From owner-freebsd-scsi@FreeBSD.ORG Mon Jan 5 11:06:59 2009 Return-Path: Delivered-To: freebsd-scsi@FreeBSD.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 409361065670 for ; Mon, 5 Jan 2009 11:06:59 +0000 (UTC) (envelope-from owner-bugmaster@FreeBSD.org) Received: from freefall.freebsd.org (freefall.freebsd.org [IPv6:2001:4f8:fff6::28]) by mx1.freebsd.org (Postfix) with ESMTP id 2D7C48FC1B for ; Mon, 5 Jan 2009 11:06:59 +0000 (UTC) (envelope-from owner-bugmaster@FreeBSD.org) Received: from freefall.freebsd.org (localhost [127.0.0.1]) by freefall.freebsd.org (8.14.3/8.14.3) with ESMTP id n05B6xBH002919 for ; Mon, 5 Jan 2009 11:06:59 GMT (envelope-from owner-bugmaster@FreeBSD.org) Received: (from gnats@localhost) by freefall.freebsd.org (8.14.3/8.14.3/Submit) id n05B6wdm002915 for freebsd-scsi@FreeBSD.org; Mon, 5 Jan 2009 11:06:58 GMT (envelope-from owner-bugmaster@FreeBSD.org) Date: Mon, 5 Jan 2009 11:06:58 GMT Message-Id: <200901051106.n05B6wdm002915@freefall.freebsd.org> X-Authentication-Warning: freefall.freebsd.org: gnats set sender to owner-bugmaster@FreeBSD.org using -f From: FreeBSD bugmaster To: freebsd-scsi@FreeBSD.org Cc: Subject: Current problem reports assigned to freebsd-scsi@FreeBSD.org X-BeenThere: freebsd-scsi@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: SCSI subsystem List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 05 Jan 2009 11:06:59 -0000 Note: to view an individual PR, use: http://www.freebsd.org/cgi/query-pr.cgi?pr=(number). The following is a listing of current problems submitted by FreeBSD users. These represent problem reports covering all versions including experimental development code and obsolete releases. S Tracker Resp. Description -------------------------------------------------------------------------------- o kern/129602 scsi [ahd] ahd(4) gets confused and wedges SCSI bus o kern/128452 scsi [sa] [panic] Accessing SCSI tape drive randomly crashe o kern/128245 scsi [scsi] "inquiry data fails comparison at DV1 step" [re o kern/127927 scsi [isp] isp(4) target driver crashes kernel when set up o kern/126866 scsi [isp] [panic] kernel panic on card initialization o kern/124667 scsi [amd] [panic] FreeBSD-7 kernel page faults at amd-scsi o kern/123674 scsi [ahc] ahc driver dumping o kern/123666 scsi [aac] attach fails with Adaptec SAS RAID 3805 controll o sparc/121676 scsi [iscsi] iscontrol do not connect iscsi-target on sparc o kern/120487 scsi [sg] scsi_sg incompatible with scanners o kern/120247 scsi [mpt] FreeBSD 6.3 and LSI Logic 1030 = only 3.300MB/s o kern/119668 scsi [cam] [patch] certain errors are too verbose comparing o kern/114597 scsi [sym] System hangs at SCSI bus reset with dual HBAs o kern/110847 scsi [ahd] Tyan U320 onboard problem with more than 3 disks o kern/99954 scsi [ahc] reading from DVD failes on 6.x [regression] o kern/94838 scsi Kernel panic while mounting SD card with lock switch o o kern/92798 scsi [ahc] SCSI problem with timeouts o kern/90282 scsi [sym] SCSI bus resets cause loss of ch device o kern/76178 scsi [ahd] Problem with ahd and large SCSI Raid system o kern/74627 scsi [ahc] [hang] Adaptec 2940U2W Can't boot 5.3 s