come from PGSemaphoreReset(), and win32 semop() looks like this: + + ret = WaitForMultipleObjectsEx(2, wh, FALSE, (sops[0].sem_flg & +IPC_NOWAIT) ? 0 : INFINITE, TRUE); + ... + else if (ret == WAIT_OBJECT_0 + 1 || ret == WAIT_IO_COMPLETION) + { + pgwin32_dispatch_queued_signals(); + errno = EINTR; + } + else if (ret == WAIT_TIMEOUT) + errno = EAGAIN; + +So it seems the EINTR is caused by an incoming signal, the EAGAIN is caused +by a TIMEOUT ... any ideas? + +Regards, +Qingqing + + + +---------------------------(end of broadcast)--------------------------- +TIP 9: In versions below 8.0, the planner will ignore your desire to + choose an index scan if your joining column's datatypes do not + match + +From pgsql-bugs-owner+M14482@postgresql.org Tue Feb 28 14:05:23 2006 +Return-path: +Received: from ams.hub.org (ams.hub.org [200.46.204.13]) + by candle.pha.pa.us (8.11.6/8.11.6) with ESMTP id k1SJ5Mk19122 + for ; Tue, 28 Feb 2006 14:05:22 -0500 (EST) +Received: from postgresql.org (postgresql.org [200.46.204.71]) + by ams.hub.org (Postfix) with ESMTP id A3FB567B7BF + for ; Tue, 28 Feb 2006 15:05:20 -0400 (AST) +X-Original-To: pgsql-bugs-postgresql.org@localhost.postgresql.org +Received: from localhost (av.hub.org [200.46.204.144]) + by postgresql.org (Postfix) with ESMTP id DB1559DCB49 + for ; Tue, 28 Feb 2006 15:05:04 -0400 (AST) +Received: from postgresql.org ([200.46.204.71]) + by localhost (av.hub.org [200.46.204.144]) (amavisd-new, port 10024) + with ESMTP id 86421-08 + for ; + Tue, 28 Feb 2006 15:05:04 -0400 (AST) +X-Greylist: from auto-whitelisted by SQLgrey- +Received: from candle.pha.pa.us (candle.pha.pa.us [70.90.9.53]) + by postgresql.org (Postfix) with ESMTP id 16D669DCAA8 + for ; Tue, 28 Feb 2006 15:05:01 -0400 (AST) +Received: (from pgman@localhost) + by candle.pha.pa.us (8.11.6/8.11.6) id k1SJ52b19021; + Tue, 28 Feb 2006 14:05:02 -0500 (EST) +From: Bruce Momjian +Message-ID: <200602281905.k1SJ52b19021@candle.pha.pa.us> +Subject: Re: [BUGS] FATAL: semctl(1672698088, 12, SETVAL, 0) failed +In-Reply-To: +To: Qingqing Zhou +Date: Tue, 28 Feb 2006 14:05:02 -0500 (EST) +cc: pgsql-bugs@postgresql.org +X-Mailer: ELM [version 2.4ME+ PL121 (25)] +MIME-Version: 1.0 +Content-Transfer-Encoding: 7bit +Content-Type: text/plain; charset=US-ASCII +X-Virus-Scanned: by amavisd-new at hub.org +X-Spam-Status: No, score=0.4 required=5 tests=[AWL=-0.100, RAZOR2_CHECK=0.5] +X-Spam-Score: 0.4 +X-Mailing-List: pgsql-bugs +List-Archive: +List-Help: +List-Id: +List-Owner: +List-Post: +List-Subscribe: +List-Unsubscribe: +Precedence: bulk +Sender: pgsql-bugs-owner@postgresql.org +Status: OR + +Qingqing Zhou wrote: +> I encountered an error when I fast shutdown 8.1.1 on Win2k: +> +> FATAL: semctl(1672698088, 12, SETVAL, 0) failed: A blocking operation +> was interrupted by a call to WSACancelBlockingCall. +> +> A similar error on 8.1/win2003 was reported on pgsql-general (sorry, I can't +> dig out the +> original post from our web archives): +> +> From: Niederland +> Date: Tues, Dec 13 2005 9:49 am +> +> 2005-12-12 