From owner-freebsd-questions@FreeBSD.ORG Mon Nov 8 19:01:07 2004 Return-Path: Delivered-To: freebsd-questions@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 3320516A4CE for ; Mon, 8 Nov 2004 19:01:07 +0000 (GMT) Received: from ms-smtp-04-eri0.ohiordc.rr.com (ms-smtp-04-smtplb.ohiordc.rr.com [65.24.5.138]) by mx1.FreeBSD.org (Postfix) with ESMTP id 863A143D2D for ; Mon, 8 Nov 2004 19:01:04 +0000 (GMT) (envelope-from joe@thefrys.com) Received: from lucy (cpe-24-33-240-92.twmi.rr.com [24.33.240.92]) iA8J11HH007059 for ; Mon, 8 Nov 2004 14:01:02 -0500 (EST) From: "Joseph H. Fry" To: Date: Mon, 8 Nov 2004 14:01:39 -0500 Message-ID: <002601c4c5c5$61306fd0$f100a8c0@lucy> X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook, Build 10.0.2616 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2900.2180 Importance: Normal X-Virus-Scanned: Symantec AntiVirus Scan Engine Subject: Slow writes using hardware ata raid on older server X-BeenThere: freebsd-questions@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: User questions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 08 Nov 2004 19:01:07 -0000 I've finally got a clean updated stable FreeBSD box, thank you all who have invested your time creating documentation that makes it so easy for us newbies. Anyway, I had gentoo on the machine for about 10 minutes... Had a problem with writing to my Megaraid i4 controller... It was terribly slow (<1MB/sec)! So I figured I'd try FreeBSD and see how that goes. I'm no expert, but as far as I can tell Bonnie output (below), things are not much better. I only tested with a 100MB file ($ bonnie++ -s 100) because I need to head to work and didn't want to wait for it to go through a whole 1-2GB file. Also below you will find the output from top while bonnie++ was "writing a byte at a time" and "writing intelligently". I only include these because I'm not sure how to read them properly for multiple processors. >From what I can tell, during the byte at a time write the processor is pegged, but only about 1/4 used during the intelligent write. Anyway, I'm stuck now with no idea what to do to fix this problem! Any guidance would be appreciated. (expecially help understanding what the different values in the bonnie++ output mean) System: Dual PII 233 (smp in kernel) on intel server board 128MB ECC SDRAM LSI MegaRaid i4 ATA RAID Controller (uses amr) in pci slot 1 Onboard Intel e100PRO 3Com 3c595 Adaptec SCSI controller onboard, and one in pci slot (I don't have the cable I need to connect my CDROM to the onboard SCSI controller so I added one temporarily) ------Sequential Output------ --Sequential Input- --Random- -Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks-- Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP /sec %CP 100M 18 99 2172 8 2105 10 54 99 29052 68 373.4 95 585ms 503ms 613ms 204ms 60668us 3941ms ------Sequential Create------ --------Random Create-------- -Create-- --Read--- -Delete-- -Create-- --Read--- -Delete-- Files /sec %CP /sec %CP /sec %CP /sec %CP /sec %CP /sec %CP 16 1204 42 5952 96 5136 99 1024 47 6246 99 4649 88 2547ms 104ms 427us 2682ms 27976us 201ms 1.93c,1.93c,fileserv.thefrys.local,1,1099936265,100M,,18,99,2172,8,2105, 10,54,99,29052,68,373.4,95,16,,,,,1204,42,5952,96,5136,99,1024,47,6246,9 9,4649,88,585ms,503ms,613ms,204ms,60668us,3941ms,2547ms,104ms,427us,2682 ms,27976us,201ms While writing byte at a time--------------------------------------------------- last pid: 2317; load averages: 0.81, 0.30, 0.17 up 0+00:54:58 13:49:24 30 processes: 2 running, 27 sleeping, 1 stopped CPU states: 4.1% user, 0.0% nice, 45.7% system, 0.3% interrupt, 49.8% idle Mem: 16M Active, 368K Inact, 27M Wired, 22M Buf, 77M Free Swap: 510M Total, 100K Used, 510M Free PID USERNAME PRI NICE SIZE RES STATE C TIME WCPU CPU COMMAND 2316 jfry 114 0 2380K 1208K CPU1 1 1:37 98.93% 98.19% bonnie++ 2265 jfry 96 0 2380K 1112K STOP 1 0:05 0.00% 0.00% bonnie++ 469 jfry 96 0 6092K 2248K select 0 0:01 0.00% 0.00% sshd 2317 root 96 0 2300K 1368K CPU0 0 0:01 0.00% 0.00% top 2260 jfry 96 0 6092K 2308K select 1 0:00 0.00% 0.00% sshd 2261 jfry 8 0 3092K 1588K wait 0 0:00 0.00% 0.00% bash 379 root 96 0 3440K 2100K select 0 0:00 0.00% 0.00% sendmail 474 root 20 0 2296K 1600K pause 0 0:00 0.00% 0.00% csh 462 root 4 0 6112K 2152K sbwait 0 0:00 0.00% 0.00% sshd 2257 root 4 0 6112K 2272K sbwait 0 0:00 0.00% 0.00% sshd . . . While writing intelligently---------------------------------------------------- last pid: 2321; load averages: 0.59, 0.41, 0.23 up 0+00:56:44 13:51:10 30 processes: 2 running, 27 sleeping, 1 stopped CPU states: 0.0% user, 0.0% nice, 4.7% system, 0.2% interrupt, 95.2% idle Mem: 16M Active, 59M Inact, 36M Wired, 6656K Cache, 22M Buf, 2448K Free Swap: 510M Total, 100K Used, 510M Free PID USERNAME PRI NICE SIZE RES STATE C TIME WCPU CPU COMMAND 2316 jfry 96 0 2380K 1200K CPU1 1 2:48 22.07% 22.07% bonnie++ 2265 jfry 96 0 2380K 1112K STOP 1 0:05 0.00% 0.00% bonnie++ 469 jfry 96 0 6092K 2248K select 1 0:01 0.00% 0.00% sshd 2260 jfry 96 0 6092K 2308K select 1 0:00 0.00% 0.00% sshd 2261 jfry 8 0 3092K 1588K wait 0 0:00 0.00% 0.00% bash 379 root 96 0 3440K 2100K select 1 0:00 0.00% 0.00% sendmail 474 root 20 0 2296K 1600K pause 1 0:00 0.00% 0.00% csh 462 root 4 0 6112K 2152K sbwait 0 0:00 0.00% 0.00% sshd 2257 root 4 0 6112K 2272K sbwait 0 0:00 0.00% 0.00% sshd 2321 root 96 0 2300K 1352K CPU0 1 0:00 0.00% 0.00% top . . . Sorry for the long post, I didn't want to leave anything important out. ---- Joe Fry ---- (joe@thefrys.com)