From owner-freebsd-stable Mon Apr 28 06:11:54 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id GAA13247 for stable-outgoing; Mon, 28 Apr 1997 06:11:54 -0700 (PDT) Received: from who.cdrom.com (who.cdrom.com [204.216.27.3]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id GAA13242 for ; Mon, 28 Apr 1997 06:11:52 -0700 (PDT) Received: from nf.jinr.ru. (root@nfsun1.jinr.ru [159.93.21.4]) by who.cdrom.com (8.8.5/8.6.11) with ESMTP id GAA01473 for ; Mon, 28 Apr 1997 06:11:13 -0700 (PDT) Received: from iren.jinr.ru (ivanov@iren.jinr.dubna.su [159.93.21.34]) by nf.jinr.ru. (8.8.4/8.8.4) with SMTP id RAA07364 for ; Mon, 28 Apr 1997 17:12:53 +0400 (MSK-DST) Message-ID: <3364A21F.167EB0E7@nf.jinr.ru> Date: Mon, 28 Apr 1997 13:11:59 +0000 From: "A.P.Ivanov" Organization: JINR X-Mailer: Mozilla 3.01 (X11; I; FreeBSD 2.2.1-RELEASE i386) MIME-Version: 1.0 To: freebsd-stable@freebsd.org Subject: Apache hangs system Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-stable@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Hi, I'm in deep trouble after upgrading to 2.2.1-RELEASE.(from2.1.5) Now apache sometimes hang the system. No logs, no dumps, machine just freezes so I have to reset it. Besides sometimes (very rare) postgres server behaves strange. I had suspicion that something wrong with SYSVSHM,IPC,SEM because postgres client sometimes couldn't connect to server complaing about SV semaphores. Besides I tested linux's httpd and it worked but complained about the same thing on start. Unfortunately linux's httpd died at last too (and hangs machine also). But it holded on much longer that native one. I tested old apache(1.1.3) with no luck. I just can't think of anything, the only way is either downgrade to 2.1.5 (which was rock solid for half year) or close http. Any suggestions? -- Alexander P. Ivanov mailto:ivanov@nf.jinr.ru FLNP JINR http://iren.jinr.ru/~ivanov From owner-freebsd-stable Mon Apr 28 07:06:03 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id HAA15452 for stable-outgoing; Mon, 28 Apr 1997 07:06:03 -0700 (PDT) Received: from pluto.plutotech.com (root@pluto100.plutotech.com [206.168.67.137]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id HAA15447 for ; Mon, 28 Apr 1997 07:06:02 -0700 (PDT) Received: from narnia.plutotech.com (narnia.plutotech.com [206.168.67.130]) by pluto.plutotech.com (8.8.5/8.8.3) with ESMTP id IAA01694; Mon, 28 Apr 1997 08:05:01 -0600 (MDT) Message-Id: <199704281405.IAA01694@pluto.plutotech.com> To: "A.P.Ivanov" cc: freebsd-stable@freebsd.org Subject: Re: Apache hangs system In-reply-to: Your message of "Mon, 28 Apr 1997 13:11:59 -0000." <3364A21F.167EB0E7@nf.jinr.ru> Date: Mon, 28 Apr 1997 09:03:07 -0600 From: "Justin T. Gibbs" Sender: owner-stable@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Could this be the exec bug, that David fixed??? Have you tried upgrading to the latest 2.2-stable snapshot? You can do this either with CVSup, CTM, or by pulling a full release from releng22.FreeBSD.org. -- Justin T. Gibbs =========================================== FreeBSD: Turning PCs into workstations =========================================== From owner-freebsd-stable Mon Apr 28 08:54:42 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id IAA21173 for stable-outgoing; Mon, 28 Apr 1997 08:54:42 -0700 (PDT) Received: from nf.jinr.ru. (root@nfsun1.jinr.ru [159.93.21.4]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id IAA21071 for ; Mon, 28 Apr 1997 08:52:32 -0700 (PDT) Received: from iren.jinr.ru (ivanov@iren.jinr.dubna.su [159.93.21.34]) by nf.jinr.ru. (8.8.4/8.8.4) with SMTP id TAA11557 for ; Mon, 28 Apr 1997 19:54:04 +0400 (MSK-DST) Message-ID: <3364C7E7.41C67EA6@nf.jinr.ru> Date: Mon, 28 Apr 1997 15:53:11 +0000 From: "A.P.Ivanov" Organization: JINR X-Mailer: Mozilla 3.01 (X11; I; FreeBSD 2.2-STABLE i386) MIME-Version: 1.0 To: freebsd-stable@freebsd.org Subject: Re: Apache hangs system References: <199704281405.IAA01694@pluto.plutotech.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-stable@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Justin T. Gibbs wrote: > > Could this be the exec bug, that David fixed??? Have you tried > upgrading to the latest 2.2-stable snapshot? You can do this No,just CVSupped latest stable,recompiled kernel -> the same shit... -- Alexander P. Ivanov mailto:ivanov@nf.jinr.ru FLNP JINR http://iren.jinr.ru/~ivanov From owner-freebsd-stable Mon Apr 28 09:40:34 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id JAA23454 for stable-outgoing; Mon, 28 Apr 1997 09:40:34 -0700 (PDT) Received: from pluto.plutotech.com (root@pluto100.plutotech.com [206.168.67.137]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id JAA23445 for ; Mon, 28 Apr 1997 09:40:32 -0700 (PDT) Received: from narnia.plutotech.com (narnia.plutotech.com [206.168.67.130]) by pluto.plutotech.com (8.8.5/8.8.3) with ESMTP id KAA04642; Mon, 28 Apr 1997 10:40:18 -0600 (MDT) Message-Id: <199704281640.KAA04642@pluto.plutotech.com> X-Mailer: exmh version 2.0beta 12/23/96 To: "A.P.Ivanov" cc: freebsd-stable@freebsd.org Subject: Re: Apache hangs system In-reply-to: Your message of "Mon, 28 Apr 1997 15:53:11 -0000." <3364C7E7.41C67EA6@nf.jinr.ru> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Mon, 28 Apr 1997 11:38:23 -0600 From: "Justin T. Gibbs" Sender: owner-stable@freebsd.org X-Loop: FreeBSD.org Precedence: bulk >Justin T. Gibbs wrote: >> >> Could this be the exec bug, that David fixed??? Have you tried >> upgrading to the latest 2.2-stable snapshot? You can do this > > No,just CVSupped latest stable,recompiled kernel -> the same shit... But DDB in your kernel and find out what it's hanging on. You can find information on how to use DDB in the handbook on www.FreeBSD.org. >-- > > Alexander P. Ivanov mailto:ivanov@nf.jinr.ru > FLNP JINR http://iren.jinr.ru/~ivanov > -- Justin T. Gibbs =========================================== FreeBSD: Turning PCs into workstations =========================================== From owner-freebsd-stable Mon Apr 28 11:30:33 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id LAA29758 for stable-outgoing; Mon, 28 Apr 1997 11:30:33 -0700 (PDT) Received: from tor-adm1.nbc.netcom.ca (taob@tor-adm1.nbc.netcom.ca [207.181.89.5]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id LAA29750 for ; Mon, 28 Apr 1997 11:30:30 -0700 (PDT) Received: from localhost (taob@localhost) by tor-adm1.nbc.netcom.ca (8.8.5/8.8.5) with SMTP id OAA16342 for ; Mon, 28 Apr 1997 14:29:23 -0400 (EDT) Date: Mon, 28 Apr 1997 14:29:23 -0400 (EDT) From: Brian Tao Reply-To: Brian Tao To: freebsd-stable@freebsd.org Subject: System freezes with 2.2-970420 In-Reply-To: <3364A21F.167EB0E7@nf.jinr.ru> Message-ID: MIME-Version: 1.0 Content-Type: MULTIPART/MIXED; BOUNDARY="-559023410-1483920592-862251942=:12135" Content-ID: Sender: owner-stable@freebsd.org X-Loop: FreeBSD.org Precedence: bulk This message is in MIME format. The first part should be readable text, while the remaining parts are likely unreadable without MIME-aware tools. Send mail to mime@docserver.cac.washington.edu for more info. ---559023410-1483920592-862251942=:12135 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Content-ID: Changing from an old 3.0 snapshot to a recent 2.2-STABLE seems to have fixed the ahc-related problems (it survived a 36-hour burn-in with two 2940UW's and seven 4.3GB drives). However, I've come across two ways to freeze up a 2.2-970420-RELENG system, one of them I was able to reproduce consistently. The first is newfs'ing a disk via the block device rather than the raw device. Doing so freezes up the machine at the same point in the newfs each time. Using the raw device works fine: # newfs /dev/sd2s1a newfs: /dev/sd2s1a: not a character-special device Warning: 560 sector(s) in last cylinder unallocated /dev/sd2s1a: 8482256 sectors in 2071 cylinders of 1 tracks, 4096 sectors 4141.7MB in 130 cyl groups (16 c/g, 32.00MB/g, 7680 i/g) super-block backups (for fsck -b #) at: 32, 65568, 131104, 196640, 262176, 327712, 393248, 458784, 524320, 589856, 655392, 720928, 786464, 852000, 917536, 983072, 1048608, 1114144, 1179680, 1245216, 1310752, 1376288, 1441824, 1507360, 1572896, 1638432, 1703968, 1769504, 1835040, 1900576, 1966112, 2031648, 2097184, 2162720, 2228256, [...machine locks up at this point...] # newfs /dev/rsd2s1a Warning: 560 sector(s) in last cylinder unallocated /dev/rsd2s1a: 8482256 sectors in 2071 cylinders of 1 tracks, 4096 sectors 4141.7MB in 130 cyl groups (16 c/g, 32.00MB/g, 7680 i/g) super-block backups (for fsck -b #) at: 32, 65568, 131104, 196640, 262176, 327712, 393248, 458784, 524320, 589856, 655392, 720928, 786464, 852000, 917536, 983072, 1048608, 1114144, 1179680, 1245216, 1310752, 1376288, 1441824, 1507360, 1572896, 1638432, 1703968, 1769504, 1835040, 1900576, 1966112, 2031648, 2097184, 2162720, 2228256, 2293792, 2359328, 2424864, 2490400, 2555936, 2621472, 2687008, 2752544, 2818080, 2883616, 2949152, 3014688, 3080224, 3145760, 3211296, 3276832, 3342368, 3407904, 3473440, 3538976, 3604512, 3670048, 3735584, 3801120, 3866656, 3932192, 3997728, 4063264, 4128800, 4194336, 4259872, 4325408, 4390944, 4456480, 4522016, 4587552, 4653088, 4718624, 4784160, 4849696, 4915232, 4980768, 5046304, 5111840, 5177376, 5242912, 5308448, 5373984, 5439520, 5505056, 5570592, 5636128, 5701664, 5767200, 5832736, 5898272, 5963808, 6029344, 6094880, 6160416, 6225952, 6291488, 6357024, 6422560, 6488096, 6553632, 6619168, 6684704, 6750240, 6815776, 6881312, 6946848, 7012384, 7077920, 7143456, 7208992, 7274528, 7340064, 7405600, 7471136, 7536672, 7602208, 7667744, 7733280, 7798816, 7864352, 7929888, 7995424, 8060960, 8126496, 8192032, 8257568, 8323104, 8388640, 8454176, # The second happened doing a "quotaon -a", with six quota filesystems, all of which were currently exported and mounted on a BSD/OS 2.1 client. quota.user files were present at the root of each filesystem, although limits for all ~30000 users were set to zero at the time. The server boots with quotas on, but I had shut off all quotas earlier. Turning them back on caused the system to freeze. I haven't tried reproducing this one. In both cases, there were no console or syslog messages indicating a problem, and a hard reboot was necessary. This is a production machine now, so I can't really do a lot of fiddling with it. I'll try the "blessed" 2.2-970422 kernel to see if that makes any difference. I can send boot messages and the kernel config if that will help. AHC_TAGENABLE, AHC_SCBPAGING_ENABLE and the SysV IPC options are enabled. bt and ccd drivers are compiled in, but not currently in active use. -- Brian Tao (BT300, taob@netcom.ca) "Though this be madness, yet there is method in't" ---559023410-1483920592-862251942=:12135-- From owner-freebsd-stable Mon Apr 28 11:58:11 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id LAA01598 for stable-outgoing; Mon, 28 Apr 1997 11:58:11 -0700 (PDT) Received: from implode.root.com (implode.root.com [198.145.90.17]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id LAA01589 for ; Mon, 28 Apr 1997 11:58:06 -0700 (PDT) Received: from localhost (localhost [127.0.0.1]) by implode.root.com (8.8.5/8.8.5) with SMTP id LAA06780; Mon, 28 Apr 1997 11:58:13 -0700 (PDT) Message-Id: <199704281858.LAA06780@implode.root.com> X-Authentication-Warning: implode.root.com: localhost [127.0.0.1] didn't use HELO protocol To: "Justin T. Gibbs" cc: "A.P.Ivanov" , freebsd-stable@FreeBSD.org Subject: Re: Apache hangs system In-reply-to: Your message of "Mon, 28 Apr 1997 09:03:07 MDT." <199704281405.IAA01694@pluto.plutotech.com> From: David Greenman Reply-To: dg@root.com Date: Mon, 28 Apr 1997 11:58:13 -0700 Sender: owner-stable@FreeBSD.org X-Loop: FreeBSD.org Precedence: bulk >Could this be the exec bug, that David fixed??? Have you tried >upgrading to the latest 2.2-stable snapshot? You can do this >either with CVSup, CTM, or by pulling a full release from >releng22.FreeBSD.org. The bug wasn't present in 2.2.1-RELEASE. -DG David Greenman Core-team/Principal Architect, The FreeBSD Project From owner-freebsd-stable Mon Apr 28 21:56:05 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id VAA24814 for stable-outgoing; Mon, 28 Apr 1997 21:56:05 -0700 (PDT) Received: from tor-adm1.nbc.netcom.ca (taob@tor-adm1.nbc.netcom.ca [207.181.89.5]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id VAA24800 for ; Mon, 28 Apr 1997 21:55:56 -0700 (PDT) Received: from localhost (taob@localhost) by tor-adm1.nbc.netcom.ca (8.8.5/8.8.5) with SMTP id AAA00497 for ; Tue, 29 Apr 1997 00:54:53 -0400 (EDT) Date: Tue, 29 Apr 1997 00:54:53 -0400 (EDT) From: Brian Tao To: FREEBSD-STABLE Subject: Memory usage on NFS server Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-stable@freebsd.org X-Loop: FreeBSD.org Precedence: bulk 2.2-970420-RELENG on an NFS server that does only that. The ~20 processes normally running on the machine add up to less than 10MB, yes `top' reports that 78MB is "active". I would have though most of the memory would be allocated as cache (which is only reported at 25MB). Am I interpreting the numbers incorrectly? last pid: 7195; load averages: 0.08, 0.10, 0.15 00:53:42 20 processes: 1 running, 19 sleeping CPU states: 0.0% user, 0.0% nice, 1.2% system, 0.8% interrupt, 98.1% idle Mem: 78M Active, 4984K Inact, 17M Wired, 25M Cache, 8343K Buf, 616K Free Swap: 256M Total, 64K Used, 256M Free PID USERNAME PRI NICE SIZE RES STATE TIME WCPU CPU COMMAND 180 root 2 0 408K 548K accept 0:14 1.37% 1.37% sshd 119 root 2 0 224K 96K nfsd 26:07 1.11% 1.11% nfsd 7195 taob 28 0 312K 852K RUN 0:00 0.47% 0.27% top 2276 taob 18 0 624K 784K pause 0:00 0.00% 0.00% tcsh 146 root 18 0 528K 440K pause 0:00 0.00% 0.00% sendmail 142 root 18 0 344K 280K pause 0:01 0.00% 0.00% cron 1 root 10 0 456K 148K wait 0:00 0.00% 0.00% init 2591 root 3 0 164K 324K ttyin 0:00 0.00% 0.00% getty 184 root 3 0 164K 248K ttyin 0:00 0.00% 0.00% getty 183 root 3 0 164K 248K ttyin 0:00 0.00% 0.00% getty 2274 root 2 0 488K 616K select 0:04 0.00% 0.00% sshd 88 root 2 0 196K 332K select 0:00 0.00% 0.00% syslogd 114 root 2 0 444K 264K select 0:00 0.00% 0.00% mountd 