kern/61165 scsi [panic] kernel page fault after calling cam_send_ccb o kern/60641 scsi [sym] Sporadic SCSI bus resets with 53C810 under load o kern/60598 scsi wire down of scsi devices conflicts with config s kern/57398 scsi [mly] Current fails to install on mly(4) based RAID di o kern/52638 scsi [panic] SCSI U320 on SMP server won't run faster than o kern/44587 scsi dev/dpt/dpt.h is missing defines required for DPT_HAND o kern/40895 scsi wierd kernel / device driver bug o kern/39388 scsi ncr/sym drivers fail with 53c810 and more than 256MB m o kern/38828 scsi [dpt] [request] DPT PM2012B/90 doesn't work o kern/35234 scsi World access to /dev/pass? (for scanner) requires acce 30 problems total. From owner-freebsd-scsi@FreeBSD.ORG Mon Jan 5 17:18:10 2009 Return-Path: Delivered-To: freebsd-scsi@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id AAF771065742 for ; Mon, 5 Jan 2009 17:18:10 +0000 (UTC) (envelope-from asossi@dotcom.ts.it) Received: from mail.dotcom.ts.it (89-96-242-204.ip14.fastwebnet.it [89.96.242.204]) by mx1.freebsd.org (Postfix) with SMTP id EF5458FC08 for ; Mon, 5 Jan 2009 17:18:09 +0000 (UTC) (envelope-from asossi@dotcom.ts.it) Received: (qmail 75774 invoked by uid 89); 5 Jan 2009 16:51:28 -0000 Received: by simscan 1.1.0 ppid: 75749, pid: 75768, t: 0.5381s scanners:none X-Spam-Checker-Version: SpamAssassin 3.2.3 (2007-08-08) on web.dotcom.ts.it X-Spam-Level: X-Spam-Status: No, hits=-3.0 required=3.0 tests=ALL_TRUSTED,AWL,BAYES_00, URIBL_BLACK,URIBL_GREY,URIBL_RED autolearn=no version=3.2.3 Received: from unknown (HELO ?192.168.0.2?) (asossi@dotcom.ts.it@192.168.254.2) by test.dotcom.ts.it with SMTP; 5 Jan 2009 16:51:28 -0000 Message-ID: <496239E8.4020607@dotcom.ts.it> Date: Mon, 05 Jan 2009 17:48:40 +0100 From: Sossi Andrej Organization: DotCom Information technology User-Agent: Thunderbird 2.0.0.18 (X11/20081105) MIME-Version: 1.0 To: freebsd-scsi@freebsd.org Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 8bit Subject: ISCSI slow to read from DELL MD3000i X-BeenThere: freebsd-scsi@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: SCSI subsystem List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 05 Jan 2009 17:18:11 -0000 Hello, I manage an web/mail/DB server. We recently purchased a DELL MD3000i server with ISCSI interface. I tested the MD3000i with FreeBSD 7.0-RELEASE, i recompiled the kernel myself but the ISCSI communication results slow. Below configuration: net configuration: MD3000i FreeBSD 7.0 controller 0 interface 0: 192.168.253.254 |__| switch 1GB |__|netcard 0: 192.168.253.1 (re0)| controller 1 interface 0: 192.168.253.253 | |192.168.253.X | | | | | | controller 0 interface 1: 192.168.252.254 |__| switch 1GB |__|netcard 1: 192.168.252.1 (re1)| controller 1 interface 1: 192.168.252.253 | |192.168.252.X | | | manage interface 0: 192.168.0.249 |__| switch 10/100 MB |_.... internal net manage interface 1: 192.168.0.248 | |192.168.0.X | /etc/iscsi.conf: port 3260 nas_dotcom { TargetAddress = 192.168.253.254 TargetName = iqn.1984-05.com.dell:powervault.6001ec9000f192210000000048a44372 initiatorName = iqn.2005-01.il.ac.huji.cs::web.dotcom.ts.it } web# sysctl net.iscsi