20:30:00 FATAL: semctl(50884184, 15, SETVAL, 0) failed: A +> non-blocking socket operation could not be completed immediately. +> +> --- +> +> There are two problems here: +> +> (1) Why a socket error? +> In port/win32.h, we have +> +> #undef EAGAIN +> #undef EINTR +> #define EINTR WSAEINTR +> #define EAGAIN WSAEWOULDBLOCK +> +> What's the rationale of doing so? + +We did this so that our code could refer to EINTR/EAGAIN without +port-specific tests. + +> (2) What's happened here? +> It may come from PGSemaphoreReset(), and win32 semop() looks like this: +> +> ret = WaitForMultipleObjectsEx(2, wh, FALSE, (sops[0].sem_flg & +> IPC_NOWAIT) ? 0 : INFINITE, TRUE); +> ... +> else if (ret == WAIT_OBJECT_0 + 1 || ret == WAIT_IO_COMPLETION) +> { +> pgwin32_dispatch_queued_signals(); +> errno = EINTR; +> } +> else if (ret == WAIT_TIMEOUT) +> errno = EAGAIN; +> +> So it seems the EINTR is caused by an incoming signal, the EAGAIN is caused +> by a TIMEOUT ... any ideas? + +I looked at the documentation for the function: + + http://msdn.microsoft.com/library/default.asp?url=/library/en-us/dllproc/base/waitformultipleobjectsex.asp + +and it isn't clear what return failure values it has. We certainly +could loop on WSAEINTR. Can you test it? + +-- + Bruce Momjian http://candle.pha.pa.us + SRA OSS, Inc. http://www.sraoss.com + + + If your life is a hard drive, Christ can be your backup. + + +---------------------------(end of broadcast)--------------------------- +TIP 3: Have you checked our extensive FAQ? + + http://www.postgresql.org/docs/faq + +From pgsql-bugs-owner+M14483@postgresql.org Tue Feb 28 21:12:39 2006 +Return-path: +Received: from ams.hub.org (ams.hub.org [200.46.204.13]) + by candle.pha.pa.us (8.11.6/8.11.6) with ESMTP id k212Cdk24423 + for ; Tue, 28 Feb 2006 21:12:39 -0500 (EST) +Received: from postgresql.org (postgresql.org [200.46.204.71]) + by ams.hub.org (Postfix) with ESMTP id 2D9CF67B7CA + for ; Tue, 28 Feb 2006 22:12:36 -0400 (AST) +X-Original-To: pgsql-bugs-postgresql.org@localhost.postgresql.org +Received: from localhost (av.hub.org [200.46.204.144]) + by postgresql.org (Postfix) with ESMTP id D8B0B9DC837 + for ; Tue, 28 Feb 2006 22:12:17 -0400 (AST) +Received: from postgresql.org ([200.46.204.71]) + by localhost (av.hub.org [200.46.204.144]) (amavisd-new, port 10024) + with ESMTP id 65124-06 + for ; + Tue, 28 Feb 2006 22:12:21 -0400 (AST) +X-Greylist: from auto-whitelisted by SQLgrey- +Received: from floppy.pyrenet.fr (news.pyrenet.fr [194.116.145.2]) + by postgresql.org (Postfix) with ESMTP id 7C5729DC81D + for ; Tue, 28 Feb 2006 22:12:15 -0400 (AST) +Received: by floppy.pyrenet.fr (Postfix, from userid 106) + id 6A8CA30BA5; Wed, 1 Mar 2006 03:12:19 +0100 (MET) +From: "Qingqing Zhou" +X-Newsgroups: pgsql.bugs +Subject: Re: [BUGS] FATAL: semctl(1672698088, 12, SETVAL, 0) failed +Date: Wed, 1 Mar 2006 10:08:58 +0800 +Organization: Hub.Org Networking Services +Lines: 68 +Message-ID: +References: <200602281905.k1SJ52b19021@candle.pha.pa.us> +X-Complaints-To: usenet@news.hub.org +X-Priority: 3 +X-MSMail-Priority: Normal +X-Newsreader: Microsoft Outlook Express 6.00.2800.1506 +X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1506 +To: pgsql-bugs@postgresql.org +X-Virus-Scanned: by amavisd-new at hub.org +X-Spam-Status: No, score=0.169 required=5 tests=[AWL=0.169] +X-Spam-Score: 0.169 +X-Mailing-List: pgsql-bugs +List-Archive: +List-Help: +List-Id: +List-Owner: +List-Post: +List-Subscribe: +List-Unsubscribe: +Precedence: bulk +Sender: pgsql-bugs-owner@postgresql.org +Status: ORr + + +"Bruce Momjian" wrote +> > In port/win32.h, we have +> > +> > #undef EAGAIN +> > #undef EINTR +> > #define EINTR WSAEINTR +> > #define EAGAIN WSAEWOULDBLOCK +> > +> > What's the rationale of doing so? +> +> We did this so that our code could refer to EINTR/EAGAIN without +> port-specific tests. +> + +AFAICS, by doing so, the EINTR/EAGAIN will be translated into +WSAINTR/WSAEWOULDBLOCK through *all* the backend code. That's seems not +appropriate for the code not involving any socket stuff ... I think we need +a fix here. + +> > (2) What's happened here? +> > It may come from PGSemaphoreReset(), and win32 semop() looks like this: +> > +> > ret = WaitForMultipleObjectsEx(2, wh, FALSE, (sops[0].sem_flg & +> > IPC_NOWAIT) ? 0 : INFINITE, TRUE); +> > ... +> > else if (ret == WAIT_OBJECT_0 + 1 || ret == WAIT_IO_COMPLETION) +> > { +> > pgwin32_dispatch_queued_signals(); +> > errno = EINTR; +> > } +> > else if (ret == WAIT_TIMEOUT) +> > errno = EAGAIN; +> > +> > So it seems the EINTR is caused by an incoming signal, the EAGAIN is +caused +> > by a TIMEOUT ... any ideas? +> +> I looked at the documentation for the function: +> +> +http://msdn.microsoft.com/library/default.asp?url=/library/en-us/dllproc/base/waitformultipleobjectsex.asp +> +> and it isn't clear what return failure values it has. We certainly +> could loop on WSAEINTR. Can you test it? +> + +Yeah, looking at other code of using semop(), we could plug in a loop in the +win32 semctl(): + + /* Quickly lock/unlock the semaphore (if we can) */ ++ do ++ { ++ errStatus = semop(semId, &sops, 1); ++ } while (errStatus < 0 && errno == EINTR); + + if (semop(semId, &sops, 1) < 0) + return -1; + +But: +(1) The EINTR problem happens rather rare, so testing it is difficult; +(2) I would rather not doing the above changes before we understand what's +happened here, especially when we have seen a EAGAIN reported here. + +Regards, +Qingqing + + + +---------------------------(end of broadcast)--------------------------- +TIP 1: if posting/reading through Usenet, please send an appropriate + subscribe-nomail command to majordomo@postgresql.org so that your + message can get through to the mailing list cleanly + +From pgsql-bugs-owner+M14485@postgresql.org Tue Feb 28 22:08:57 2006 +Return-path: +Received: from ams.hub.org (ams.hub.org [200.46.204.13]) + by candle.pha.pa.us (8.11.6/8.11.6) with ESMTP id k2138uk09057 + for ; Tue, 28 Feb 2006 22:08:56 -0500 (EST) +Received: from postgresql.org (postgresql.org [200.46.204.71]) + by ams.hub.org (Postfix) with