99 daemon 2 0 176K 264K select 0:00 0.00% 0.00% portmap 123 root 2 0 176K 248K select 0:00 0.00% 0.00% rpc.statd 140 root 2 0 216K 216K select 0:00 0.00% 0.00% inetd 117 root 2 0 232K 108K accept 0:00 0.00% 0.00% nfsd 120 root 2 0 224K 96K nfsd 5:34 0.00% 0.00% nfsd 121 root 2 0 224K 96K nfsd 1:16 0.00% 0.00% nfsd 122 root 2 0 224K 96K nfsd 0:18 0.00% 0.00% nfsd -- Brian Tao (BT300, taob@netcom.ca) "Though this be madness, yet there is method in't" From owner-freebsd-stable Tue Apr 29 00:18:37 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id AAA01258 for stable-outgoing; Tue, 29 Apr 1997 00:18:37 -0700 (PDT) Received: from implode.root.com (implode.root.com [198.145.90.17]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id AAA01251 for ; Tue, 29 Apr 1997 00:18:34 -0700 (PDT) Received: from localhost (localhost [127.0.0.1]) by implode.root.com (8.8.5/8.8.5) with SMTP id AAA11888; Tue, 29 Apr 1997 00:20:09 -0700 (PDT) Message-Id: <199704290720.AAA11888@implode.root.com> X-Authentication-Warning: implode.root.com: localhost [127.0.0.1] didn't use HELO protocol To: Brian Tao cc: FREEBSD-STABLE Subject: Re: Memory usage on NFS server In-reply-to: Your message of "Tue, 29 Apr 1997 00:54:53 EDT." From: David Greenman Reply-To: dg@root.com Date: Tue, 29 Apr 1997 00:20:09 -0700 Sender: owner-stable@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > 2.2-970420-RELENG on an NFS server that does only that. The ~20 >processes normally running on the machine add up to less than 10MB, >yes `top' reports that 78MB is "active". I would have though most of >the memory would be allocated as cache (which is only reported at >25MB). Am I interpreting the numbers incorrectly? ... >Mem: 78M Active, 4984K Inact, 17M Wired, 25M Cache, 8343K Buf, 616K Free The numbers lie; don't believe them. :-) It's not as simple as looking at the "cache" number - pages in "active" and "inactive" are also part of the file cache. I know this makes it very difficult to see how much memory is actually available for caching...I have the same problem on wcarchive. We need to provide an additional metric, but I don't know at the moment how to create the desired information (it's difficult and perhaps impossible in the current architecture). -DG David Greenman Core-team/Principal Architect, The FreeBSD Project From owner-freebsd-stable Tue Apr 29 01:06:31 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id BAA03532 for stable-outgoing; Tue, 29 Apr 1997 01:06:31 -0700 (PDT) Received: from zibbi.mikom.csir.co.za (zibbi.mikom.csir.co.za [146.64.24.58]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id BAA03526 for ; Tue, 29 Apr 1997 01:06:14 -0700 (PDT) Received: (from jhay@localhost) by zibbi.mikom.csir.co.za (8.8.5/8.8.5) id KAA18509; Tue, 29 Apr 1997 10:03:57 +0200 (SAT) From: John Hay Message-Id: <199704290803.KAA18509@zibbi.mikom.csir.co.za> Subject: Re: Memory usage on NFS server In-Reply-To: <199704290720.AAA11888@implode.root.com> from David Greenman at "Apr 29, 97 00:20:09 am" To: dg@root.com Date: Tue, 29 Apr 1997 10:03:57 +0200 (SAT) Cc: freebsd-stable@freebsd.org X-Mailer: ELM [version 2.4ME+ PL31 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-stable@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > > 2.2-970420-RELENG on an NFS server that does only that. The ~20 > >processes normally running on the machine add up to less than 10MB, > >yes `top' reports that 78MB is "active". I would have though most of > >the memory would be allocated as cache (which is only reported at > >25MB). Am I interpreting the numbers incorrectly? > ... > >Mem: 78M Active, 4984K Inact, 17M Wired, 25M Cache, 8343K Buf, 616K Free > > The numbers lie; don't believe them. :-) It's not as simple as looking > at the "cache" number - pages in "active" and "inactive" are also part of > the file cache. I know this makes it very difficult to see how much memory > is actually available for caching...I have the same problem on wcarchive. > We need to provide an additional metric, but I don't know at the moment > how to create the desired information (it's difficult and perhaps impossible > in the current architecture). > Do the numbers that top and ps show for per process memory usage also lie? What I see here on my news server, is that I run out of swap (256M), but according to top and ps a rough calculation of the total of all the processes is less than half that. Inn then typically show a usage of ~70M according to top, but as soon as I kill and restart it the swap usage go down to ~5M. I once even added a 128M vn swapfile and it filled that also without inn showing a usage of more than 70M, but killing and restarting it takes the swap usage down to ~5M. John -- John Hay -- John.Hay@mikom.csir.co.za From owner-freebsd-stable Tue Apr 29 01:50:03 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id BAA05638 for stable-outgoing; Tue, 29 Apr 1997 01:50:03 -0700 (PDT) Received: from relay.nuxi.com (nuxi.ucdavis.edu [128.120.175.23]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id BAA05609; Tue, 29 Apr 1997 01:49:54 -0700 (PDT) Received: from dragon.nuxi.com (dav1-6.calweb.com [207.211.82.6]) by relay.nuxi.com (8.8.5/8.6.12) with ESMTP id BAA16176; Tue, 29 Apr 1997 01:03:07 GMT Received: (from obrien@localhost) by dragon.nuxi.com (8.8.5/8.7.3) id IAA01252; Tue, 29 Apr 1997 08:49:45 GMT Message-ID: <19970429014944.36331@dragon.nuxi.com> Date: Tue, 29 Apr 1997 01:49:44 -0700 From: "David O'Brien" To: gibbs@freebsd.org Cc: stable@freebsd.org Subject: AHA-2940U problem in -STABLE Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.69 X-Warning: Mutt Bites! X-Operating-System: FreeBSD 2.2-STABLE Organization: The NUXI *BSD group X-PGP-Fingerprint: B7 4D 3E E9 11 39 5F A3 90 76 5D 69 58 D9 98 7A X-Pgp-Keyid: 34F9F9D5 Sender: owner-stable@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Hi Justin, I'm having a little problem with my AHA-2940. I can no longer use my Exbyte 8505 8mm drive to back up my SCSI hard drive. Using a 2.2-STABLE kernel in single user mode from 4/24 I got: DUMP: 91.33% done, finished in 0:03 st0(ahc0:4:0): SCB 0x2 - timeed out while idle, LASTPHASE == 0x1, SCSISIGI == 0x0 SEQADDR == 0x8 st0(ahc0:4:0): Queueing an Abort SCB st0(ahc0:4:0): Abort Mesasge sent st0(ahc0:4:0): SCB 2 - Abort Completed st0(ahc0:4:0): no longer in timeout DUMP: End of tape detected sd0(ahc0:0:0): SCB 0x1 - timed out in message out phase, SCSISIGI == 0x0 SEQADDR == 0x0 sd0(ahc0:0:0): SCB 1: Yucky Immediate reset. Flags = 0x1 sd0(ahc0:0:0): no longer in timeout sd0(ahc0:0:0): Issued Channel A Bus Reset. 3 SCBs aborted. And then the machine froze (not even a panic). Note, that Dump shouldn't have detected EOT as I was dumping a 2gig partition on a tape that should hold 5gig minium. I've used this tape to backup the SCSI drive before, even while multiuser. So I think it has something to do with an aic7xxx change. Maybe? -- -- David (obrien@NUXI.com -or- obrien@FreeBSD.org) From owner-freebsd-stable Tue Apr 29 02:55:51 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id CAA08174 for stable-outgoing; Tue, 29 Apr 1997 02:55:51 -0700 (PDT) Received: from gate.ticl.co.uk (gate.ticl.co.uk [193.32.1.5]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id CAA08169 for ; Tue, 29 Apr 1997 02:55:48 -0700 (PDT) Received: (from peter@localhost) by gate.ticl.co.uk (8.8.5/8.8.5) id KAA11005 for stable@freebsd.org; Tue, 29 Apr 1997 10:53:58 +0100 (BST) From: Peter Curran Message-Id: <199704290953.KAA11005@gate.ticl.co.uk> Subject: Funny ARP problem To: stable@freebsd.org Date: Tue, 29 Apr 1997 10:53:58 +0100 (BST) X-Mailer: ELM [version 2.4ME+ PL31 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-stable@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Hi guys I got a funny problem with ARP on a 2.1.7-R box. I have an ethernet interface (3com 3c509) which has two IP addresses, one the primary is (e.g.) 200.1.1.1 and the alias is 200.1.1.1. If I ping another box on the 100.1.1 net, say 100.1.1.2, then I get no reply. If I ping on the 200.1.1 net I get a reply. Putting an analyser on the net I can see that if I ping 100.1.1.2 then the arp request is FROM 200.1.1.1 -> 100.1.1.2. If I ping 200.1.1.2 then the arp message is correct. I have never observed this behaviour before and normally have multiple addresses bound to a single interface from different nets. I have only very recently upgraded this box to 2.1.7 and the extra address is new so I am not sure what has caused the breakage. Am I missing something really obvious here? Any help or suggestions gratefully received. Cheers Peter From owner-freebsd-stable Tue Apr 29 07:27:26 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id HAA18450 for stable-outgoing; Tue, 29 Apr 1997 07:27:26 -0700 (PDT) Received: from tor-adm1.nbc.netcom.ca (taob@tor-adm1.nbc.netcom.ca [207.181.89.5]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id HAA18445 for ; Tue, 29 Apr 1997 07:27:24 -0700 (PDT) Received: from localhost (taob@localhost) by tor-adm1.nbc.netcom.ca (8.8.5/8.8.5) with SMTP id KAA06836; Tue, 29 Apr 1997 10:26:19 -0400 (EDT) Date: Tue, 29 Apr 1997 10:26:19 -0400 (EDT) From: Brian Tao To: David Greenman cc: FREEBSD-STABLE Subject: Re: Memory usage on NFS server In-Reply-To: <199704290720.AAA11888@implode.root.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-stable@freebsd.org X-Loop: FreeBSD.org Precedence: bulk On Tue, 29 Apr 1997, David Greenman wrote: > ... > >Mem: 78M Active, 4984K Inact, 17M Wired, 25M Cache, 8343K Buf, 616K Free > > The numbers lie; don't believe them. :-) It's not as simple as > looking at the "cache" number - pages in "active" and "inactive" are > also part of the file cache. But in the end, FreeBSD *will* try to use as much physical memory as possible as a file cache, regardless of what `top' reports? The owner of the machine thought 128MB was overkill for just an NFS server, but I assured him all of it would be put to good use. ;-) -- Brian Tao (BT300, taob@netcom.ca) "Though this be madness, yet there is method in't" From owner-freebsd-stable Tue Apr 29 08:39:05 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id IAA21423 for stable-outgoing; Tue, 29 Apr 1997 08:39:05 -0700 (PDT) Received: from pluto.plutotech.com (root@pluto100.plutotech.com [206.168.67.137]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id IAA21418; Tue, 29 Apr 1997 08:38:59 -0700 (PDT) Received: from narnia.plutotech.com (narnia.plutotech.com [206.168.67.130]) by pluto.plutotech.com (8.8.5/8.8.3) with ESMTP id JAA27224; Tue, 29 Apr 1997 09:38:47 -0600 (MDT) Message-Id: <199704291538.JAA27224@pluto.plutotech.com> X-Mailer: exmh version 2.0beta 12/23/96 To: "David O'Brien" cc: gibbs@freebsd.org, stable@freebsd.org Subject: Re: AHA-2940U problem in -STABLE In-reply-to: Your message of "Tue, 29 Apr 1997 01:49:44 PDT." <19970429014944.36331@dragon.nuxi.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Tue, 29 Apr 1997 10:37:15 -0600 From: "Justin T. Gibbs" Sender: owner-stable@freebsd.org X-Loop: FreeBSD.org Precedence: bulk >Hi Justin, > >I'm having a little problem with my AHA-2940. I can no longer use my >Exbyte 8505 8mm drive to back up my SCSI hard drive. Using a 2.2-STABLE >kernel in single user mode from 4/24 I got: > > >DUMP: 91.33% done, finished in 0:03 >st0(ahc0:4:0): SCB 0x2 - timeed out while idle, LASTPHASE == 0x1, SCSISIGI == >0x0 >SEQADDR == 0x8 My guess is that you are using the rewindind device and that the rewind is taking longer than the st driver allows (5 minutes I think) causing the initial timeout. This is the first problem. >st0(ahc0:4:0): Queueing an Abort SCB >st0(ahc0:4:0): Abort Mesasge sent >st0(ahc0:4:0): SCB 2 - Abort Completed >st0(ahc0:4:0): no longer in timeout >DUMP: End of tape detected >sd0(ahc0:0:0): SCB 0x1 - timed out in message out phase, SCSISIGI == 0x0 >SEQADDR == 0x0 >sd0(ahc0:0:0): SCB 1: Yucky Immediate reset. Flags = 0x1 >sd0(ahc0:0:0): no longer in timeout >sd0(ahc0:0:0): Issued Channel A Bus Reset. 3 SCBs aborted. > > >And then the machine froze (not even a panic). This is bug #2. For some reason, the recovery code is botching it. Can you try the attached patch and see if the recovery code behaves correctly? Once I get the recovery code to work for you, you should be able to bump up the timeout in scsi/st.c:st_rewind and make the problem completly go away. >Note, that Dump shouldn't >have detected EOT as I was dumping a 2gig partition on a tape that should >hold 5gig minium. I've used this tape to backup the SCSI drive before, >even while multiuser. So I think it has something to do with an aic7xxx >change. Maybe? My guess is that you have more data to dump now, making the rewind take longer, which is why you are just now running into this problem. >-- >-- David (obrien@NUXI.com -or- obrien@FreeBSD.org) -- Justin T. Gibbs =========================================== FreeBSD: Turning PCs into workstations =========================================== Index: i386/scsi/aic7xxx.c =================================================================== RCS file: /usr/cvs/src/sys/i386/scsi/aic7xxx.c,v retrieving revision 1.118 diff -c -r1.118 aic7xxx.c *** aic7xxx.c 1997/04/26 05:03:18 1.118 --- aic7xxx.c 1997/04/29 16:31:33 *************** *** 178,191 **** int unpause_always)); static inline void restart_sequencer __P((struct ahc_softc *ahc)); - static int timeouts_work; /* - * Boolean that lets the driver know - * that it can safely use timeouts - * for event scheduling. Timeouts - * don't work early in the boot - * process. - */ - #define AHC_BUSRESET_DELAY 1000 /* Reset delay in us */ #define AHC_BUSSETTLE_DELAY (100 * 1000)/* Delay before taking commands * after a reset in us --- 178,183 ---- *************** *** 308,314 **** static void ahc_clear_intstat __P((struct ahc_softc *ahc)); static void ahc_reset_current_bus __P((struct ahc_softc *ahc)); static void ahc_run_done_queue __P((struct ahc_softc *ahc)); - static void ahc_untimeout_done_queue __P((struct ahc_softc *ahc)); static void ahc_scsirate __P((struct ahc_softc* ahc, u_int8_t *scsirate, u_int8_t *period, u_int8_t *offset, char channel, int