From owner-freebsd-proliant@FreeBSD.ORG Sun Feb 5 01:31:49 2006 Return-Path: X-Original-To: freebsd-proliant@freebsd.org Delivered-To: freebsd-proliant@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 77A4516A420; Sun, 5 Feb 2006 01:31:49 +0000 (GMT) (envelope-from scrappy@hub.org) Received: from hub.org (hub.org [200.46.204.220]) by mx1.FreeBSD.org (Postfix) with ESMTP id 019C543D46; Sun, 5 Feb 2006 01:31:48 +0000 (GMT) (envelope-from scrappy@hub.org) Received: from localhost (unknown [200.46.204.144]) by hub.org (Postfix) with ESMTP id 0D63362C8D4; Sat, 4 Feb 2006 21:31:48 -0400 (AST) Received: from hub.org ([200.46.204.220]) by localhost (av.hub.org [200.46.204.144]) (amavisd-new, port 10024) with ESMTP id 85471-08; Sat, 4 Feb 2006 21:31:47 -0400 (AST) Received: from ganymede.hub.org (blk-222-82-85.eastlink.ca [24.222.82.85]) by hub.org (Postfix) with ESMTP id 8A64F62C856; Sat, 4 Feb 2006 21:31:47 -0400 (AST) Received: by ganymede.hub.org (Postfix, from userid 1000) id 5AA7947363; Sat, 4 Feb 2006 21:31:51 -0400 (AST) Received: from localhost (localhost [127.0.0.1]) by ganymede.hub.org (Postfix) with ESMTP id 59D2446D04; Sat, 4 Feb 2006 21:31:51 -0400 (AST) Date: Sat, 4 Feb 2006 21:31:51 -0400 (AST) From: "Marc G. Fournier" To: Ivan Voras In-Reply-To: <43E52A89.4020700@fer.hr> Message-ID: <20060204213028.D3207@ganymede.hub.org> References: <43E52A89.4020700@fer.hr> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII; format=flowed X-Virus-Scanned: by amavisd-new at hub.org Cc: stable@freebsd.org, freebsd-proliant@freebsd.org Subject: Re: Compaq Proliant CISS slow writes X-BeenThere: freebsd-proliant@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: "Technical discussion of FreeBSD on HP ProLiant server platforms." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 05 Feb 2006 01:31:49 -0000 Not sure if this is standard, but on an idle machine (ie. just set it up, nothing running on it yet), with RAID1+0 across 4 drives: Writing the 853 Megabyte file, 'iozone.tmp'...15.656250 seconds Reading the file...3.921875 seconds IOZONE performance measurements: 57129601 bytes/second for writing the file 228063191 bytes/second for reading the file The server is a new DL360 G4p, drives are SAS ... On Sat, 4 Feb 2006, Ivan Voras wrote: > I need to get a Proliant machine with 2 P3 processors running FreeBSD 6. I > don't know much about the machine, I think it's ML 380 G2 or close to that, > but I have physical access. So far, everything is fine (once the inability to > boot from CD-ROM is circumvented), except one "detail": horrible write > performance on its CISS 5 RAID5 array. I get ~75MB/s burst (large blocks) > reads, and only 5MB/s burst writes. I know how RAID5 works, but still, this > is bad. The machine has been running Linux before this and performance was Ok > - I didn't benchmark it but the "feeling" when working on it was normal, > while on FreeBSD it's noticably slow in mixed read/write load. > > Is there anything I can try to improve this? In the verbose boot log there's > a line that says the controller supports "simple, performant and MEMQ" modes, > and the one that's used is "simple" - does this have any influence? If so, > how to change it? > > Thanks! > ---- Marc G. Fournier Hub.Org Networking Services (http://www.hub.org) Email: scrappy@hub.org Yahoo!: yscrappy ICQ: 7615664 From owner-freebsd-proliant@FreeBSD.ORG Sun Feb 5 10:42:20 2006 Return-Path: X-Original-To: freebsd-proliant@freebsd.org Delivered-To: freebsd-proliant@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id A08BA16A422; Sun, 5 Feb 2006 10:42:20 +0000 (GMT) (envelope-from ivoras@fer.hr) Received: from pinus.cc.fer.hr (pinus.cc.fer.hr [161.53.73.18]) by mx1.FreeBSD.org (Postfix) with ESMTP id C48E443D48; Sun, 5 Feb 2006 10:42:17 +0000 (GMT) (envelope-from ivoras@fer.hr) Received: from [161.53.72.113] (lara.cc.fer.hr [161.53.72.113]) by pinus.cc.fer.hr (8.12.2/8.12.2) with ESMTP id k15Ag7Fx013123; Sun, 5 Feb 2006 11:42:08 +0100 (MET) Message-ID: <43E5D652.6010205@fer.hr> Date: Sun, 05 Feb 2006 11:41:22 +0100 From: Ivan Voras User-Agent: Mozilla Thunderbird 1.0.6 (X11/20050921) X-Accept-Language: en-us, en MIME-Version: 1.0 To: "Marc G. Fournier" References: <43E52A89.4020700@fer.hr> <20060204213028.D3207@ganymede.hub.org> In-Reply-To: <20060204213028.D3207@ganymede.hub.org> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Cc: stable@freebsd.org, freebsd-proliant@freebsd.org Subject: Re: Compaq Proliant CISS slow writes X-BeenThere: freebsd-proliant@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: "Technical discussion of FreeBSD on HP ProLiant server platforms." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 05 Feb 2006 10:42:20 -0000 Marc G. Fournier wrote: > > Not sure if this is standard, but on an idle machine (ie. just set it > up, nothing running on it yet), with RAID1+0 across 4 drives: > > Writing the 853 Megabyte file, 'iozone.tmp'...15.656250 seconds > Reading the file...3.921875 seconds > > IOZONE performance measurements: > 57129601 bytes/second for writing the file > 228063191 bytes/second for reading the file I could live with 4x slower writes - that would get me ~18MB/s; I'd settle for ~10MB/s as the server is on a 100Mbit/s network. If it's the standard situation then ok, it's "higher power" but somehow I doubt 3 10kRPM SCSI drives in RAID5 should be this slow. Toying with mount options doesn't help. (btw. everything is fine on the attached desktop-grade IDE disk [very recent model]: 45MB/s writes, 50MB/s reads) From owner-freebsd-proliant@FreeBSD.ORG Mon Feb 6 23:57:58 2006 Return-Path: X-Original-To: freebsd-proliant@freebsd.org Delivered-To: freebsd-proliant@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 461DB16A420 for ; Mon, 6 Feb 2006 23:57:58 +0000 (GMT) (envelope-from freebsd@henriklidstrom.se) Received: from pne-smtpout2-sn1.fre.skanova.net (pne-smtpout2-sn1.fre.skanova.net [81.228.11.159]) by mx1.FreeBSD.org (Postfix) with ESMTP id BB4C943D4C for ; Mon, 6 Feb 2006 23:57:57 +0000 (GMT) (envelope-from freebsd@henriklidstrom.se) Received: from mailgw.henriklidstrom.se (83.250.79.110) by pne-smtpout2-sn1.fre.skanova.net (7.2.069.1) id 43E6239C000722A6; Tue, 7 Feb 2006 00:57:56 +0100 Received: from localhost (localhost.henriklidstrom.se [127.0.0.1]) by mailgw.henriklidstrom.se (Postfix) with ESMTP id B53D2C0D7; Tue, 7 Feb 2006 00:57:55 +0100 (CET) Received: from mailgw.henriklidstrom.se ([127.0.0.1]) by localhost (mailgw.henriklidstrom.se [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 20825-09; Tue, 7 Feb 2006 00:57:42 +0100 (CET) Received: from [192.168.0.2] (henkezan.henriklidstrom.se [192.168.0.2]) by mailgw.henriklidstrom.se (Postfix) with ESMTP id B23B8B8E0; Tue, 7 Feb 2006 00:57:42 +0100 (CET) Message-ID: <43E7E27E.9060501@henriklidstrom.se> Date: Tue, 07 Feb 2006 00:57:50 +0100 From: =?ISO-8859-1?Q?Henrik_Lidstr=F6m?= User-Agent: Thunderbird 1.5 (Windows/20051201) MIME-Version: 1.0 To: greg@wholemeal.net Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-Virus-Scanned: amavisd-new at henriklidstrom.se Cc: freebsd-proliant@freebsd.org Subject: Re: hpasmcli locks up a DL380G3 X-BeenThere: freebsd-proliant@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: "Technical discussion of FreeBSD on HP ProLiant server platforms." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 06 Feb 2006 23:57:58 -0000 > > Hello > > First of all many thanks for the port of these tools to FreeBSD, > something I'd dreamt about for a long time. The lack of RAID > management tools for FreeBSD on DLxxx hardware was quite crippling. > OK, they have been removed but at least I know they are actively > being worked on. > > Sadly though I've been experiencing problems with the hpasmcli > utility on a number of machines, which today culminated in the > locking up of a production webserver. I was taking what seem to be a > small risk running it in that environment but initial tests had shown > it to be quite stable. > > The box that