ESMTP id B4EFF67B7CA + for ; Tue, 28 Feb 2006 23:08:53 -0400 (AST) +X-Original-To: pgsql-bugs-postgresql.org@localhost.postgresql.org +Received: from localhost (av.hub.org [200.46.204.144]) + by postgresql.org (Postfix) with ESMTP id 4A2B39DCB5D + for ; Tue, 28 Feb 2006 23:08:34 -0400 (AST) +Received: from postgresql.org ([200.46.204.71]) + by localhost (av.hub.org [200.46.204.144]) (amavisd-new, port 10024) + with ESMTP id 77300-07 + for ; + Tue, 28 Feb 2006 23:08:38 -0400 (AST) +X-Greylist: from auto-whitelisted by SQLgrey- +Received: from candle.pha.pa.us (candle.pha.pa.us [70.90.9.53]) + by postgresql.org (Postfix) with ESMTP id E8B109DCB0D + for ; Tue, 28 Feb 2006 23:08:31 -0400 (AST) +Received: (from pgman@localhost) + by candle.pha.pa.us (8.11.6/8.11.6) id k2138bO09022; + Tue, 28 Feb 2006 22:08:37 -0500 (EST) +From: Bruce Momjian +Message-ID: <200603010308.k2138bO09022@candle.pha.pa.us> +Subject: Re: [BUGS] FATAL: semctl(1672698088, 12, SETVAL, 0) failed +In-Reply-To: +To: Qingqing Zhou +Date: Tue, 28 Feb 2006 22:08:37 -0500 (EST) +cc: pgsql-bugs@postgresql.org +X-Mailer: ELM [version 2.4ME+ PL121 (25)] +MIME-Version: 1.0 +Content-Transfer-Encoding: 7bit +Content-Type: text/plain; charset=US-ASCII +X-Virus-Scanned: by amavisd-new at hub.org +X-Spam-Status: No, score=0.404 required=5 tests=[AWL=-0.096, RAZOR2_CHECK=0.5] +X-Spam-Score: 0.404 +X-Mailing-List: pgsql-bugs +List-Archive: +List-Help: +List-Id: +List-Owner: +List-Post: +List-Subscribe: +List-Unsubscribe: +Precedence: bulk +Sender: pgsql-bugs-owner@postgresql.org +Status: OR + +Qingqing Zhou wrote: +> +> "Bruce Momjian" wrote +> > > In port/win32.h, we have +> > > +> > > #undef EAGAIN +> > > #undef EINTR +> > > #define EINTR WSAEINTR +> > > #define EAGAIN WSAEWOULDBLOCK +> > > +> > > What's the rationale of doing so? +> > +> > We did this so that our code could refer to EINTR/EAGAIN without +> > port-specific tests. +> > +> +> AFAICS, by doing so, the EINTR/EAGAIN will be translated into +> WSAINTR/WSAEWOULDBLOCK through *all* the backend code. That's seems not +> appropriate for the code not involving any socket stuff ... I think we need +> a fix here. + +Uh, how do we handle it now? I thought we did just that. + +> http://msdn.microsoft.com/library/default.asp?url=/library/en-us/dllproc/base/waitformultipleobjectsex.asp +> > +> > and it isn't clear what return failure values it has. We certainly +> > could loop on WSAEINTR. Can you test it? +> > +> +> Yeah, looking at other code of using semop(), we could plug in a loop in the +> win32 semctl(): +> +> /* Quickly lock/unlock the semaphore (if we can) */ +> + do +> + { +> + errStatus = semop(semId, &sops, 1); +> + } while (errStatus < 0 && errno == EINTR); +> +> if (semop(semId, &sops, 1) < 0) +> return -1; +> +> But: +> (1) The EINTR problem happens rather rare, so testing it is difficult; +> (2) I would rather not doing the above changes before we understand what's +> happened here, especially when we have seen a EAGAIN reported here. + +OK, so how do we find the