target)); --- 300,305 ---- *************** *** 319,333 **** static void ahc_timeout __P((void *)); #endif - static timeout_t - ahc_first_timeout; - - static timeout_t - ahc_busreset_complete; - - static timeout_t - ahc_busreset_settle_complete; - static u_int8_t ahc_index_busy_target __P((struct ahc_softc *ahc, int target, char channel, int unbusy)); --- 310,315 ---- *************** *** 489,500 **** ahc->flags = flags; ahc->unpause = (ahc_inb(ahc, HCNTRL) & IRQMS) | INTEN; ahc->pause = ahc->unpause | PAUSE; - - ahc->busreset_args.ahc = ahc; - ahc->busreset_args.bus = 'A'; - ahc->busreset_args_b.ahc = ahc; - ahc->busreset_args_b.bus = 'B'; - #if defined(__FreeBSD__) return (ahc); #endif --- 471,476 ---- *************** *** 2326,2334 **** unpause_sequencer(ahc, /*unpause_always*/TRUE); - /* Notify us when the system can handle timeouts */ - timeout(ahc_first_timeout, NULL, 0); - /* * Note that we are going and return (to probe) */ --- 2302,2307 ---- *************** *** 2377,2399 **** SC_DEBUG(xs->sc_link, SDEV_DB2, ("ahc_scsi_cmd\n")); flags = xs->flags; - if (ahc->in_reset != 0) { - /* - * If we are still in a reset, send the transaction - * back indicating that it was aborted due to a - * reset. - */ - if ((IS_SCSIBUS_B(ahc, xs->sc_link) - && (ahc->in_reset & CHANNEL_B_RESET) != 0) - || (!IS_SCSIBUS_B(ahc, xs->sc_link) - && (ahc->in_reset & CHANNEL_A_RESET) != 0)) { - /* Ick, but I don't want it to abort this */ - xs->retries++; - xs->error = XS_BUSY; - return(COMPLETE); - } - } - /* * get an scb to use. If the transfer * is from a buf (possibly from interrupt time) --- 2350,2355 ---- *************** *** 2893,2911 **** return (0); } - /* - * Handler to register that the system is capable of - * delivering timeouts. We register this timeout in - * ahc_init and it should go off as soon as we're through - * the boot process. - */ - static void - ahc_first_timeout(arg) - void *arg; - { - timeouts_work = 1; - } - static void ahc_timeout(arg) void *arg; --- 2849,2854 ---- *************** *** 3654,3714 **** ahc_outb(ahc, SIMODE1, ahc_inb(ahc, SIMODE1) & ~ENSCSIRST); scsiseq = ahc_inb(ahc, SCSISEQ); ahc_outb(ahc, SCSISEQ, scsiseq | SCSIRSTO); ! if (timeouts_work != 0) { ! struct ahc_busreset_args *args; ! ! sblkctl = ahc_inb(ahc, SBLKCTL); ! args = (sblkctl & SELBUSB) ? &ahc->busreset_args_b ! : &ahc->busreset_args; ! ahc->in_reset |= (args->bus == 'A' ? CHANNEL_A_RESET ! : CHANNEL_B_RESET); ! timeout(ahc_busreset_complete, args, ! MAX((AHC_BUSRESET_DELAY * hz) / 1000000, 1)); ! } else { ! DELAY(AHC_BUSRESET_DELAY); ! /* Turn off the bus reset */ ! ahc_outb(ahc, SCSISEQ, scsiseq & ~SCSIRSTO); ! ! ahc_clear_intstat(ahc); ! ! /* Re-enable reset interrupts */ ! ahc_outb(ahc, SIMODE1, ahc_inb(ahc, SIMODE1) | ENSCSIRST); ! ! /* Wait a minimal bus settle delay */ ! DELAY(AHC_BUSSETTLE_DELAY); ! ! ahc_run_done_queue(ahc); ! } ! } ! ! static void ! ahc_busreset_complete(arg) ! void *arg; ! { ! struct ahc_busreset_args *args; ! struct ahc_softc *ahc; ! int s; ! u_int8_t scsiseq; ! u_int8_t sblkctl; ! ! args = (struct ahc_busreset_args *)arg; ! ahc = (struct ahc_softc *)args->ahc; ! ! s = splbio(); ! pause_sequencer(ahc); ! ! /* Save the current block control state */ ! sblkctl = ahc_inb(ahc, SBLKCTL); ! ! /* Switch to the right bus */ ! if (args->bus == 'A') ! ahc_outb(ahc, SBLKCTL, sblkctl & ~SELBUSB); ! else ! ahc_outb(ahc, SBLKCTL, sblkctl | SELBUSB); ! /* Turn off the bus reset */ - printf("Clearing bus reset\n"); - scsiseq = ahc_inb(ahc, SCSISEQ); ahc_outb(ahc, SCSISEQ, scsiseq & ~SCSIRSTO); ahc_clear_intstat(ahc); --- 3597,3604 ---- ahc_outb(ahc, SIMODE1, ahc_inb(ahc, SIMODE1) & ~ENSCSIRST); scsiseq = ahc_inb(ahc, SCSISEQ); ahc_outb(ahc, SCSISEQ, scsiseq | SCSIRSTO); ! DELAY(AHC_BUSRESET_DELAY); /* Turn off the bus reset */ ahc_outb(ahc, SCSISEQ, scsiseq & ~SCSIRSTO); ahc_clear_intstat(ahc); *************** *** 3716,3752 **** /* Re-enable reset interrupts */ ahc_outb(ahc, SIMODE1, ahc_inb(ahc, SIMODE1) | ENSCSIRST); ! /* ! * Set a 100ms timeout to clear our in_reset flag. ! * It will be *at-least* that long before any targets ! * respond to real commands on the bus. ! */ ! timeout(ahc_busreset_settle_complete, args, ! (AHC_BUSSETTLE_DELAY * hz) / 1000000); ! ! /* Restore the original channel */ ! ahc_outb(ahc, SBLKCTL, sblkctl); ! unpause_sequencer(ahc, /*unpause_always*/FALSE); ! splx(s); ! } ! ! static void ! ahc_busreset_settle_complete(arg) ! void *arg; ! { ! struct ahc_busreset_args *args; ! struct ahc_softc *ahc; ! int s; ! ! args = (struct ahc_busreset_args *)arg; ! ahc = (struct ahc_softc *)args->ahc; ! /* Clear the reset flag */ ! s = splbio(); ! printf("Clearing 'in-reset' flag\n"); ! ahc->in_reset &= (args->bus == 'A' ? ~CHANNEL_A_RESET ! : ~CHANNEL_B_RESET); ! ahc_run_done_queue(ahc); ! splx(s); } static int --- 3606,3613 ---- /* Re-enable reset interrupts */ ahc_outb(ahc, SIMODE1, ahc_inb(ahc, SIMODE1) | ENSCSIRST); ! /* Wait a minimal bus settle delay */ ! DELAY(AHC_BUSSETTLE_DELAY); } static int *************** *** 3831,3848 **** ahc_clear_intstat(ahc); restart_sequencer(ahc); } ! /* ! * Untimeout our scbs now in case we have to delay our done ! * processing. ! */ ! ahc_untimeout_done_queue(ahc); ! if (initiate_reset == 0) { ! /* ! * If we initiated the reset, we'll run the queue ! * once our bus-settle delay has expired. ! */ ! ahc_run_done_queue(ahc); ! } return found; } --- 3692,3698 ---- ahc_clear_intstat(ahc); restart_sequencer(ahc); } ! ahc_run_done_queue(ahc); return found; } *************** *** 3860,3879 **** } } - static void - ahc_untimeout_done_queue(ahc) - struct ahc_softc *ahc; - { - int i; - struct scb *scbp; - - for (i = 0; i < ahc->scb_data->numscbs; i++) { - scbp = ahc->scb_data->scbarray[i]; - if (scbp->flags & SCB_QUEUED_FOR_DONE) - untimeout(ahc_timeout, (caddr_t)scbp); - } - } - static int ahc_match_scb (scb, target, channel, lun, tag) struct scb *scb; --- 3710,3715 ---- Index: i386/scsi/aic7xxx.h =================================================================== RCS file: /usr/cvs/src/sys/i386/scsi/aic7xxx.h,v retrieving revision 1.40 diff -c -r1.40 aic7xxx.h *** aic7xxx.h 1997/02/25 03:05:35 1.40 --- aic7xxx.h 1997/04/29 16:28:40 *************** *** 228,238 **** */ }; - struct ahc_busreset_args { - struct ahc_softc *ahc; - char bus; - }; - struct ahc_softc { #if defined(__FreeBSD__) int unit; --- 228,233 ---- *************** *** 249,256 **** #endif volatile u_int8_t *maddr; struct scb_data *scb_data; - struct ahc_busreset_args busreset_args; - struct ahc_busreset_args busreset_args_b; struct scsi_link sc_link; struct scsi_link sc_link_b; /* Second bus for Twin channel cards */ STAILQ_HEAD(, scb) waiting_scbs;/* --- 244,249 ---- *************** *** 289,297 **** u_int8_t unpause; u_int8_t pause; u_int8_t in_timeout; - u_int8_t in_reset; - #define CHANNEL_A_RESET 0x01 - #define CHANNEL_B_RESET 0x02 }; struct full_ahc_softc { --- 282,287 ---- From owner-freebsd-stable Tue Apr 29 09:28:04 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id JAA24021 for stable-outgoing; Tue, 29 Apr 1997 09:28:04 -0700 (PDT) Received: from pahtoh.cwu.edu (root@pahtoh.cwu.edu [198.104.65.27]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id JAA23988 for ; Tue, 29 Apr 1997 09:27:57 -0700 (PDT) Received: from opus.cts.cwu.edu (skynyrd@opus.cts.cwu.edu [198.104.92.71]) by pahtoh.cwu.edu (8.8.5/8.8.5) with ESMTP id JAA25385; Tue, 29 Apr 1997 09:27:53 -0700 (PDT) Received: from localhost (skynyrd@localhost) by opus.cts.cwu.edu (8.8.5/8.8.5) with SMTP id JAA06658; Tue, 29 Apr 1997 09:27:53 -0700 (PDT) Date: Tue, 29 Apr 1997 09:27:53 -0700 (PDT) From: Chris Timmons To: Brian Tao cc: freebsd-stable@freebsd.org Subject: Re: System freezes with 2.2-970420 In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-stable@freebsd.org X-Loop: FreeBSD.org Precedence: bulk I've seen the newfs'ing-the-block-device situation in recent RELENG_2_2 kernels and even 3.0-current. On the 2.2 machine I hit ^C - and at about that time everything hung. I didn't have physical access to the machine at the time so in desperation I ssh'd it a 'reboot' just in case. Nothing happened at first, and I then noticed that even ping requests of the machine were failing. But before the ssh command timed out the machine fully recovered... and then rebooted :( -Chris On Mon, 28 Apr 1997, Brian Tao wrote: > Changing from an old 3.0 snapshot to a recent 2.2-STABLE seems to > have fixed the ahc-related problems (it survived a 36-hour burn-in > with two 2940UW's and seven 4.3GB drives). However, I've come across > two ways to freeze up a 2.2-970420-RELENG system, one of them I was > able to reproduce consistently. > > The first is newfs'ing a disk via the block device rather than the > raw device. Doing so freezes up the machine at the same point in the > newfs each time. Using the raw device works fine: [...] From owner-freebsd-stable Tue Apr 29 09:28:28 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id JAA24130 for stable-outgoing; Tue, 29 Apr 1997 09:28:28 -0700 (PDT) Received: from relay.nuxi.com (nuxi.ucdavis.edu [128.120.175.23]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id JAA24116 for ; Tue, 29 Apr 1997 09:28:25 -0700 (PDT) Received: from dragon.nuxi.com (reqf-013.ucdavis.edu [128.120.253.133]) by relay.nuxi.com (8.8.5/8.6.12) with ESMTP id IAA17405; Tue, 29 Apr 1997 08:41:47 GMT Received: (from obrien@localhost) by dragon.nuxi.com (8.8.5/8.7.3) id QAA02242; Tue, 29 Apr 1997 16:28:17 GMT Message-ID: <19970429092817.18346@dragon.nuxi.com> Date: Tue, 29 Apr 1997 09:28:17 -0700 From: "David O'Brien" To: "Justin T. Gibbs" Cc: stable@freebsd.org Subject: Re: AHA-2940U problem in -STABLE References: <19970429014944.36331@dragon.nuxi.com> <199704291538.JAA27224@pluto.plutotech.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.69 In-Reply-To: <199704291538.JAA27224@pluto.plutotech.com>; from Justin T. Gibbs on Tue, Apr 29, 1997 at 10:37:15AM -0600 X-Warning: Mutt Bites! X-Operating-System: FreeBSD 2.2-STABLE Organization: The NUXI *BSD group X-PGP-Fingerprint: B7 4D 3E E9 11 39 5F A3 90 76 5D 69 58 D9 98 7A X-Pgp-Keyid: 34F9F9D5 Sender: owner-stable@freebsd.org X-Loop: FreeBSD.org Precedence: bulk On Tue, Apr 29, 1997 at 10:37:15AM -0600, Justin T. Gibbs wrote: > >I'm having a little problem with my AHA-2940. I can no longer use my > >Exbyte 8505 8mm drive to back up my SCSI hard drive. Using a 2.2-STABLE > >kernel in single user mode from 4/24 I got: > > > >DUMP: 91.33% done, finished in 0:03 > >st0(ahc0:4:0): SCB 0x2 - timeed out while idle, LASTPHASE == 0x1, SCSISIGI == > >0x0 > >SEQADDR == 0x8 > > My guess is that you are using the rewindind device and that the rewind is Nope. The exact command was "dump 0udsf 54000 13000 /dev/nrst0 /files". > >And then the machine froze (not even a panic). > > This is bug #2. For some reason, the recovery code is botching it. Can > you try the attached patch and see if the recovery code behaves correctly? > Once I get the recovery code to work for you, you should be able to bump > up the timeout in scsi/st.c:st_rewind and make the problem completly go > away. Will try it. BTW, lastnight, I CVSup'ed src-sys and got a revision 1.81.2.25 of 1997/04/26. So I will obviously be patching agaist that. I'll make the new kernel now, and test the dump later today (trying to get an assignemnt done now (isn't school a pain? really gets in the way of FreeBSD hacking :-)))) -- -- David (obrien@NUXI.com -or- obrien@FreeBSD.org) From owner-freebsd-stable Tue Apr 29 09:33:34 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id JAA24532 for stable-outgoing; Tue, 29 Apr 1997 09:33:34 -0700 (PDT) Received: from pluto.plutotech.com (root@pluto100.plutotech.com [206.168.67.137]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id JAA24523 for ; Tue, 29 Apr 1997 09:33:31 -0700 (PDT) Received: from narnia.plutotech.com (narnia.plutotech.com [206.168.67.130]) by pluto.plutotech.com (8.8.5/8.8.3) with ESMTP id KAA28926; Tue, 29 Apr 1997 10:33:29 -0600 (MDT) Message-Id: <199704291633.KAA28926@pluto.plutotech.com> X-Mailer: exmh version 2.0beta 12/23/96 To: "David O'Brien" cc: "Justin T. Gibbs" , stable@freebsd.org Subject: Re: AHA-2940U problem in -STABLE In-reply-to: Your message of "Tue, 29 Apr 1997 09:28:17 PDT." <19970429092817.18346@dragon.nuxi.