net.iscsi.driver_version: 2.1.0 net.iscsi.isid: €DIB00 net.iscsi.sessions: 1 net.iscsi.0.targetname: iqn.1984-05.com.dell:powervault.6001ec9000f192210000000048a44372 net.iscsi.0.targeaddress: 192.168.253.254 net.iscsi.0.stats: recv=71479 sent=71479 flags=0x0000019f pdus-alloc=0 pdus-max=5 cws=385 cmd=b499 exp=b499 max=b619 stat=b49b itt=b499 net.iscsi.0.douio: 0 net.iscsi.0.pid: 612 web# sysctl kern.cam kern.cam.cam_srch_hi: 0 kern.cam.scsi_delay: 5000 kern.cam.cd.changer.max_busy_seconds: 15 kern.cam.cd.changer.min_busy_seconds: 5 kern.cam.da.da_send_ordered: 1 kern.cam.da.default_timeout: 60 kern.cam.da.retry_count: 4 kern.cam.da.0.minimum_cmd_size: 6 kern.cam.da.1.minimum_cmd_size: 6 kern.cam.da.2.minimum_cmd_size: 6 kern.cam.da.3.minimum_cmd_size: 6 I tryed with the ISCSI driver I found in FreeBSD 7.0, with the same driver patched with the patch I found there (http://lists.freebsd.org/pipermail/freebsd-scsi/2008-February/003383.html) and alse with the driver ISCSI 2.1 downloaded from Danny's ftp. All test have same results: about 820~850KB/s. eg: web# dd if=/dev/zero of=/usr/local/www/tmp count=1m 1048576+0 records in 1048576+0 records out 536870912 bytes transferred in 654.398700 secs (820403 bytes/sec) and simultaneously web# netstat -w 1 -I re0 input (re0) output packets errs bytes packets errs bytes colls 0 0 0 0 0 0 0 123 0 8550 234 0 343180 0 300 0 20952 564 0 824808 0 325 0 22698 611 0 893542 0 300 0 20952 564 0 824808 0 325 0 22698 611 0 893542 0 300 0 20952 564 0 824808 0 .... and more re1 has no traffic. But at the end i try read from ISCSI. Read is more quickly. eg: web# dd if=/usr/local/www/tmp of=/usr/tmp 1048576+0 records in 1048576+0 records out 536870912 bytes transferred in 19.636577 secs (27340351 bytes/sec) and simultaneously: web# netstat -w 1 -I re0 input (re0) output packets errs bytes packets errs bytes colls 0 0 0 0 0 0 0 2210 0 3296084 1156 0 78744 0 16420 0 24493888 8567 0 582558 0 13031 0 19427130 6809 0 463074 0 20638 0 30786236 10768 0 732288 0 17296 0 25791540 9032 0 614160 0 20347 0 30347462 10615 0 721854 0 20254 0 30210156 10567 0 718590 0 .... and more re1 has no traffic. I searched but found nothing that helps me, the only thing that triggered my attention was this Microsoft's User's Guide (http://www.microsoft.com/downloads/details.aspx?familyid=12CB3C1A-15D6-4585-B385-BEFD1319F825&displaylang=en). It seems interesting at least the chapter 'Addressing Slow Performance with iSCSI clusters' Thanks in advance for yours help. -- Sossi Andrej ------------------------- DotCom Information technology Via Biancospino, 9 34151 - Opicina (TS) Italy tel: +39 040 2158191 fax: +39 040 0641954 E-mail: asossi@dotcom.ts.it ---------------------------- Ai sensi del D.lgs n. 196 del 30.06.03 (Codice Privacy) si precisa che le informazioni contenute in questo messaggio sono riservate e ad uso esclusivo del destinatario. Qualora il messaggio in parola Le fosse pervenuto per errore, La preghiamo di eliminarlo senza copiarlo e di non inoltrarlo a terzi, dandocene gentilmente comunicazione. Grazie This message, for the D.lgs n. 196 / 30.06.03 (Privacy Code), may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation. From