answer? + +-- + Bruce Momjian http://candle.pha.pa.us + SRA OSS, Inc. http://www.sraoss.com + + + If your life is a hard drive, Christ can be your backup. + + +---------------------------(end of broadcast)--------------------------- +TIP 2: Don't 'kill -9' the postmaster + +From pgsql-bugs-owner+M14486@postgresql.org Tue Feb 28 22:21:21 2006 +Return-path: +Received: from ams.hub.org (ams.hub.org [200.46.204.13]) + by candle.pha.pa.us (8.11.6/8.11.6) with ESMTP id k213LKk11157 + for ; Tue, 28 Feb 2006 22:21:20 -0500 (EST) +Received: from postgresql.org (postgresql.org [200.46.204.71]) + by ams.hub.org (Postfix) with ESMTP id 9747667B7CC + for ; Tue, 28 Feb 2006 23:21:17 -0400 (AST) +X-Original-To: pgsql-bugs-postgresql.org@localhost.postgresql.org +Received: from localhost (av.hub.org [200.46.204.144]) + by postgresql.org (Postfix) with ESMTP id 6FC2C9DC9A1 + for ; Tue, 28 Feb 2006 23:20:52 -0400 (AST) +Received: from postgresql.org ([200.46.204.71]) + by localhost (av.hub.org [200.46.204.144]) (amavisd-new, port 10024) + with ESMTP id 79679-03 + for ; + Tue, 28 Feb 2006 23:20:56 -0400 (AST) +Received: from svr4.postgresql.org (svr4.postgresql.org [66.98.251.159]) + by postgresql.org (Postfix) with ESMTP id CD1939DC846 + for ; Tue, 28 Feb 2006 23:20:49 -0400 (AST) +Received: from cliff.cs.toronto.edu (cliff.cs.toronto.edu [128.100.3.120]) + by svr4.postgresql.org (Postfix) with ESMTP id 656B95AF07C + for ; Wed, 1 Mar 2006 03:20:55 +0000 (GMT) +Received: from eon.cs (eon.cs.toronto.edu [128.100.3.15]) + by cliff.cs.toronto.edu (Postfix) with ESMTP id DE4EB5FD0C; + Tue, 28 Feb 2006 22:20:52 -0500 (EST) +Received: by eon.cs (Postfix, from userid 1300) + id 127BA7BE; Tue, 28 Feb 2006 22:20:53 -0500 (EST) +Received: from localhost (localhost [127.0.0.1]) + by eon.cs (Postfix) with ESMTP + id 0653653C; Tue, 28 Feb 2006 22:20:53 -0500 (EST) +Date: Tue, 28 Feb 2006 22:20:52 -0500 (EST) +From: Qingqing Zhou +X-X-Sender: zhouqq@eon.cs +To: Bruce Momjian +cc: pgsql-bugs@postgresql.org +Subject: Re: [BUGS] FATAL: semctl(1672698088, 12, SETVAL, 0) failed +In-Reply-To: <200603010308.k2138bO09022@candle.pha.pa.us> +Message-ID: +References: <200603010308.k2138bO09022@candle.pha.pa.us> +MIME-Version: 1.0 +Content-Type: TEXT/PLAIN; charset=US-ASCII +X-Virus-Scanned: by amavisd-new at hub.org +X-Spam-Status: No, score=0.442 required=5 tests=[AWL=-0.037, + DNS_FROM_RFC_ABUSE=0.479] +X-Spam-Score: 0.442 +X-Mailing-List: pgsql-bugs +List-Archive: +List-Help: +List-Id: +List-Owner: +List-Post: +List-Subscribe: +List-Unsubscribe: +Precedence: bulk +Sender: pgsql-bugs-owner@postgresql.org +Status: OR + + + +On Tue, 28 Feb 2006, Bruce Momjian wrote: + +> +> Uh, how do we handle it now? I thought we did just that. +> +> OK, so how do we find the answer? +> + +For both problems, I am uncertain (or I've sent a patch already :-(). Call +more artillery support here ... + +Regards, +Qingqing + +---------------------------(end of broadcast)--------------------------- +TIP 5: don't forget to increase your free space map settings +