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Tue, 29 Apr 1997 11:31:57 -0600 From: "Justin T. Gibbs" Sender: owner-stable@freebsd.org X-Loop: FreeBSD.org Precedence: bulk >Nope. The exact command was "dump 0udsf 54000 13000 /dev/nrst0 /files". Use the 'a' option. It's much easier. Okay, if it wasn't a rewind option, then the drive hit a bad tape spot and took too long to write the block. The drive waits 100seconds which seems like more than enough time, but who knows. When was the last time you cleaned the heads on your drive? -- Justin T. Gibbs =========================================== FreeBSD: Turning PCs into workstations =========================================== From owner-freebsd-stable Tue Apr 29 10:25:35 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id KAA27299 for stable-outgoing; Tue, 29 Apr 1997 10:25:35 -0700 (PDT) Received: from sierra.zyzzyva.com (ppp01-58.zyzzyva.com [208.214.58.10]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id KAA27291 for ; Tue, 29 Apr 1997 10:25:30 -0700 (PDT) Received: from sierra (localhost [127.0.0.1]) by sierra.zyzzyva.com (8.8.5/8.8.2) with ESMTP id MAA00757; Tue, 29 Apr 1997 12:24:18 -0500 (CDT) Message-Id: <199704291724.MAA00757@sierra.zyzzyva.com> X-Mailer: exmh version 2.0gamma 1/27/96 To: John Hay cc: dg@root.com, freebsd-stable@freebsd.org Subject: Re: Memory usage on NFS server In-reply-to: jhay's message of Sat, 29 Apr 1997 10:03:57 +0200. <199704290803.KAA18509@zibbi.mikom.csir.co.za> X-uri: http://www.zyzzyva.com/ Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Tue, 29 Apr 1997 12:24:18 -0500 From: Randy Terbush Sender: owner-stable@freebsd.org X-Loop: FreeBSD.org Precedence: bulk For the record, I've seen the same behavior on a machine whose primary job is webservices. This is both for 2.1.7 and 2.2.1. The swap will be freed by killing and restarting the webservers. The machine in question has 64MB RAM and 150MB swap. The webserver is Apache. > Do the numbers that top and ps show for per process memory usage also lie? > What I see here on my news server, is that I run out of swap (256M), but > according to top and ps a rough calculation of the total of all the > processes is less than half that. Inn then typically show a usage of > ~70M according to top, but as soon as I kill and restart it the swap usage > go down to ~5M. I once even added a 128M vn swapfile and it filled that > also without inn showing a usage of more than 70M, but killing and > restarting it takes the swap usage down to ~5M. > > John > -- > John Hay -- John.Hay@mikom.csir.co.za From owner-freebsd-stable Tue Apr 29 10:33:14 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id KAA27904 for stable-outgoing; Tue, 29 Apr 1997 10:33:14 -0700 (PDT) Received: from tor-adm1.nbc.netcom.ca (taob@tor-adm1.nbc.netcom.ca [207.181.89.5]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id KAA27897 for ; Tue, 29 Apr 1997 10:33:09 -0700 (PDT) Received: from localhost (taob@localhost) by tor-adm1.nbc.netcom.ca (8.8.5/8.8.5) with SMTP id NAA08727; Tue, 29 Apr 1997 13:28:56 -0400 (EDT) Date: Tue, 29 Apr 1997 13:28:56 -0400 (EDT) From: Brian Tao To: Chris Timmons cc: freebsd-stable@freebsd.org Subject: Re: System freezes with 2.2-970420 In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-stable@freebsd.org X-Loop: FreeBSD.org Precedence: bulk On Tue, 29 Apr 1997, Chris Timmons wrote: > > But before the ssh command timed out the machine fully recovered... > and then rebooted :( In my case, the server required a physical reset each time. It was pingable, but everything else was locked up. No console switching, no keyboard LED activity, no keypresses echoed. -- Brian Tao (BT300, taob@netcom.ca) "Though this be madness, yet there is method in't" From owner-freebsd-stable Tue Apr 29 10:57:06 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id KAA29157 for stable-outgoing; Tue, 29 Apr 1997 10:57:06 -0700 (PDT) Received: from draw.ubet.com (draw.ubet.com [206.79.156.2]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id KAA29151 for ; Tue, 29 Apr 1997 10:57:02 -0700 (PDT) Received: from roulette.ubet.com (roulette.ubet.com [10.10.10.11]) by draw.ubet.com with SMTP (8.7.5/8.7.3) id KAA25090 for ; Tue, 29 Apr 1997 10:55:06 -0700 (PDT) Received: by roulette.ubet.com with SMTP (Microsoft Exchange Server Internet Mail Connector Version 4.0.993.5) id <01BC548C.4350BA90@roulette.ubet.com>; Tue, 29 Apr 1997 10:58:26 -0700 Message-ID: From: Sam Wu To: "'freebsd-stable@FreeBSD.org'" Subject: remove Date: Tue, 29 Apr 1997 10:58:23 -0700 X-Mailer: Microsoft Exchange Server Internet Mail Connector Version 4.0.993.5 MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: owner-stable@FreeBSD.org X-Loop: FreeBSD.org Precedence: bulk Remove saminfo@ubet.com From owner-freebsd-stable Tue Apr 29 11:47:28 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id LAA02427 for stable-outgoing; Tue, 29 Apr 1997 11:47:28 -0700 (PDT) Received: from mercury.uniserve.com (mercury.uniserve.com [204.191.197.248]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id LAA02417 for ; Tue, 29 Apr 1997 11:47:25 -0700 (PDT) Received: from haven.uniserve.com (shell.uniserve.com [198.53.215.121]) by mercury.uniserve.com with SMTP id LAA15339; Tue, 29 Apr 1997 11:46:04 -0700 (PDT) Date: Tue, 29 Apr 1997 11:52:18 -0700 (PDT) From: Tom Samplonius To: Randy Terbush cc: John Hay , dg@root.com, freebsd-stable@FreeBSD.org Subject: Re: Memory usage on NFS server In-Reply-To: <199704291724.MAA00757@sierra.zyzzyva.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-stable@FreeBSD.org X-Loop: FreeBSD.org Precedence: bulk On Tue, 29 Apr 1997, Randy Terbush wrote: > > For the record, I've seen the same behavior on a machine whose > primary job is webservices. This is both for 2.1.7 and 2.2.1. > The swap will be freed by killing and restarting the webservers. > > The machine in question has 64MB RAM and 150MB swap. > > The webserver is Apache. Both inn and apache used mmap'ed memory. When a process mmaps a shared file, to what process is the memory usage associated with? It could be what you are seeing here is a mmap() leak in these applications, and that mmap'ed memory is not associated with any process. Which version of apache? Which version of inn? Tom > > Do the numbers that top and ps show for per process memory usage also lie? > > What I see here on my news server, is that I run out of swap (256M), but > > according to top and ps a rough calculation of the total of all the > > processes is less than half that. Inn then typically show a usage of > > ~70M according to top, but as soon as I kill and restart it the swap usage > > go down to ~5M. I once even added a 128M vn swapfile and it filled that > > also without inn showing a usage of more than 70M, but killing and > > restarting it takes the swap usage down to ~5M. > > > > John > > -- > > John Hay -- John.Hay@mikom.csir.co.za > > > > From owner-freebsd-stable Tue Apr 29 15:02:38 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id PAA13744 for stable-outgoing; Tue, 29 Apr 1997 15:02:38 -0700 (PDT) Received: from gatekeeper.alcatel.com.au (gatekeeper.alcatel.com.au [203.17.66.1]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id PAA13731 for ; Tue, 29 Apr 1997 15:02:21 -0700 (PDT) Received: from 139.188.23.1 (139.188.23.1) by gatekeeper.alcatel.com.au (PMDF V5.0-5 #11861) id <01IIB550JGU80004D5@gatekeeper.alcatel.com.au> for freebsd-stable@freefall.FreeBSD.ORG; Wed, 30 Apr 1997 07:57:31 +1000 Received: from cbd.alcatel.com.au (cbd.alcatel.com.au) by cim.alcatel.com.au (PMDF V5.0-6 #9239) id <01IIB598E2OG8ZEOY4@cim.alcatel.com.au> for freebsd-stable@freefall.FreeBSD.ORG; Wed, 30 Apr 1997 08:00:55 +1000 Received: from gsms01.alcatel.com.au (gsms01.alcatel.com.au) by cbd.alcatel.com.au (PMDF V5.0-5 #9241) id <01IIB55L0IE89YDT0T@cbd.alcatel.com.au> for freebsd-stable@freefall.FreeBSD.ORG; Wed, 30 Apr 1997 07:57:59 +1100 Received: (from jeremyp@localhost) by gsms01.alcatel.com.au (8.7.5/8.7.3) id IAA11983 for freebsd-stable@freefall.FreeBSD.ORG; Wed, 30 Apr 1997 08:01:22 +1000 (EST) Date: Wed, 30 Apr 1997 08:01:22 +1000 (EST) From: Peter Jeremy Subject: Swap leaks (was Re: Memory usage on NFS server) To: freebsd-stable@freefall.FreeBSD.ORG Message-id: <199704292201.IAA11983@gsms01.alcatel.com.au> Content-transfer-encoding: 7BIT Sender: owner-stable@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk On Tue, 29 Apr 1997 10:03:57 +0200 (SAT), John Hay wrote: > Inn then typically show a usage of >~70M according to top, but as soon as I kill and restart it the swap >usage go down to ~5M. Some of the BSD/386-derived Unices have a memory leak in the VM subsystem which matches the above behaviour: When a process fork()d whilst the system was paging, swap space allocated to the child was not released until the _parent_ exit()d. This primarily bit long- running daemons that fork() regularly: cron, inetd, innd and sendmail. I thought FreeBSD 2.1 and 2.2 included fixes for this VM problem. Can someone please confirm this. (In which case John's problem is something else). Peter -- Peter Jeremy (VK2PJ) peter.jeremy@alcatel.com.au Alcatel Australia Limited 41 Mandible St Phone: +61 2 9690 5019 ALEXANDRIA NSW 2015 Fax: +61 2 9690 5247 From owner-freebsd-stable Tue Apr 29 15:31:25 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id PAA14927 for stable-outgoing; Tue, 29 Apr 1997 15:31:25 -0700 (PDT) Received: from halyard.my.domain (schieffe@netcom23.netcom.com [192.100.81.137]) by hub.freebsd.org (8.8.5/8.8.5) with SMTP id PAA14922 for ; Tue, 29 Apr 1997 15:31:21 -0700 (PDT) Received: from tiller.my.domain (tiller [192.168.1.4]) by halyard.my.domain (8.6.12/8.6.12) with SMTP id RAA00585 for ; Tue, 29 Apr 1997 17:31:14 -0500 Received: by tiller.my.domain with Microsoft Mail id <01BC54C3.250B8880@tiller.my.domain>; Tue, 29 Apr 1997 17:31:17 -0500 Message-ID: <01BC54C3.250B8880@tiller.my.domain> From: "Jerry K. Schieffer" To: "'freebsd-stable@freebsd.org'" Subject: Adaptec 2842 SCSI controller issues Date: Tue, 29 Apr 1997 17:31:16 -0500 MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: owner-stable@freebsd.org X-Loop: FreeBSD.org Precedence: bulk I haven't posted to any freebsd lists before, so if I am in the wrong place, please let me know which list this ought to be directed to. I chose this one since I recall having seen similar postings here in the last few days. I am having problems with a configuration as follows: 486/66 VL mainboard VL EIDE controller Adaptec 2842 (VL) SCSI controller etc. FreeBSD 2.1.7.1+ CTM fixes from src-2.1.0284xEmpty.gz through src-2.1.0299.gz used to build kernel. During boot, the AHA controller seems to hang after hardware resets (more details in a moment), but all seems to work if I boot DOS first (DOS is on the IDE drive, the AHA BIOS is disabled, DOS loads the ASPI drivers v3.0) and then reboot using ctl-alt-del. I ran a short experiment to see what conditions failed and what didn't. For now, I use the workaround, but won't recover from a power fail restart. Here is what happens - (Hope this helps) Booting the system, three partitions on wd0 (root, usr, var ) get mounted OK. In the middle of mounts, mounts rwd0a, rwd0s2f, rwds2e (all clean) should go get sd0 stuff next - 1 partition. Timedout SCB handled by another timeout Timedout SCB handled by another timeout Timedout SCB handled by another timeout sd0(ahc0:0:0): SCB 0x0 - timed out in message phase, SCSISIGI == 0xe6 SEQADR == 0x16c sd0(ahc0:0:0): abort message in message buffer ahc0:a:0: Missed busfree. sd0(ahc0:0:0): SCB 0x0 - timed out while idle, LASTPHASE == 0x1, SCSISIGI == 0xb 6 SEQADDR == 0x4 sd0(ahc0:0:0): no longer in timeout ahc0: Issued Channel A Bus Reset. 1 SCBs aborted Then system hangs there. reset to reboot same results power down and restart same result reset and boot dos reset and boot FreeBSD same result reset and boot dos dir d: reset and boot FreeBSD same result reset and boot dos ctl-alt-del everything works conclusion: must have something to do with hardware reset. This same behavior was seen with a kernel including fixes thru ...94 also. regards, Jerry Schieffer schieffe@netcom.com From owner-freebsd-stable Tue Apr 29 16:07:02 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id QAA16481 for stable-outgoing; Tue, 29 Apr 1997 16:07:02 -0700 (PDT) Received: from dyson.iquest.net (dyson.iquest.net [198.70.144.127]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id QAA16469 for ; Tue, 29 Apr 1997 16:06:52 -0700 (PDT) Received: (from root@localhost) by dyson.iquest.net (8.8.4/8.6.9) id SAA07175; Tue, 29 Apr 1997 18:05:13 -0500 (EST) From: "John S. Dyson" Message-Id: <199704292305.SAA07175@dyson.iquest.net> Subject: Re: Swap leaks (was Re: Memory usage on NFS server) In-Reply-To: <199704292201.IAA11983@gsms01.alcatel.com.au> from Peter Jeremy at "Apr 30, 97 08:01:22 am" To: peter.jeremy@alcatel.com.au (Peter Jeremy) Date: Tue, 29 Apr 1997 18:05:03 -0500 (EST) Cc: freebsd-stable@freefall.FreeBSD.ORG X-Mailer: ELM [version 2.4ME+ PL31 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-stable@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > On Tue, 29 Apr 1997 10:03:57 +0200 (SAT), John Hay > wrote: > > Inn then typically show a usage of > >~70M according to top, but as soon as I kill and restart it the swap > >usage go down to ~5M. > > Some of the BSD/386-derived Unices have a memory leak in the VM > subsystem which matches the above behaviour: When a process fork()d > whilst the system was paging, swap space allocated to the child was > not released until the _parent_ exit()d. This primarily bit long- > running daemons that fork() regularly: cron, inetd, innd and sendmail. > > I thought FreeBSD 2.1 and 2.2 included fixes for this VM problem. > Can someone please confirm this. (In which case John's problem is > something else). > The swap-leak problem is fixed as completely as resonable. Once swap space is allocated for a given object in a process, it stays around until the object is destroyed. Swap space is not recovered when pages are paged back in. That is a performance design decision. John From owner-freebsd-stable Tue Apr 29 16:08:50 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id QAA16702 for stable-outgoing; Tue, 29 Apr 1997 16:08:50 -0700 (PDT) Received: from pluto.plutotech.com (root@pluto100.plutotech.com [206.168.67.137]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id QAA16692 for ; Tue, 29 Apr 1997 16:08:45 -0700 (PDT) Received: from narnia.plutotech.com (narnia.plutotech.com [206.168.67.130]) by pluto.plutotech.com (8.8.5/8.8.3) with ESMTP id RAA01221; Tue, 29 Apr 1997 17:08:35 -0600 (MDT) Message-Id: <199704292308.RAA01221@pluto.plutotech.com> X-Mailer: exmh version 2.0beta 12/23/96 To: "Jerry K. Schieffer" cc: "'freebsd-stable@freebsd.org'" Subject: Re: Adaptec 2842 SCSI controller issues In-reply-to: Your message of "Tue, 29 Apr 1997 17:31:16 CDT." <01BC54C3.250B8880@tiller.my.domain> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Tue, 29 Apr 1997 18:07:08 -0600 From: "Justin T. Gibbs" Sender: owner-stable@freebsd.org X-Loop: FreeBSD.org Precedence: bulk >conclusion: must have something to do with hardware reset. Actually it probably means that I should be relying on the serial eeprom on the 284X cards instead of using the values left over in scratch ram. This is probalby screwing up the parameters used in sync negotiation. -- Justin T. Gibbs =========================================== FreeBSD: Turning PCs into workstations =========================================== From owner-freebsd-stable Tue Apr 29 17:52:52 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id RAA21274 for stable-outgoing; Tue, 29 Apr 1997 17:52:52 -0700 (PDT) Received: from sierra.zyzzyva.com (ppp01-58.zyzzyva.com [208.214.58.10]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id RAA21264 for ; Tue, 29 Apr 1997 17:52:48 -0700 (PDT) Received: from sierra (localhost [127.0.0.1]) by