owner-freebsd-scsi@FreeBSD.ORG Tue Jan 6 13:49:34 2009 Return-Path: Delivered-To: freebsd-scsi@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 8EAAA106566B for ; Tue, 6 Jan 2009 13:49:34 +0000 (UTC) (envelope-from danny@cs.huji.ac.il) Received: from kabab.cs.huji.ac.il (kabab.cs.huji.ac.il [132.65.16.84]) by mx1.freebsd.org (Postfix) with ESMTP id 208DA8FC1C for ; Tue, 6 Jan 2009 13:49:34 +0000 (UTC) (envelope-from danny@cs.huji.ac.il) Received: from pampa.cs.huji.ac.il ([132.65.80.32]) by kabab.cs.huji.ac.il with esmtp id 1LKCJI-000Crk-2p; Tue, 06 Jan 2009 15:49:32 +0200 X-Mailer: exmh version 2.7.2 01/07/2005 with nmh-1.2 To: Sossi Andrej In-reply-to: <496239E8.4020607@dotcom.ts.it> References: <496239E8.4020607@dotcom.ts.it> Comments: In-reply-to Sossi Andrej message dated "Mon, 05 Jan 2009 17:48:40 +0100." Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: quoted-printable Date: Tue, 06 Jan 2009 15:49:32 +0200 From: Danny Braniss Message-ID: Cc: freebsd-scsi@freebsd.org Subject: Re: ISCSI slow to read from DELL MD3000i X-BeenThere: freebsd-scsi@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: SCSI subsystem List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 06 Jan 2009 13:49:35 -0000 sorry for the top potst, my MUI screws up the quoatet message. check: ftp://ftp.cs.huji.ac.il/users/danny/freebsd/iscsi-2.1.tar.gz danny > Hello,> I manage an web/mail/DB server. We recently purchased a DELL M= D3000i > server with ISCSI interface. I tested the MD3000i with FreeBSD >= 7.0-RELEASE, i recompiled the kernel myself but the ISCSI communication= > results slow.> Below configuration:> > net configuration:> MD3000i = > = FreeBSD 7.0> controller 0 interface 0: 192.168.253.2= 54 =7C__=7C switch 1GB > =7C__=7Cnetcard 0: 192.168.253.1 (re0)=7C> c= ontroller 1 interface 0: 192.168.253.253 =7C =7C192.168.253.X =7C = =7C > =7C> = =7C > = =7C > = =7C> controller 0 interface 1: 192.168.252.254 =7C__=7C switch 1GB = > =7C__=7Cnetcard 1: 192.168.252.1 (re1)=7C> controller 1 interface 1:= 192.168.252.253 =7C =7C192.168.252.X =7C =7C > = =7C> > manage interface 0: 192.168.0.249 = =7C__=7C switch 10/100 MB =7C_.... > internal net> manage interf= ace 1: 192.168.0.248 =7C =7C192.168.0.X =3D=7C> >= /etc/iscsi.conf:> port 3260> nas_dotcom =7B> TargetAddress =3D 19= 2.168.253.254> TargetName =3D > iqn.1984-05.com.dell:powervaul= t.6001ec9000f192210000000048a44372> initiatorName =3D iqn.2005-01.= il.ac.huji.cs::web.dotcom.ts.it> =7D> > web=23 sysctl net.iscsi> net.iscs= i.driver_version: 2.1.0> net.iscsi.isid: =C2=80DIB00> net.iscsi.sessions:= 1> net.iscsi.0.targetname: > iqn.1984-05.com.dell:powervault.6001ec9000f= 192210000000048a44372> net.iscsi.0.targeaddress: 192.168.253.254> net.isc= si.0.stats: recv=3D71479 sent=3D71479 flags=3D0x0000019f pdus-allo=3Dc=3D= 0 > pdus-max=3D5 cws=3D385 cmd=3Db499 exp=3Db499 max=3Db619 stat=3Db49b i= tt=3D=3Db499> net.iscsi.0.douio: 0> net.iscsi.0.pid: 612> > web=23 sysctl= kern.cam> kern.cam.cam_srch_hi: 0> kern.cam.scsi_delay: 5000> kern.cam.c= d.changer.max_busy_seconds: 15> kern.cam.cd.changer.min_busy_seconds: 5> = kern.cam.da.da_send_ordered: 