locked up was a DL380G3 which reports as below, running > FreebSD 5.3R-p5 > > System : ProLiant DL380 G3 > Serial No. : xxxxxxxxxxxx > ROM version : P29 05/01/2004 > iLo present : Yes > Embedded NICs : 2 > NIC1 MAC: xxxxxxxxxxxxxxxxxx > NIC2 MAC: xxxxxxxxxxxxxxxxxx > > Processor: 0 > Name : Intel Xeon > Stepping : 9 > Speed : 2800 MHz > Bus : 400 MHz > Socket : 1 > Level2 Cache : 512 KBytes > Status : Ok > > Processor total : 1 > > Memory installed : 1024 MBytes > ECC supported : Yes > > What appears to happen is thus: > > 1. For a period of days or weeks the hpasmcli and hpasmd work just fine. > 2. After some period of time, in the order of weeks, when hpasmcli is > run it will just print: > > HP management CLI for FreeBSD (v7.22) > Copyright 2003,2005 Hewlett-Packard Development Group, L.P. > > ------------------------------------------------------------------------ > -- > NOTE: Some hpasmcli commands may not be supported on all Proliant > servers. > Type 'help' to get a list of all top level commands. > ------------------------------------------------------------------------ > -- > > and then hang, without showing the hpasmcli> prompt for input. In the > past it's been possible to Ctrl-C break out. Restarting the hpasmd > daemon will see hpasmcli working fine again. Given the usefulness of > the utility I've been willing to live with that. However today > running hpasmcli seems to have instantly locked the box, both at the > console and via ssh. I've been using hpasmcli since it's public release. > > The other HP boxes I have include two DL360G4s with FreeBSD 5.4-p1 > and a DL360G3 with FreeBSD 5.3-p4 . They have shown the hpasmcli > hanging behaviour, but never locked up hard. > > Unfortunately all the HP boxes I have are currently in production and > I can't run any further diagnostics, traces etc. > > Has anyone else seen this behaviour? Is a fix likely? > > best wishes > > greg. > Hi! Sorry for being late on this one, found this browsing around. Yes, I have had ONE machine lock up on me once. And older HP Proliant DL380G1 UP. Just as you describe, it had been working great for a couple of weeks, then suddenly when starting hpasmcli it froze. Couldn't even ping the machine. This particualar machine really is not doing anything, and as I belive it still is running (Moved/changed job) and I could probably recreate the lockup. I still have access to this machine, so if anyone want me to try something, I can do it. The machine is 600km away from me now, so if lockup occurs it can take some time to get it powercycled though. Oh, 5.3 or 5.4 as I recall. Have you seen any other lockup Greg? /Henrik PS. Yes, these tools have been _very_ useful otherwise. Excellent job! From owner-freebsd-proliant@FreeBSD.ORG Tue Feb 7 10:13:13 2006 Return-Path: X-Original-To: freebsd-proliant@freebsd.org Delivered-To: freebsd-proliant@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 893CB16A420 for ; Tue, 7 Feb 2006 10:13:13 +0000 (GMT) (envelope-from greg@wholemeal.net) Received: from warprecords.com (dsl-212-135-210-82.dsl.easynet.co.uk [212.135.210.82]) by mx1.FreeBSD.org (Postfix) with ESMTP id B110843D49 for ; Tue, 7 Feb 2006 10:13:11 +0000 (GMT) (envelope-from greg@wholemeal.net) Received: from [192.168.100.33] (HELO [192.168.0.2]) by warprecords.com (CommuniGate Pro SMTP 4.2.10) with ESMTP-TLS id 5916477; Tue, 07 Feb 2006 10:13:10 +0000 In-Reply-To: <43E7E27E.9060501@henriklidstrom.se> References: <43E7E27E.9060501@henriklidstrom.se> Mime-Version: 1.0 (Apple Message framework v746.2) Content-Type: text/plain; charset=ISO-8859-1; delsp=yes; format=flowed Message-Id: <214C059F-21D7-4827-801A-BD823D9F7CF7@wholemeal.net> Content-Transfer-Encoding: quoted-printable From: Greg Eden Date: Tue, 7 Feb 2006 10:13:10 +0000 To: =?ISO-8859-1?Q?Henrik_Lidstr=F6m?= X-Mailer: Apple Mail (2.746.2) Cc: freebsd-proliant@freebsd.org Subject: Re: hpasmcli locks up a DL380G3 X-BeenThere: freebsd-proliant@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: "Technical discussion of FreeBSD on HP ProLiant server platforms." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 07 Feb 2006 10:13:13 -0000 On 6 Feb 2006, at 23:57, Henrik Lidstr=F6m wrote: >> >> Hello >> >> First of all many thanks for the port of these tools to FreeBSD, =20 >> something I'd dreamt about for a long time. The lack of RAID =20 >> management tools for FreeBSD on DLxxx hardware was quite =20 >> crippling. OK, they have been removed but at least I know they =20 >> are actively being worked on. >> >> Sadly though I've been experiencing problems with the hpasmcli =20 >> utility on a number of machines, which today culminated in the =20 >> locking up of a production webserver. I was taking what seem to be =20= >> a small risk running it in that environment but initial tests had =20= >> shown it to be quite stable. >> >> The box that locked up was a DL380G3 which reports as below, =20 >> running FreebSD 5.3R-p5 >> >> System : ProLiant DL380 G3 >> Serial No. : xxxxxxxxxxxx >> ROM version : P29 05/01/2004 >> iLo present : Yes >> Embedded NICs : 2 >> NIC1 MAC: xxxxxxxxxxxxxxxxxx >> NIC2 MAC: xxxxxxxxxxxxxxxxxx >> >> Processor: 0 >> Name : Intel Xeon >> Stepping : 9 >> Speed : 2800 MHz >> Bus : 400 MHz >> Socket : 1 >> Level2 Cache : 512 KBytes >> Status : Ok >> >> Processor total : 1 >> >> Memory installed : 1024 MBytes >> ECC supported : Yes >> >> What appears to happen is thus: >> >> 1. For a period of days or weeks the hpasmcli and hpasmd work just =20= >> fine. >> 2. After some period of time, in the order of weeks, when hpasmcli =20= >> is run it will just print: >> >> HP management CLI for FreeBSD (v7.22) >> Copyright 2003,2005 Hewlett-Packard Development Group, L.P. >> >> ---------------------------------------------------------------------=20= >> --- -- >> NOTE: Some hpasmcli commands may not be supported on all Proliant =20= >> servers. >> Type 'help' to get a list of all top level commands. >> ---------------------------------------------------------------------=20= >> --- -- >> >> and then hang, without showing the hpasmcli> prompt for input. In =20 >> the past it's been possible to Ctrl-C break out. Restarting the =20 >> hpasmd daemon will see hpasmcli working fine again. Given the =20 >> usefulness of the utility I've been willing to live with that. =20 >> However today running hpasmcli seems to have instantly locked the =20= >> box, both at the console and via ssh. I've been using hpasmcli =20 >> since it's public release. >> >> The other HP boxes I have include two DL360G4s with FreeBSD 5.4-=20 >> p1 and a DL360G3 with FreeBSD 5.3-p4 . They have shown the =20 >> hpasmcli hanging behaviour, but never locked up hard. >> >> Unfortunately all the HP boxes I have are currently in production =20 >> and I can't run any further diagnostics, traces etc. >> >> Has anyone else seen this behaviour? Is a fix likely? >> >> best wishes >> >> greg. >> > > Hi! > > Sorry for being late on this one, found this browsing around. > > Yes, I have had ONE machine lock up on me once. > And older HP Proliant DL380G1 UP. Just as you describe, it had been =20= > working great for > a couple of weeks, then suddenly when starting hpasmcli it froze. > Couldn't even ping the machine. > > This particualar machine really is not doing anything, and as I belive > it still is running (Moved/changed job) and I could probably =20 > recreate the lockup. > I still have access to this machine, so if anyone want me to try =20 > something, I can do it. > The machine is 600km away from me now, so if lockup occurs it can =20 > take some time to get it > powercycled though. > > Oh, 5.3 or 5.4 as I recall. > > Have you seen any other lockup Greg? I haven't tempted fate that way yet. I always restart the hpasmd =20 before using the client on a machine. This seems to avoid the problem. Thanks for responding to my mail, you're the third person to confirm =20 the problem, which given that it locks the machine up hard, is a very =20= serious one. best. greg.=