sierra.zyzzyva.com (8.8.5/8.8.2) with ESMTP id TAA05475; Tue, 29 Apr 1997 19:51:04 -0500 (CDT) Message-Id: <199704300051.TAA05475@sierra.zyzzyva.com> X-Mailer: exmh version 2.0gamma 1/27/96 To: Tom Samplonius cc: Randy Terbush , John Hay , dg@root.com, freebsd-stable@FreeBSD.org Subject: Re: Memory usage on NFS server In-reply-to: tom's message of Tue, 29 Apr 1997 11:52:18 -0700. X-uri: http://www.zyzzyva.com/ Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Tue, 29 Apr 1997 19:51:04 -0500 From: Randy Terbush Sender: owner-stable@FreeBSD.org X-Loop: FreeBSD.org Precedence: bulk This behavior has been observed with Apache 1.2 throughout the development cycle through 1.2b10 (just released). > On Tue, 29 Apr 1997, Randy Terbush wrote: > > > > > For the record, I've seen the same behavior on a machine whose > > primary job is webservices. This is both for 2.1.7 and 2.2.1. > > The swap will be freed by killing and restarting the webservers. > > > > The machine in question has 64MB RAM and 150MB swap. > > > > The webserver is Apache. > > Both inn and apache used mmap'ed memory. When a process mmaps a shared > file, to what process is the memory usage associated with? > > It could be what you are seeing here is a mmap() leak in these > applications, and that mmap'ed memory is not associated with any process. > > Which version of apache? Which version of inn? > > Tom > > > > > Do the numbers that top and ps show for per process memory usage also lie? > > > What I see here on my news server, is that I run out of swap (256M), but > > > according to top and ps a rough calculation of the total of all the > > > processes is less than half that. Inn then typically show a usage of > > > ~70M according to top, but as soon as I kill and restart it the swap usage > > > go down to ~5M. I once even added a 128M vn swapfile and it filled that > > > also without inn showing a usage of more than 70M, but killing and > > > restarting it takes the swap usage down to ~5M. > > > > > > John > > > -- > > > John Hay -- John.Hay@mikom.csir.co.za > > > > > > > > From owner-freebsd-stable Tue Apr 29 18:05:56 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id SAA21766 for stable-outgoing; Tue, 29 Apr 1997 18:05:56 -0700 (PDT) Received: from mercury.uniserve.com (mercury.uniserve.com [204.191.197.248]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id SAA21761 for ; Tue, 29 Apr 1997 18:05:51 -0700 (PDT) Received: from haven.uniserve.com (shell.uniserve.com [198.53.215.121]) by mercury.uniserve.com with SMTP id SAA28090; Tue, 29 Apr 1997 18:05:00 -0700 (PDT) Date: Tue, 29 Apr 1997 18:11:15 -0700 (PDT) From: Tom Samplonius To: Randy Terbush cc: John Hay , dg@root.com, freebsd-stable@FreeBSD.org Subject: Re: Memory usage on NFS server In-Reply-To: <199704300051.TAA05475@sierra.zyzzyva.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-stable@FreeBSD.org X-Loop: FreeBSD.org Precedence: bulk On Tue, 29 Apr 1997, Randy Terbush wrote: > This behavior has been observed with Apache 1.2 throughout the > development cycle through 1.2b10 (just released). Definitely not true for earlier versions. I've got Apache 1.0.5 running on a 2.1-stable system here for over 183 days. It is only mildly busy though: 70,000 access_log entries in 24 hours. No evidence of this memory leak here. Tom > > On Tue, 29 Apr 1997, Randy Terbush wrote: > > > > > > > > For the record, I've seen the same behavior on a machine whose > > > primary job is webservices. This is both for 2.1.7 and 2.2.1. > > > The swap will be freed by killing and restarting the webservers. > > > > > > The machine in question has 64MB RAM and 150MB swap. > > > > > > The webserver is Apache. > > > > Both inn and apache used mmap'ed memory. When a process mmaps a shared > > file, to what process is the memory usage associated with? > > > > It could be what you are seeing here is a mmap() leak in these > > applications, and that mmap'ed memory is not associated with any process. > > > > Which version of apache? Which version of inn? > > > > Tom > > > > > > > > Do the numbers that top and ps show for per process memory usage also lie? > > > > What I see here on my news server, is that I run out of swap (256M), but > > > > according to top and ps a rough calculation of the total of all the > > > > processes is less than half that. Inn then typically show a usage of > > > > ~70M according to top, but as soon as I kill and restart it the swap usage > > > > go down to ~5M. I once even added a 128M vn swapfile and it filled that > > > > also without inn showing a usage of more than 70M, but killing and > > > > restarting it takes the swap usage down to ~5M. > > > > > > > > John > > > > -- > > > > John Hay -- John.Hay@mikom.csir.co.za > > > > > > > > > > > > > > > > From owner-freebsd-stable Tue Apr 29 22:46:07 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id WAA06584 for stable-outgoing; Tue, 29 Apr 1997 22:46:07 -0700 (PDT) Received: from zibbi.mikom.csir.co.za (zibbi.mikom.csir.co.za [146.64.24.58]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id WAA06575 for ; Tue, 29 Apr 1997 22:45:37 -0700 (PDT) Received: (from jhay@localhost) by zibbi.mikom.csir.co.za (8.8.5/8.8.5) id HAA19298; Wed, 30 Apr 1997 07:44:26 +0200 (SAT) From: John Hay Message-Id: <199704300544.HAA19298@zibbi.mikom.csir.co.za> Subject: Re: Memory usage on NFS server In-Reply-To: from Tom Samplonius at "Apr 29, 97 11:52:18 am" To: tom@uniserve.com (Tom Samplonius) Date: Wed, 30 Apr 1997 07:44:26 +0200 (SAT) Cc: freebsd-stable@FreeBSD.org X-Mailer: ELM [version 2.4ME+ PL31 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-stable@FreeBSD.org X-Loop: FreeBSD.org Precedence: bulk > > > > For the record, I've seen the same behavior on a machine whose > > primary job is webservices. This is both for 2.1.7 and 2.2.1. > > The swap will be freed by killing and restarting the webservers. > > > > The machine in question has 64MB RAM and 150MB swap. > > > > The webserver is Apache. > > Both inn and apache used mmap'ed memory. When a process mmaps a shared > file, to what process is the memory usage associated with? > > It could be what you are seeing here is a mmap() leak in these > applications, and that mmap'ed memory is not associated with any process. > > Which version of apache? Which version of inn? > I'm using inn 1.5.1 without mmap(), though I have tried the mmap() patch for nnrpd the last few days, but it didn't make a difference to the "swap leakage". I think it must be some setting in inn because between December and February it was very stable (running on 2.2-ALPHA also using inn 1.5.1). have already tried the old kernel but that didn't help, so I will try to backtrack my "tweaks" to the inn config file. I'm also using a patch to fix the pipe bug in inn which shows mostly as dropped articles to overchan. I'll back that out also. > > > > > Do the numbers that top and ps show for per process memory usage also lie? > > > What I see here on my news server, is that I run out of swap (256M), but > > > according to top and ps a rough calculation of the total of all the > > > processes is less than half that. Inn then typically show a usage of > > > ~70M according to top, but as soon as I kill and restart it the swap usage > > > go down to ~5M. I once even added a 128M vn swapfile and it filled that > > > also without inn showing a usage of more than 70M, but killing and > > > restarting it takes the swap usage down to ~5M. > > > John -- John Hay -- John.Hay@mikom.csir.co.za From owner-freebsd-stable Wed Apr 30 06:40:05 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id GAA26569 for stable-outgoing; Wed, 30 Apr 1997 06:40:05 -0700 (PDT) Received: from dyson.iquest.net (dyson.iquest.net [198.70.144.127]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id GAA26560 for ; Wed, 30 Apr 1997 06:40:02 -0700 (PDT) Received: (from root@localhost) by dyson.iquest.net (8.8.4/8.6.9) id JAA06469; Wed, 30 Apr 1997 09:39:15 -0500 (EST) From: "John S. Dyson" Message-Id: <199704301439.JAA06469@dyson.iquest.net> Subject: Re: Memory usage on NFS server In-Reply-To: <199704300051.TAA05475@sierra.zyzzyva.com> from Randy Terbush at "Apr 29, 97 07:51:04 pm" To: randy@zyzzyva.com (Randy Terbush) Date: Wed, 30 Apr 1997 09:39:15 -0500 (EST) Cc: tom@uniserve.com, randy@zyzzyva.com, jhay@zibbi.mikom.csir.co.za, dg@root.com, freebsd-stable@FreeBSD.ORG X-Mailer: ELM [version 2.4ME+ PL31 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-stable@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > > > This behavior has been observed with Apache 1.2 throughout the > development cycle through 1.2b10 (just released). > Okay -- I'll take a look at it. I'll run some mmap tests, after looking at the apache sources. John From owner-freebsd-stable Wed Apr 30 09:02:11 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id JAA03271 for stable-outgoing; Wed, 30 Apr 1997 09:02:11 -0700 (PDT) Received: from dopey.pathlink.com (dopey.pathlink.com [204.30.237.226]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id JAA03260 for ; Wed, 30 Apr 1997 09:02:08 -0700 (PDT) Received: from dvl-1.pathlink.com (dvl-1.pathlink.com [204.30.237.241]) by dopey.pathlink.com (8.8.5/8.8.5) with SMTP id JAA09733 for ; Wed, 30 Apr 1997 09:10:31 -0700 (PDT) Message-Id: <1.5.4.32.19970430090159.006da760@dopey.pathlink.com> X-Sender: kachun@dopey.pathlink.com X-Mailer: Windows Eudora Light Version 1.5.4 (32) Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Wed, 30 Apr 1997 09:01:59 +0000 To: freebsd-stable@freebsd.org From: Kachun Lee Subject: kmem_map full with 2.2beta to 2.2-releng-970422 Sender: owner-stable@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Since did not get a response from freebsd-question yet, I am trying my luck here. After upgraded some of our news servers to 2.2, they would panic about every other days with the kmem_map full (Just for info, they went fine with 2.1.6). I increased NMBCLUSTERS from 3000 to 4000 to 5000 to 6000 and that did not help. I logged the netstat -m every 15 minutes and the mbufs usage was in the 2000s when the system crashed. Any suggestion, or where or how I should look for the problem? Thank you in advance. PS: system conf-> maxusers 128 # options MATH_EMULATE #Support for x87 emulation options INET #InterNETworking options FFS #Berkeley Fast Filesystem options NFS #Network Filesystem options MFS #Memory File System options MSDOSFS #MSDOS Filesystem options "CD9660" #ISO 9660 Filesystem options PROCFS #Process filesystem options "COMPAT_43" #Compatible with BSD 4.3 options SCSI_DELAY=15 #Be pessimistic about Joe SCSI device options UCONSOLE #Allow users to grab the console options OPEN_MAX=4000 options CHILD_MAX=2000 options NMBCLUSTERS=5000 options "MAXMEM=(192*1024)" options SYSVSHM options SYSVSEM options SYSVMSG From owner-freebsd-stable Wed Apr 30 13:37:15 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id NAA16758 for stable-outgoing; Wed, 30 Apr 1997 13:37:15 -0700 (PDT) Received: from who.cdrom.com (who.cdrom.com [204.216.27.3]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id NAA16747 for ; Wed, 30 Apr 1997 13:37:11 -0700 (PDT) Received: from mercury.uniserve.com (mercury.uniserve.com [204.191.197.248]) by who.cdrom.com (8.8.5/8.6.11) with ESMTP id KAA06388 for ; Wed, 30 Apr 1997 10:38:04 -0700 (PDT) Received: from haven.uniserve.com (shell.uniserve.com [198.53.215.121]) by mercury.uniserve.com with SMTP id KAA18352; Wed, 30 Apr 1997 10:37:21 -0700 (PDT) Date: Wed, 30 Apr 1997 10:43:40 -0700 (PDT) From: Tom Samplonius To: Kachun Lee cc: freebsd-stable@freebsd.org Subject: Re: kmem_map full with 2.2beta to 2.2-releng-970422 In-Reply-To: <1.5.4.32.19970430090159.006da760@dopey.pathlink.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-stable@freebsd.org X-Loop: FreeBSD.org Precedence: bulk On Wed, 30 Apr 1997, Kachun Lee wrote: > Since did not get a response from freebsd-question yet, I am trying my luck > here. > > After upgraded some of our news servers to 2.2, they would panic about every > other days with the kmem_map full (Just for info, they went fine with > 2.1.6). I increased NMBCLUSTERS from 3000 to 4000 to 5000 to 6000 and that > did not help. I logged the netstat -m every 15 minutes and the mbufs usage > was in the 2000s when the system crashed. > > Any suggestion, or where or how I should look for the problem? > > Thank you in advance. > > PS: system conf-> > > maxusers 128 > > # options MATH_EMULATE #Support for x87 emulation > options INET #InterNETworking > options FFS #Berkeley Fast Filesystem > options NFS #Network Filesystem > options MFS #Memory File System > options MSDOSFS #MSDOS Filesystem > options "CD9660" #ISO 9660 Filesystem > options PROCFS #Process filesystem > options "COMPAT_43" #Compatible with BSD 4.3 > options SCSI_DELAY=15 #Be pessimistic about Joe SCSI device > options UCONSOLE #Allow users to grab the console > > options OPEN_MAX=4000 > options CHILD_MAX=2000 These two are insanely large. You might want to bring these both down to 500 or so. > options NMBCLUSTERS=5000 > > options "MAXMEM=(192*1024)" > > options SYSVSHM > options SYSVSEM > options SYSVMSG If you remove this SYS* stuff you get a smaller kernel. Tom From owner-freebsd-stable Wed Apr 30 14:02:05 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id OAA17849 for stable-outgoing; Wed, 30 Apr 1997 14:02:05 -0700 (PDT) Received: from tor-adm1.nbc.netcom.ca (taob@tor-adm1.nbc.netcom.ca [207.181.89.5]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id OAA17838 for ; Wed, 30 Apr 1997 14:01:55 -0700 (PDT) Received: from localhost (taob@localhost) by tor-adm1.nbc.netcom.ca (8.8.5/8.8.5) with SMTP id RAA20227; Wed, 30 Apr 1997 17:00:30 -0400 (EDT) Date: Wed, 30 Apr 1997 17:00:30 -0400 (EDT) From: Brian Tao To: Kachun Lee cc: freebsd-stable@FreeBSD.org Subject: Re: kmem_map full with 2.2beta to 2.2-releng-970422 In-Reply-To: <1.5.4.32.19970430090159.006da760@dopey.pathlink.