1> kern.cam.da.default_timeout: 60> kern.cam= .da.retry_count: 4> kern.cam.da.0.minimum_cmd_size: 6> kern.cam.da.1.mini= mum_cmd_size: 6> kern.cam.da.2.minimum_cmd_size: 6> kern.cam.da.3.minimum= _cmd_size: 6> > > I tryed with the ISCSI driver I found in FreeBSD 7.0, w= ith the same > driver patched with the patch I found there > (http://list= s.freebsd.org/pipermail/freebsd-scsi/2008-February/003383.htm=3Dl) > and = alse with the driver ISCSI 2.1 downloaded from Danny's ftp.> All test hav= e same results: about 820=7E850KB/s.> > eg:> web=23 dd if=3D/dev/zero of= =3D/usr/local/www/tmp count=3D1m> 1048576+0 records in> 1048576+0 records= out> 536870912 bytes transferred in 654.398700 secs (820403 bytes/sec)> = > and simultaneously> web=23 netstat -w 1 -I re0> input = (re0) output> packets errs bytes packets errs = bytes colls> 0 0 0 0 0 0= 0> 123 0 8550 234 0 343180 0> = 300 0 20952 564 0 824808 0> 325 = 0 22698 611 0 893542 0> 300 0 = 20952 564 0 824808 0> 325 0 22698 = 611 0 893542 0> 300 0 20952 564 = 0 824808 0> .... and more> re1 has no traffic.> > But at the en= d i try read from ISCSI. Read is more quickly.> eg:> web=23 dd if=3D/usr/= local/www/tmp of=3D/usr/tmp> 1048576+0 records in> 1048576+0 records out>= 536870912 bytes transferred in 19.636577 secs (27340351 bytes/sec)> > an= d simultaneously:> web=23 netstat -w 1 -I re0> input = (re0) output> packets errs bytes packets errs = bytes colls> 0 0 0 0 0 0 = 0> 2210 0 3296084 1156 0 78744 0> = 16420 0 24493888 8567 0 582558 0> 13031 = 0 19427130 6809 0 463074 0> 20638 0 30786= 236 10768 0 732288 0> 17296 0 25791540 = 9032 0 614160 0> 20347 0 30347462 10615 0= 721854 0> 20254 0 30210156 10567 0 71859= 0 0> .... and more> re1 has no traffic.> > I searched but found noth= ing that helps me, the only thing that > triggered my attention was this= Microsoft's User's Guide > (http://www.microsoft.com/downloads/details.a= spx?familyid=3D12CB3C1A-15D6=3D-4585-B385-BEFD1319F825&displaylang=3Den).= > > It seems interesting at least the chapter 'Addressing Slow Performan= ce > with iSCSI clusters'> > Thanks in advance for yours help.> > -- > So= ssi Andrej> -------------------------> DotCom Information technology> > V= ia Biancospino, 9> 34151 - Opicina (TS)> Italy> > tel: +39 040 2158191> f= ax: +39 040 0641954> E-mail: asossi=40dotcom.ts.it> ---------------------= -------> > Ai sensi del D.lgs n. 196 del 30.06.03 (Codice Privacy) si pre= cisa che le=3D informazioni contenute in questo messaggio sono riservate = e ad uso esclu=3Dsivo del destinatario. Qualora il messaggio in parola Le= fosse pervenuto =3Dper errore, La preghiamo di eliminarlo senza copiarlo= e di non inoltrarlo=3D a terzi, dandocene gentilmente comunicazione. Gra= zie> > This message, for the D.lgs n. 196 / 30.06.03 (Privacy Code), may = contain=3D confidential and/or privileged information. If you are not the= addressee=3D or authorized to receive this for the addressee, you must n= ot use, copy,=3D disclose or take any action based on this message or any= information her=3Dein. If you have received this message in error, pleas= e advise the sender=3D immediately by reply e-mail and delete this messag= e. Thank you for your =3Dcooperation.