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-stable@FreeBSD.org X-Loop: FreeBSD.org Precedence: bulk On Wed, 30 Apr 1997, Kachun Lee wrote: > > options OPEN_MAX=4000 > options CHILD_MAX=2000 Wow, do you really need to set them that high? -- Brian Tao (BT300, taob@netcom.ca) "Though this be madness, yet there is method in't" From owner-freebsd-stable Wed Apr 30 20:36:13 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id UAA04646 for stable-outgoing; Wed, 30 Apr 1997 20:36:13 -0700 (PDT) Received: from who.cdrom.com (who.cdrom.com [204.216.27.3]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id UAA04631 for ; Wed, 30 Apr 1997 20:36:10 -0700 (PDT) Received: from eris.quintessential.com ([206.9.129.30]) by who.cdrom.com (8.8.5/8.6.11) with ESMTP id UAA09476 for ; Wed, 30 Apr 1997 20:07:49 -0700 (PDT) Received: from pm1-20.pclink.com (pm4-08.pclink.com [206.11.0.138]) by eris.quintessential.com (8.8.3/8.8.3) with SMTP id WAA18298 for ; Wed, 30 Apr 1997 22:06:34 -0500 (CDT) Message-ID: <3368254C.679D@freeq.com> Date: Wed, 30 Apr 1997 22:08:28 -0700 From: Gypsy Rogers Organization: Quintessential Communications X-Mailer: Mozilla 3.01 (Win95; I; 16bit) MIME-Version: 1.0 To: freebsd-stable@FreeBSD.ORG Subject: subscribe freebsd-stable Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-stable@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk subscribe freebsd-stable -- =+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+= SysAdmin: Quintessential Communications | Youth is not a Disability. http://www.quintessential.com | It's a wealth of energy and http://freeq.com | innovation, use it or lose it. =+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+= From owner-freebsd-stable Thu May 1 05:34:52 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id FAA21794 for stable-outgoing; Thu, 1 May 1997 05:34:52 -0700 (PDT) Received: from Pkrw.tcn.net (Pkrw.tcn.net [199.166.4.58]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id FAA21789 for ; Thu, 1 May 1997 05:34:49 -0700 (PDT) Received: from localhost (krw@localhost) by Pkrw.tcn.net (8.8.5/8.8.5) with SMTP id IAA07390 for ; Thu, 1 May 1997 08:38:33 -0400 (EDT) X-Authentication-Warning: Pkrw.tcn.net: krw owned process doing -bs Date: Thu, 1 May 1997 08:38:31 -0400 (EDT) From: "Kenneth R. Westerback" To: stable@freebsd.org Subject: 2.2.1 kernel compile fails with ctm src-2.2.0258 - opt_wd.h missing? Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-stable@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Having done a make world with src-2.2.0258 I tried to make my customized kernel and got the following: cc -c -O -Wreturn-type -Wcomment -Wredundant-decls -Wimplicit -Wnested-externs -Wstrict-prototypes -Wmissing-prototypes -Wpointer-arith -nostdinc -I- -I. -I../.. -I../../../include -DSB16MIDI_BASE=0x300 -DSB16_DMA=7 -DSBC_IRQ=10 -DFAILSAFE -DCOMPAT_43 -DCD9660 -DEXT2FS -DMSDOSFS -DFFS -DINET -DKERNEL ../../pci/wdc_p.c ../../pci/wdc_p.c:29: opt_wd.h: No such file or directory *** Error code 1 Stop. I tried a 'make clean; make depend; make' and got the same error message. I believe this actually started one or two ctm updates ago and I was hoping such an apparently obvious problem would be quickly fixed. I did a 'find / -name opt_wd.h -print' and no hits, so I guess there really IS no file opt_wd.h. ---- Ken From owner-freebsd-stable Thu May 1 09:50:55 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id JAA02222 for stable-outgoing; Thu, 1 May 1997 09:50:55 -0700 (PDT) Received: from dopey.pathlink.com (dopey.pathlink.com [204.30.237.226]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id JAA02217 for ; Thu, 1 May 1997 09:50:52 -0700 (PDT) Received: from dvl-1.pathlink.com (dvl-1.pathlink.com [204.30.237.241]) by dopey.pathlink.com (8.8.5/8.8.5) with SMTP id JAA11640; Thu, 1 May 1997 09:59:24 -0700 (PDT) Message-Id: <1.5.4.32.19970501095033.006e6dc8@dopey.pathlink.com> X-Sender: kachun@dopey.pathlink.com X-Mailer: Windows Eudora Light Version 1.5.4 (32) Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Thu, 01 May 1997 09:50:33 +0000 To: dg@root.com From: Kachun Lee Subject: Re: kmem_map full with 2.2beta to 2.2-releng-970422 Cc: freebsd-stable@freebsd.org Sender: owner-stable@freebsd.org X-Loop: FreeBSD.org Precedence: bulk At 11:14 PM 4/30/97 -0700, you wrote: >> [snip] > > That panic is not related to the mbuf clusters problem. The kmem_map full >panic indicates that the kernel has run out of malloc space. The standard >default size is 32MB. You can increase it by changing VM_KMEM_SIZE in >i386/include/vmparam.h. > >-DG > >David Greenman >Core-team/Principal Architect, The FreeBSD Project Thank you! That's exactly what the patient needed. While I got your attention, the kmem_map full was happening at our NFS clients, and our NFS servers have quite a lot of these: /kernel: nfsd send error 55 They does not seem cause any ill effect. I assume 55=ENOBUFS No buffer space available. But I thought with the unified VM/Buffer, there are no MAXBUFFER setting any more. Do you know what parameter I should adjust? Thank you again. From owner-freebsd-stable Thu May 1 09:51:07 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id JAA02249 for stable-outgoing; Thu, 1 May 1997 09:51:07 -0700 (PDT) Received: from sliphost37.uni-trier.de (root@sliphost37.uni-trier.de [136.199.240.37]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id JAA02242 for ; Thu, 1 May 1997 09:51:02 -0700 (PDT) Received: (from blank@localhost) by sliphost37.uni-trier.de (8.8.5/8.8.5) id SAA01255 for stable@freebsd.org; Thu, 1 May 1997 18:48:42 +0200 (CEST) From: Sascha Blank Message-Id: <199705011648.SAA01255@sliphost37.uni-trier.de> Subject: Re: 2.2.1 kernel compile fails with ctm src-2.2.0258 - opt_wd.h missing? In-Reply-To: from "Kenneth R. Westerback" at "May 1, 97 08:38:31 am" To: stable@freebsd.org Date: Thu, 1 May 1997 18:48:42 +0200 (CEST) Reply-To: blank@fox.uni-trier.de (Sascha Blank) X-System: FreeBSD 2.2-STABLE X-Mailer: ELM [version 2.4ME+ PL31H (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-stable@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Hello, Kenneth R. Westerback has written recently: > Having done a make world with src-2.2.0258 I tried to make my customized > kernel and got the following: > > cc -c -O -Wreturn-type -Wcomment -Wredundant-decls -Wimplicit -Wnested-externs -Wstrict-prototypes -Wmissing-prototypes -Wpointer-arith -nostdinc -I- -I. -I../.. -I../../../include -DSB16MIDI_BASE=0x300 -DSB16_DMA=7 -DSBC_IRQ=10 -DFAILSAFE -DCOMPAT_43 > -DCD9660 -DEXT2FS -DMSDOSFS -DFFS -DINET -DKERNEL ../../pci/wdc_p.c > ../../pci/wdc_p.c:29: opt_wd.h: No such file or directory > *** Error code 1 > > Stop. You must reconfigure your kernel from its associated configuration file using "config YOUR_KERNEL_CONFIG_FILE". This will create the "opt_wd.h" file you are currently lacking, so compilation should work again. -- Sascha Blank - mailto:blank@fox.uni-trier.de Student and System Administrator at the University of Trier, Germany Finger my account to receive my Public PGP key I don't speak for my employers, they don't pay me enough for that. From owner-freebsd-stable Thu May 1 09:58:51 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id JAA02690 for stable-outgoing; Thu, 1 May 1997 09:58:51 -0700 (PDT) Received: from dopey.pathlink.com (dopey.pathlink.com [204.30.237.226]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id JAA02685 for ; Thu, 1 May 1997 09:58:44 -0700 (PDT) Received: from dvl-1.pathlink.com (dvl-1.pathlink.com [204.30.237.241]) by dopey.pathlink.com (8.8.5/8.8.5) with SMTP id KAA11656; Thu, 1 May 1997 10:07:08 -0700 (PDT) Message-Id: <1.5.4.32.19970501095817.006df320@dopey.pathlink.com> X-Sender: kachun@dopey.pathlink.com X-Mailer: Windows Eudora Light Version 1.5.4 (32) Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Thu, 01 May 1997 09:58:17 +0000 To: Tom Samplonius From: Kachun Lee Subject: Re: kmem_map full with 2.2beta to 2.2-releng-970422 Cc: freebsd-stable@freebsd.org Sender: owner-stable@freebsd.org X-Loop: FreeBSD.org Precedence: bulk At 10:43 AM 4/30/97 -0700, you wrote: > >On Wed, 30 Apr 1997, Kachun Lee wrote: >> [snip] >> options SYSVSHM >> options SYSVSEM >> options SYSVMSG > > If you remove this SYS* stuff you get a smaller kernel. Thank you for the info. The nnrpd that we are running uses the SYSV share memory. I should get the mmap version, then I can remove them. From owner-freebsd-stable Thu May 1 10:03:33 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id KAA03017 for stable-outgoing; Thu, 1 May 1997 10:03:33 -0700 (PDT) Received: from dopey.pathlink.com (dopey.pathlink.com [204.30.237.226]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id KAA03012 for ; Thu, 1 May 1997 10:03:31 -0700 (PDT) Received: from dvl-1.pathlink.com (dvl-1.pathlink.com [204.30.237.241]) by dopey.pathlink.com (8.8.5/8.8.5) with SMTP id KAA11671; Thu, 1 May 1997 10:12:12 -0700 (PDT) Message-Id: <1.5.4.32.19970501100321.006e13c0@dopey.pathlink.com> X-Sender: kachun@dopey.pathlink.com X-Mailer: Windows Eudora Light Version 1.5.4 (32) Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Thu, 01 May 1997 10:03:21 +0000 To: "Jeffrey J. Mountin" From: Kachun Lee Subject: Re: kmem_map full with 2.2beta to 2.2-releng-970422 Cc: freebsd-stable@freebsd.org Sender: owner-stable@freebsd.org X-Loop: FreeBSD.org Precedence: bulk At 06:34 AM 5/1/97 -0500, you wrote: >At 09:01 AM 4/30/97 +0000, Kachun Lee wrote: >>Since did not get a response from freebsd-question yet, I am trying my luck >>here. > >Don't ya just hate that. ;-) Yep! Thank you for the reply. >>After upgraded some of our news servers to 2.2, they would panic about every >>other days with the kmem_map full (Just for info, they went fine with >>2.1.6). I increased NMBCLUSTERS from 3000 to 4000 to 5000 to 6000 and that >>did not help. I logged the netstat -m every 15 minutes and the mbufs usage >>was in the 2000s when the system crashed. >> >>Any suggestion, or where or how I should look for the problem? > >Have to wonder what you are doing/expecting to do with the server.... Just trying to make it running a LOT of nnrpd, and FreeBSD 2.2 is doing it, except the occasional kmem_map full. >> [snip] From owner-freebsd-stable Thu May 1 10:53:59 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id KAA05888 for stable-outgoing; Thu, 1 May 1997 10:53:59 -0700 (PDT) Received: from mercury.uniserve.com (mercury.uniserve.com [204.191.197.248]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id KAA05880 for ; Thu, 1 May 1997 10:53:56 -0700 (PDT) Received: from haven.uniserve.com (shell.uniserve.com [198.53.215.121]) by mercury.uniserve.com with SMTP id KAA10662; Thu, 1 May 1997 10:52:54 -0700 (PDT) Date: Thu, 1 May 1997 10:59:18 -0700 (PDT) From: Tom Samplonius To: Kachun Lee cc: freebsd-stable@freebsd.org Subject: Re: kmem_map full with 2.2beta to 2.2-releng-970422 In-Reply-To: <1.5.4.32.19970501095817.006df320@dopey.pathlink.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-stable@freebsd.org X-Loop: FreeBSD.org Precedence: bulk On Thu, 1 May 1997, Kachun Lee wrote: > At 10:43 AM 4/30/97 -0700, you wrote: > > > >On Wed, 30 Apr 1997, Kachun Lee wrote: > >> [snip] > >> options SYSVSHM > >> options SYSVSEM > >> options SYSVMSG > > > > If you remove this SYS* stuff you get a smaller kernel. > > Thank you for the info. The nnrpd that we are running uses the SYSV share > memory. I should get the mmap version, then I can remove them. > > The mmap() version is also more reliable, due to the strange way that SYSV shared memory works. Tom From owner-freebsd-stable Thu May 1 14:04:17 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id OAA16767 for stable-outgoing; Thu, 1 May 1997 14:04:17 -0700 (PDT) Received: from implode.root.com (implode.root.com [198.145.90.17]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id OAA16762 for ; Thu, 1 May 1997 14:04:14 -0700 (PDT) Received: from localhost (localhost [127.0.0.1]) by implode.root.com (8.8.5/8.8.5) with SMTP id OAA29970; Thu, 1 May 1997 14:05:27 -0700 (PDT) Message-Id: <199705012105.OAA29970@implode.root.com> X-Authentication-Warning: implode.root.com: localhost [127.0.0.1] didn't use HELO protocol To: Kachun Lee cc: freebsd-stable@freebsd.org Subject: Re: kmem_map full with 2.2beta to 2.2-releng-970422 In-reply-to: Your message of "Thu, 01 May 1997 09:50:33 -0000." <1.5.4.32.19970501095033.006e6dc8@dopey.pathlink.com> From: David Greenman Reply-To: dg@root.com Date: Thu, 01 May 1997 14:05:27 -0700 Sender: owner-stable@freebsd.org X-Loop: FreeBSD.org Precedence: bulk >clients, and our NFS servers have quite a lot of these: > > /kernel: nfsd send error 55 > >They does not seem cause any ill effect. I assume 55=ENOBUFS No buffer space >available. But I thought with the unified VM/Buffer, there are no MAXBUFFER >setting any more. Do you know what parameter I should adjust? ...that one indicates that network buffers were depleted. This usually happens when the interface queue reaches it's limit of 50 queued packets, although it can also happen if the global pool of network buffers is deleted. The most common cause of this error is a stuck network device driver - some NICs, for example will stop sending if the cable is unplugged. It can also happen on PPP/SLIP lines if hardware flow control (CTS) stays de-asserted too long. -DG David Greenman Core-team/Principal Architect, The FreeBSD Project From owner-freebsd-stable Thu May 1 14:49:46 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id OAA19444 for stable-outgoing; Thu, 1 May 1997 14:49:46 -0700 (PDT) Received: from implode.root.com (implode.root.com [198.145.90.17]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id OAA19439 for ; Thu, 1 May 1997 14:49:44 -0700 (PDT) Received: from localhost (localhost [127.0.0.1]) by implode.root.com (8.8.5/8.8.5) with SMTP id OAA00450; Thu, 1 May 1997 14:49:40 -0700 (PDT) Message-Id: <199705012149.OAA00450@implode.root.com> X-Authentication-Warning: implode.root.com: localhost [127.0.0.1] didn't use HELO protocol To: Kachun Lee cc: freebsd-stable@FreeBSD.ORG Subject: Re: kmem_map full with 2.2beta to 2.2-releng-970422 From: David Greenman Reply-To: dg@root.com Date: Thu, 01 May 1997 14:49:40 -0700 Sender: owner-stable@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk >>clients, and our NFS servers have quite a lot of these: >> >> /kernel: nfsd send error 55 >> >>They does not seem cause any ill effect. I assume 55=ENOBUFS No buffer space >>available. But I thought with the unified VM/Buffer, there are no MAXBUFFER >>setting any more. Do you know what parameter I should adjust? > > ...that one indicates that network buffers were depleted. This usually >happens when the interface queue reaches it's limit of 50 queued packets, >although it can also happen if the global pool of network buffers is deleted. ^^^^^^^ Oops, that should have been "depleted". -DG David Greenman Core-team/Principal Architect, The FreeBSD Project From owner-freebsd-stable Thu May 1 16:51:47 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id QAA26489 for stable-outgoing; Thu, 1 May 1997 16:51:47 -0700 (PDT) Received: from mail.ai.com (Mail.ai.com [204.144.182.5]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id QAA26484 for ; Thu, 1 May 1997 16:51:45 -0700 (PDT) From: rich@vegasone.com Message-Id: <199705012351.QAA26484@hub.freebsd.org> Received: from Dial-up323 ([208.193.202.83]) by mail.ai.com (Post.Office MTA v3.0 release 0121 ID# 0-32332U1000L100S10000) with SMTP id ACR193; Thu, 1 May 1997 17:25:44 -0600 Date: Thu, 01 May 1997 16:22:54 PDT Subject: Add LIVE Girls to your website! Sender: owner-stable@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk To: undisclosed-recipients:; Want to increase your profits by adding LIVE strippers to your adult website? Check out what Virtual Net Productions has to offer: We will provide a Live Stripshow with chat for your website starting at only $500 per month and NEVER more than $1600.00 a month!! (read on for details) The $500 rate is based on 744 hours (The amount of hours in a month for 1 person). - Additional hours beyond 744 will be rated at 60 cents per hour. You may or may not use additional hours depending on the amount of traffic you receive on your website. I can tell you that we have had our stream on over 100 different websites for the past year and only the top 10 sites exceeded 700 hours for any given month, so for the majority of adult websites our stream will never cost more than $500 per month. But what if you do use more? No problem cause we are capping the stream at $1600! You will never pay over $1600 no matter how many hours you use. And we are throwing in chat at no additional cost! So just to recap, get our stream on your website for $500 - no setup fees, you are paid in advance for the month for only $500. Then if you use more than 744 hours you pay .60 cents per hour up until you reach $1600, then you pay no more, and chat is included! Our stream has been used reliably on sites like Cybererotica, Sexmuseum and Newd.com just to name a few. Don't waste your money with untried services that can't match our quality or reliability, not to mention our prices! But don't take my word for it, see our girls in action for yourself. Just visit http://www.mojave.net/stream USER ID: visitor PASSWORD: live213 If you need any additional information ask for Rich or Glenn.. Rich Olson Virtual Net Productions OFFICE: 1-800-314-4285 or 1-702-360-4939 E-MAIL: rich@vegasone.com From owner-freebsd-stable Thu May 1 18:19:34 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id SAA02075 for stable-outgoing; Thu, 1 May 1997 18:19:34 -0700 (PDT) Received: from eris.quintessential.com ([206.9.129.30]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id SAA02066 for ; Thu, 1 May 1997 18:19:30 -0700 (PDT) Received: (from gypsy@localhost) by eris.quintessential.com (8.8.3/8.8.3) id UAA19182; Thu, 1 May 1997 20:18:16 -0500 (CDT) Date: Thu, 1 May 1997 20:18:15 -0500 (CDT) From: Gypsy Rogers X-Sender: gypsy@eris.quintessential.com To: stable@freebsd.org Subject: error Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-stable@freebsd.org X-Loop: FreeBSD.org Precedence: bulk ok, I've never had to upgrade the OS on one of my servers before, and now I am trying to do it, so I went and got 2.1.7.1 and did a make world on it, but it keeps crashing in the same place, any Ideas where I should look on this? (see below) =+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+= SysAdmin: Quintessential Communications | Youth is not a Disability. http://www.Quintessential.com | It's a wealth of energy and http://FreeQ.com | innovation, use it or lose it. =+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+= ===> sbin/fsirand cc -O -c /usr/homes/gypsy/freebsd/FreeBSD-stable/src/sbin/fsirand/fsirand.c /usr/homes/gypsy/freebsd/FreeBSD-stable/src/sbin/fsirand/fsirand.c: In function `fsirand': /usr/homes/gypsy/freebsd/FreeBSD-stable/src/sbin/fsirand/fsirand.c:203: structure has no member named `fs_id' /usr/homes/gypsy/freebsd/FreeBSD-stable/src/sbin/fsirand/fsirand.c:203: structure has no member named `fs_id' /usr/homes/gypsy/freebsd/FreeBSD-stable/src/sbin/fsirand/fsirand.c:204: structure has no member named `fs_id' /usr/homes/gypsy/freebsd/FreeBSD-stable/src/sbin/fsirand/fsirand.c:206: structure has no member named `fs_id' /usr/homes/gypsy/freebsd/FreeBSD-stable/src/sbin/fsirand/fsirand.c:207: structure has no member named `fs_id' /usr/homes/gypsy/freebsd/FreeBSD-stable/src/sbin/fsirand/fsirand.c:208: structure has no member named `fs_id' /usr/homes/gypsy/freebsd/FreeBSD-stable/src/sbin/fsirand/fsirand.c:214: structure has no member named `fs_id' /usr/homes/gypsy/freebsd/FreeBSD-stable/src/sbin/fsirand/fsirand.c:215: structure has no member named `fs_id' *** Error code 1 Stop. *** Error code 1 Stop. *** Error code 1 Stop. *** Error code 1 Stop. From owner-freebsd-stable Thu May 1 19:27:56 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id TAA05514 for stable-outgoing; Thu, 1 May 1997 19:27:56 -0700 (PDT) Received: from bofh.noc.best.net (rone@ennui.org [205.149.163.54]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id TAA05509 for ; Thu, 1 May 1997 19:27:54 -0700 (PDT) Received: (from rone@localhost) by bofh.noc.best.net (8.8.5/8.7.3) id TAA04716 for stable@freebsd.org; Thu, 1 May 1997 19:27:48 -0700 (PDT) From: Ron Echeverri Message-Id: <199705020227.TAA04716@bofh.noc.best.net> Subject: Re: Add LIVE Girls to your website! In-Reply-To: <199705012351.QAA26484@hub.freebsd.org> from "rich@vegasone.com" at "May 1, 97 04:22:54 pm" To: stable@freebsd.org Date: Thu, 1 May 1997 19:27:48 -0700 (PDT) X-GmbH: Gesellschaft mit beschränkter Haftung X-Mailer: ELM [version 2.4ME+ PL31 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-stable@freebsd.org X-Loop: FreeBSD.org Precedence: bulk rich@vegasone.com writes: We will provide a Live Stripshow with chat for your website starting at only $500 per month and NEVER more than $1600.00 a month!! (read on for details) Is this in the ports directory? rone -- Ron Echeverri Best Internet Usenet Administration rone@best.net From owner-freebsd-stable Thu May 1 20:22:22 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id UAA08360 for stable-outgoing; Thu, 1 May 1997 20:22:22 -0700 (PDT) Received: from lithium.maximum.net (daleg@x.maximum.net [205.161.109.86]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id UAA08341 for ; Thu, 1 May 1997 20:22:19 -0700 (PDT) Received: from localhost (daleg@localhost) by lithium.maximum.net (8.8.5/8.8.4) with SMTP id XAA03131; Thu, 1 May 1997 23:17:46 -0400 Date: Thu, 1 May 1997 23:17:46 -0400 (EDT) From: Dale Ghent Reply-To: Dale Ghent To: Ron Echeverri cc: stable@FreeBSD.ORG Subject: Re: Add LIVE Girls to your website! In-Reply-To: <199705020227.TAA04716@bofh.noc.best.net> Message-ID: X-President-Clinton: On Crack X-LART: Homelite Chainsaw MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-stable@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk On Thu, 1 May 1997, Ron Echeverri wrote: | rich@vegasone.com writes: | We will provide a Live Stripshow with chat for your website starting | at only $500 per month and NEVER more than $1600.00 a month!! (read | on for details) | | Is this in the ports directory? hmmm... adds a whole new meaning to `cvsup` -Dale G. From owner-freebsd-stable Thu May 1 20:32:44 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id UAA08723 for stable-outgoing; Thu, 1 May 1997 20:32:44 -0700 (PDT) Received: from dilbert.my.net (haplo@slip129-37-233-204.mi.us.ibm.net [129.37.233.204]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id UAA08694 for ; Thu, 1 May 1997 20:31:55 -0700 (PDT) Received: from localhost (haplo@localhost) by dilbert.my.net (8.8.5/8.8.5) with SMTP id WAA04180; Thu, 1 May 1997 22:31:30 -0500 (CDT) X-Authentication-Warning: dilbert.my.net: haplo owned process doing -bs Date: Thu, 1 May 1997 22:31:29 -0500 (CDT) From: Jordan Klein X-Sender: haplo@dilbert.my.net To: Ron Echeverri cc: stable@FreeBSD.ORG Subject: Re: Add LIVE Girls to your website! In-Reply-To: <199705020227.TAA04716@bofh.noc.best.net> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-stable@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk So, THAT'S where it came from? I was wondering... Jordan Klein Playing with FreeBSD and loving it jklein @ theramp . net haplo @ ibm . net http://www.theramp.net/jklein http://www.ibm.net Spam me and die Remove spaces from email addr to reply On Thu, 1 May 1997, Ron Echeverri wrote: > rich@vegasone.com writes: > We will provide a Live Stripshow with chat for your website starting > at only $500 per month and NEVER more than $1600.00 a month!! (read > on for details) > > Is this in the ports directory? > > rone > -- > Ron Echeverri Best Internet Usenet Administration rone@best.net > > From owner-freebsd-stable Fri May 2 01:19:40 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id BAA19881 for stable-outgoing; Fri, 2 May 1997 01:19:40 -0700 (PDT) Received: from xwing.wcape.gov.za (xwing.wcape.gov.za [164.151.101.253]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id BAA19849 for ; Fri, 2 May 1997 01:18:38 -0700 (PDT) Received: from jedi.wcape.gov.za. (jedi.wcape.gov.za [164.151.101.248]) by xwing.wcape.gov.za (8.8.5/8.6.12) with ESMTP id KAA18021; Fri, 2 May 1997 10:16:38 +0200 (SAT) Received: from JEDI/MAILQ by jedi.wcape.gov.za. (Mercury 1.21); 2 May 97 10:32:45 +0200 Received: from MAILQ by JEDI (Mercury 1.21); 2 May 97 10:32:36 +0200 From: "Sean White" Organization: CCS: Western Cape - OpeNET project To: "Justin T. Gibbs" Date: Fri, 2 May 1997 10:32:27 +0200 Subject: Re: Adaptec 2842 SCSI controller issues Reply-to: swhite@gov.za CC: "'freebsd-stable@freebsd.org'" Priority: normal X-mailer: Pegasus Mail for Windows (v2.40) Message-ID: <33E4FCF2125@jedi.wcape.gov.za.> Sender: owner-stable@freebsd.org X-Loop: FreeBSD.org Precedence: bulk On 29 Apr 97 at 18:07, Justin T. Gibbs wrote the following:- > Actually it probably means that I should be relying on the > serial eeprom on the 284X cards instead of using the values > left over in scratch ram. This is probalby screwing up the > parameters used in sync negotiation. I don't know if this will help at all, but I've run into similar problems on some systems that have AIC7870 controllers embedded on the motherboard with *no* SEEPROM. It appears as though SCSI bus disconnect is not being set - writes to the disk while the tape drive is busy guarantees a panic. I've experienced this with both 2.1.0 and 2.1.5 - not tried it with 2.1.7.1 yet. The only workaround I was able to do was to install an AHA2742 and stick all my devices on that. If you want me to recreate the crash and provide info on it, lemme know. Can do. Not a problem. :) Regards, Sean. +--------------------+----------------------+------------------------------+ | CCS: Western Cape | Tel: +27-21-462-2780 | DISCLAIMER: Any opinions ex- | | Private Bag X1 | Fax: +27-21-462-2791 | pressed above are *MINE*! If | | 8012 ROGGEBAAI | swhite@gov.za. | you want my boss's, ask him. | +--------------------+----------------------+------------------------------+ "Eagles may soar, but weasels don't get sucked into jet engine intakes" From owner-freebsd-stable Fri May 2 09:07:56 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id JAA08012 for stable-outgoing; Fri, 2 May 1997 09:07:56 -0700 (PDT) Received: from austin.polstra.com (austin.polstra.com [206.213.73.10]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id JAA08002 for ; Fri, 2 May 1997 09:07:36 -0700 (PDT) Received: (from jdp@localhost) by austin.polstra.com (8.8.5/8.8.5) id JAA21362; Fri, 2 May 1997 09:07:35 -0700 (PDT) To: freebsd-stable@freebsd.org Path: not-for-mail From: jdp@polstra.com (John Polstra) Newsgroups: polstra.freebsd.stable Subject: Re: Add LIVE Girls to your website! Date: 2 May 1997 09:07:34 -0700 Organization: Polstra & Co., Seattle, WA Lines: 10 Distribution: local Message-ID: <5kd3g6$krf@austin.polstra.com> References: <199705012351.QAA26484@hub.freebsd.org> Sender: owner-stable@freebsd.org X-Loop: FreeBSD.org Precedence: bulk In article <199705012351.QAA26484@hub.freebsd.org>, wrote: > We will provide a Live Stripshow with chat for your website starting > at only $500 per month and NEVER more than $1600.00 a month!! RTFM, Bozo. We already have that as part of the system. Try "man 1 strip". -- John Polstra jdp@polstra.com John D. Polstra & Co., Inc. Seattle, Washington USA "Self-knowledge is always bad news." -- John Barth From owner-freebsd-stable Fri May 2 09:08:38 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id JAA08131 for stable-outgoing; Fri, 2 May 1997 09:08:38 -0700 (PDT) Received: from desiree.teleport.com (desiree.teleport.com [192.108.254.21]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id JAA08116 for ; Fri, 2 May 1997 09:08:34 -0700 (PDT) Received: from linda.teleport.com (linda.teleport.com [192.108.254.12]) by desiree.teleport.com (8.8.5/8.7.3) with ESMTP id JAA16122; Fri, 2 May 1997 09:07:46 -0700 (PDT) From: Mostyn/Annabella Received: (from mrl@localhost) by linda.teleport.com (8.8.5/8.8.4) id JAA03377; Fri, 2 May 1997 09:07:45 -0700 (PDT) Message-Id: <199705021607.JAA03377@linda.teleport.com> Subject: No src-2.2.0259.gz in CTM? To: freebsd-stable@freebsd.org Date: Fri, 2 May 1997 09:07:45 -0700 (PDT) Cc: mrl@teleport.com (Mostyn/Annabella), rkw@dataplex.net X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-stable@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Hi, I noticed that src-2.2.0260.gz is in the CTM area but not src-2.2.0259.gz? Is this going to cause trouble? Regards, Mostyn From owner-freebsd-stable Fri May 2 09:19:52 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id JAA08710 for stable-outgoing; Fri, 2 May 1997 09:19:52 -0700 (PDT) Received: from dopey.pathlink.com (dopey.pathlink.com [204.30.237.226]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id JAA08705 for ; Fri, 2 May 1997 09:19:49 -0700 (PDT) Received: from dvl-1.pathlink.com (dvl-1.pathlink.com [204.30.237.241]) by dopey.pathlink.com (8.8.5/8.8.5) with SMTP id JAA13147; Fri, 2 May 1997 09:28:36 -0700 (PDT) Message-Id: <1.5.4.32.19970502091932.006d6bc4@dopey.pathlink.com> X-Sender: kachun@dopey.pathlink.com X-Mailer: Windows Eudora Light Version 1.5.4 (32) Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Fri, 02 May 1997 09:19:32 +0000 To: dg@root.com From: Kachun Lee Subject: Re: kmem_map full with 2.2beta to 2.2-releng-970422 Cc: freebsd-stable@freebsd.org Sender: owner-stable@freebsd.org X-Loop: FreeBSD.org Precedence: bulk At 02:05 PM 5/1/97 -0700, you wrote: >>clients, and our NFS servers have quite a lot of these: >> >> /kernel: nfsd send error 55 >> >>They does not seem cause any ill effect. I assume 55=ENOBUFS No buffer space >>available. But I thought with the unified VM/Buffer, there are no MAXBUFFER >>setting any more. Do you know what parameter I should adjust? > > ...that one indicates that network buffers were depleted. This usually >happens when the interface queue reaches it's limit of 50 queued packets, >although it can also happen if the global pool of network buffers is deleted. >The most common cause of this error is a stuck network device driver - some >NICs, for example will stop sending if the cable is unplugged. It can also >happen on PPP/SLIP lines if hardware flow control (CTS) stays de-asserted too >long. I kind of know it is probably related to the 'de' drivers and the 50+% collicion rate. I am almost there giving up waiting for the 'new' de driver to make it to stable. I should just throw away all the SMC cards, get some Intel ones and use your driver. Thank again for the info. From owner-freebsd-stable Fri May 2 09:33:09 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id JAA09195 for stable-outgoing; Fri, 2 May 1997 09:33:09 -0700 (PDT) Received: from shrimp.dataplex.net (shrimp.dataplex.net [208.2.87.3]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id JAA09176 for ; Fri, 2 May 1997 09:32:40 -0700 (PDT) Received: from [204.69.236.50] (GATEWAY.SKIPSTONE.COM [198.214.10.129]) by shrimp.dataplex.net (8.8.5/8.8.5) with SMTP id LAA22241; Fri, 2 May 1997 11:32:31 -0500 (CDT) Date: 2 May 97 11:33:20 -0500 Subject: Re: No src-2.2.0259.gz in CTM? From: "Richard Wackerbarth" To: "Mostyn/Annabella" Cc: freebsd-stable@FreeBSD.ORG X-Mailer: Cyberdog/2.0 MIME-Version: 1.0 Message-Id: Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-stable@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk On Fri, May 2, 1997 11:07 AM, Mostyn/Annabella wrote: >Hi, > >I noticed that src-2.2.0260.gz is in the CTM area but not >src-2.2.0259.gz? Is this going to cause trouble? That depend on whether or not you want to update beyond 0258 :-) 260 and later will "queue up", but not get applied until 259 is available. This is caused by problems in the FreeBSD mail system. Continue to expect problems until they get it repaired. I have manually forced src-2.2.0259.gz to be retransmitted. From owner-freebsd-stable Fri May 2 14:50:28 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id OAA25300 for stable-outgoing; Fri, 2 May 1997 14:50:28 -0700 (PDT) Received: from gatekeeper.tsc.tdk.com (root@gatekeeper.tsc.tdk.com [207.113.159.21]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id OAA25295 for ; Fri, 2 May 1997 14:50:26 -0700 (PDT) Received: from sunrise.gv.tsc.tdk.com (root@sunrise.gv.tsc.tdk.com [192.168.241.191]) by gatekeeper.tsc.tdk.com (8.8.4/8.8.4) with ESMTP id OAA00626; Fri, 2 May 1997 14:50:14 -0700 (PDT) Received: from salsa.gv.tsc.tdk.com (salsa.gv.tsc.tdk.com [192.168.241.194]) by sunrise.gv.tsc.tdk.com (8.8.5/8.8.5) with ESMTP id OAA10203; Fri, 2 May 1997 14:50:13 -0700 (PDT) Received: (from gdonl@localhost) by salsa.gv.tsc.tdk.com (8.8.5/8.8.5) id OAA25847; Fri, 2 May 1997 14:50:11 -0700 (PDT) From: Don Lewis Message-Id: <199705022150.OAA25847@salsa.gv.tsc.tdk.com> Date: Fri, 2 May 1997 14:50:11 -0700 In-Reply-To: Kachun Lee "Re: kmem_map full with 2.2beta to 2.2-releng-970422" (May 2, 9:19am) X-Mailer: Mail User's Shell (7.2.6 alpha(3) 7/19/95) To: Kachun Lee , dg@root.com Subject: Re: kmem_map full with 2.2beta to 2.2-releng-970422 Cc: freebsd-stable@FreeBSD.ORG Sender: owner-stable@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk On May 2, 9:19am, Kachun Lee wrote: } Subject: Re: kmem_map full with 2.2beta to 2.2-releng-970422 } } I kind of know it is probably related to the 'de' drivers and the 50+% } collicion rate. I am almost there giving up waiting for the 'new' de driver } to make it to stable. I should just throw away all the SMC cards, get some } Intel ones and use your driver. So far as I know the collision statistics reporting problem was fixed months ago. Here are the statistics on our 2.1.6+ news server. Name Mtu Network Address Ipkts Ierrs Opkts Oerrs Coll de0 1500 00.00.c0.ae.d2.e4 31402509 2 33107898 2 3589950 This looks like a 10.8% collision rate to me. The network it's connected to is not bridged or switched, and has 60 to 80 hosts on it. --- Truck From owner-freebsd-stable Fri May 2 17:16:03 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id RAA01904 for stable-outgoing; Fri, 2 May 1997 17:16:03 -0700 (PDT) Received: from tok.qiv.com ([204.214.141.211]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id RAA01840 for ; Fri, 2 May 1997 17:15:49 -0700 (PDT) Received: (from uucp@localhost) by tok.qiv.com (8.8.5/8.8.5) with UUCP id TAA09523; Fri, 2 May 1997 19:15:42 -0500 (CDT) Received: from localhost (jdn@localhost) by acp.qiv.com (8.8.5/8.8.5) with SMTP id SAA00432; Fri, 2 May 1997 18:56:40 -0500 (CDT) X-Authentication-Warning: acp.qiv.com: jdn owned process doing -bs Date: Fri, 2 May 1997 18:56:39 -0500 (CDT) From: "Jay D. Nelson" To: John Polstra cc: freebsd-stable@FreeBSD.ORG Subject: Re: Add LIVE Girls to your website! In-Reply-To: <5kd3g6$krf@austin.polstra.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-stable@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Of course, if you have emacs installed, you also get the man pages for sex(6) and celibacy(1). -- Jay On 2 May 1997, John Polstra wrote: ->In article <199705012351.QAA26484@hub.freebsd.org>, wrote: -> ->> We will provide a Live Stripshow with chat for your website starting ->> at only $500 per month and NEVER more than $1600.00 a month!! -> ->RTFM, Bozo. We already have that as part of the system. Try "man 1 strip". ->-- -> John Polstra jdp@polstra.com -> John D. Polstra & Co., Inc. Seattle, Washington USA -> "Self-knowledge is always bad news." -- John Barth -> From owner-freebsd-stable Fri May 2 18:19:50 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id SAA04285 for stable-outgoing; Fri, 2 May 1997 18:19:50 -0700 (PDT) Received: from netcomsv.netcom.com (uucp12.netcom.com [163.179.3.12]) by hub.freebsd.org (8.8.5/8.8.5) with SMTP id SAA04276 for ; Fri, 2 May 1997 18:19:47 -0700 (PDT) Received: from iecgate.UUCP by netcomsv.netcom.com with UUCP (8.6.12/SMI-4.1) id SAA20973; Fri, 2 May 1997 18:16:44 -0700 Received: from madmax.iecorp.com by iecgate.iecorp.com (4.1/SMI-4.1) id AA28765; Fri, 2 May 97 18:07:37 PDT Received: by madmax.iecorp.com (4.1/SMI-4.1) id AA08100; Fri, 2 May 97 18:07:24 PDT From: bartleym@iecorp.com (Matt Bartley) Message-Id: <9705030107.AA08100@madmax.iecorp.com> Subject: Re: Add LIVE Girls to your website! To: freebsd-stable@freebsd.org Date: Fri, 2 May 1997 18:07:24 -0700 (PDT) In-Reply-To: from "Jay D. Nelson" at May 2, 97 06:56:39 pm X-Mailer: ELM [version 2.4 PL23] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-stable@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Sorry for continuing this thread; perhaps someone at a backbone site should email these spammers a copy of one of the source code CTM "A" deltas. They did post to FreeBSD-stable after all - they must be interested... :-) From owner-freebsd-stable Fri May 2 18:51:17 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id SAA05135 for stable-outgoing; Fri, 2 May 1997 18:51:17 -0700 (PDT) Received: from dopey.pathlink.com (dopey.pathlink.com [204.30.237.226]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id SAA05130 for ; Fri, 2 May 1997 18:51:13 -0700 (PDT) Received: from dvl-1.pathlink.com (dvl-1.pathlink.com [204.30.237.241]) by dopey.pathlink.com (8.8.5/8.8.5) with SMTP id TAA14011; Fri, 2 May 1997 19:00:04 -0700 (PDT) Message-Id: <1.5.4.32.19970502185049.007589a0@dopey.pathlink.com> X-Sender: kachun@dopey.pathlink.com X-Mailer: Windows Eudora Light Version 1.5.4 (32) Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Fri, 02 May 1997 18:50:49 +0000 To: Don Lewis From: Kachun Lee Subject: Re: kmem_map full with 2.2beta to 2.2-releng-970422 Cc: freebsd-stable@freebsd.org Sender: owner-stable@freebsd.org X-Loop: FreeBSD.org Precedence: bulk At 02:50 PM 5/2/97 -0700, you wrote: >On May 2, 9:19am, Kachun Lee wrote: >} Subject: Re: kmem_map full with 2.2beta to 2.2-releng-970422 >} >} I kind of know it is probably related to the 'de' drivers and the 50+% >} collicion rate. I am almost there giving up waiting for the 'new' de driver >} to make it to stable. I should just throw away all the SMC cards, get some >} Intel ones and use your driver. > >So far as I know the collision statistics reporting problem was fixed >months ago. Here are the statistics on our 2.1.6+ news server. > >Name Mtu Network Address Ipkts Ierrs Opkts Oerrs Coll >de0 1500 00.00.c0.ae.d2.e4 31402509 2 33107898 2 3589950 > >This looks like a 10.8% collision rate to me. The network it's connected to >is not bridged or switched, and has 60 to 80 hosts on it. Oh! I don't think that is a bug... I believe we are actually getting 50+% collision rate :-( Well, our systems in aggregate are putting about 40 mbit/sec on the LAN though. I always assume the high collision rate is because of half duplex. ------- 6:52PM up 6 days, 12:13, 1 user, load averages: 0.00, 0.00, 0.00 Name Mtu Network Address Ipkts Ierrs Opkts Oerrs Coll de0 1500 00.00.c0.cd.f9.f2 791136217 0 956939840 441395 493282445 -------- Best regards From owner-freebsd-stable Sat May 3 18:22:04 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id SAA28779 for stable-outgoing; Sat, 3 May 1997 18:22:04 -0700 (PDT) Received: from doleh.com (mail.doleh.com [192.231.91.254]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id SAA28774 for ; Sat, 3 May 1997 18:22:00 -0700 (PDT) Received: (from yaser@localhost) by doleh.com (8.8.5/96.03.01.0) id VAA03635 for freebsd-stable@FreeBSD.ORG; Sat, 3 May 1997 21:22:09 -0400 (EDT) From: "Yaser K. Doleh" Message-Id: <199705040122.VAA03635@doleh.com> Subject: Where is netiso includes To: freebsd-stable@FreeBSD.ORG Date: Sat, 3 May 1997 21:22:08 -0400 (EDT) X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-stable@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk I tried to upgrade from 2.1.7 to 2.2 using cvsup. For some reason I am missing the following the directory /usr/src/sys/netiso This is my supfile *default tag=RELENG_2_2 *default host=cvsup.FreeBSD.org *default prefix=/usr *default base=/usr/local/etc/cvsup *default release=cvs delete use-rel-suffix compress src-base src-bin src-contrib src-etc src-gnu src-include src-lib src-libexec src-release src-share src-sys src-tools src-usrbin src-usrsbin cvs-crypto This is error I get when I try make world mkdep -f .depend -a /usr/src/sbin/ifconfig/ifconfig.c /usr/src/sbin/ifconfig/ifconfig.c:65: netiso/iso.h: No such file or directory /usr/src/sbin/ifconfig/ifconfig.c:66: netiso/iso_var.h: No such file or directory mkdep: compile failed. *** Error code 1 Stop. *** Error code 1 Stop. *** Error code 1 Stop. *** Error code 1 Stop. Very truly yours Yaser Doleh Yaser@Doleh.Com From owner-freebsd-stable Sat May 3 23:15:06 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id XAA11493 for stable-outgoing; Sat, 3 May 1997 23:15:06 -0700 (PDT) Received: from austin.polstra.com (austin.polstra.com [206.213.73.10]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id XAA11488 for ; Sat, 3 May 1997 23:15:03 -0700 (PDT) Received: from austin.polstra.com (jdp@localhost) by austin.polstra.com (8.8.5/8.8.5) with ESMTP id XAA04575; Sat, 3 May 1997 23:14:57 -0700 (PDT) Message-Id: <199705040614.XAA04575@austin.polstra.com> To: yaser@doleh.com Subject: Re: Where is netiso includes Newsgroups: polstra.freebsd.stable In-Reply-To: <199705040122.VAA03635@doleh.com> References: <199705040122.VAA03635@doleh.com> Organization: Polstra & Co., Seattle, WA Cc: stable@freebsd.org Date: Sat, 03 May 1997 23:14:57 -0700 From: John Polstra Sender: owner-stable@freebsd.org X-Loop: FreeBSD.org Precedence: bulk In article <199705040122.VAA03635@doleh.com>, Yaser K. Doleh wrote: > > I tried to upgrade from 2.1.7 to 2.2 using cvsup. For some reason I am missing > the following the directory /usr/src/sys/netiso [...] > This is error I get when I try make world > > mkdep -f .depend -a /usr/src/sbin/ifconfig/ifconfig.c > /usr/src/sbin/ifconfig/ifconfig.c:65: netiso/iso.h: No such file or directory > /usr/src/sbin/ifconfig/ifconfig.c:66: netiso/iso_var.h: No such file or I think you have some old .depend files or "obj" links somewhere. I'd recommend doing this before upgrading from 2.1.7 to 2.2: * rm -rf /usr/obj/* * find /usr/src -name obj -type l | xargs rm Then do your make world, and I bet it will work OK. John -- John Polstra jdp@polstra.com John D. Polstra & Co., Inc. Seattle, Washington USA "Self-knowledge is always bad news." -- John Barth