> > ________________________________= _______________> freebsd-scsi=40freebsd.org mailing list> http://lists.fr= eebsd.org/mailman/listinfo/freebsd-scsi> To unsubscribe, send any mail to= =22freebsd-scsi-unsubscribe=40freebsd.org=22>=20 From owner-freebsd-scsi@FreeBSD.ORG Tue Jan 6 18:03:14 2009 Return-Path: Delivered-To: freebsd-scsi@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 780CE1065672 for ; Tue, 6 Jan 2009 18:03:14 +0000 (UTC) (envelope-from nglrossi@gmail.com) Received: from yw-out-2324.google.com (yw-out-2324.google.com [74.125.46.31]) by mx1.freebsd.org (Postfix) with ESMTP id 299388FC1A for ; Tue, 6 Jan 2009 18:03:14 +0000 (UTC) (envelope-from nglrossi@gmail.com) Received: by yw-out-2324.google.com with SMTP id 9so3433599ywe.13 for ; Tue, 06 Jan 2009 10:03:13 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:message-id:date:from:to :subject:mime-version:content-type; bh=ioNgNdPHLWcvf1xvaZ1XCVT8cf7QjdU99ueF6zIpsrg=; b=flZ6a9NFAhwVgP30gIBlxxIhq4vNKeSnlL5xlRD7K3sHrvVmfpUly2IIqwV0USnqKG SdVYW+5ceP4MNwuu5BhGAxHjBR56FUcea3JJWhNNqnjg/0h4IDMZF6Afl8Kl8i6lKiaK /xKl6u6j7SG5e/x79MnGNMpMr0vDe4nrDVhj4= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=message-id:date:from:to:subject:mime-version:content-type; b=NbWhYixzZjl1gncstLFcPBDzSIndYHNdTfOklGl4J+VzA9abtaSp7W6WQij6PMcBAl uFJ6Zn5/a377FqE5TOa8xDwkTAVP8AJgoKaDvypfvdk6y8zuHtbQzIVow7/lqf2jwpXv 9vVpt+ecWUDOlqfdJxQ3pat1+dGOIBF1reO1A= Received: by 10.100.207.14 with SMTP id e14mr11837802ang.60.1231263507088; Tue, 06 Jan 2009 09:38:27 -0800 (PST) Received: by 10.100.136.13 with HTTP; Tue, 6 Jan 2009 09:38:27 -0800 (PST) Message-ID: <6c1e076a0901060938i5450a1fnc4dd68ff9690862d@mail.gmail.com> Date: Tue, 6 Jan 2009 18:38:27 +0100 From: Angelo To: freebsd-scsi@freebsd.org MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline X-Content-Filtered-By: Mailman/MimeDel 2.1.5 Subject: iscsi client hangs performing I/O on a dead target X-BeenThere: freebsd-scsi@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: SCSI subsystem List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 06 Jan 2009 18:03:14 -0000 Hi, I'm experimenting on iscsi; both the target and the initiator are vmware virtual machines and are running 7.0-RELEASE. So far everything works fine except that the client hangs forever (state is PHYSWR) if it is accessing (dd if=/dev/zero of=/dev/da0) the device while it "dies" (shutdown of the target machine). Is this the expected behaviour? I would like to be able to configure a timeout so the command that is trying to access the unavailable device fails and I can handle the problem but I can't find anything in iscontrol man pages. Thanks, Angelo From owner-freebsd-scsi@FreeBSD.ORG Wed Jan 7 03:38:33 2009 Return-Path: Delivered-To: scsi@FreeBSD.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 50B80106564A for ; Wed, 7 Jan 2009 03:38:33 +0000 (UTC) (envelope-from imp@bsdimp.com) Received: from harmony.bsdimp.com (bsdimp.com [199.45.160.85]) by mx1.freebsd.org (Postfix) with ESMTP id EC83C8FC17 for ; Wed, 7 Jan 2009 03:38:32 +0000 (UTC) (envelope-from imp@bsdimp.com) Received: from localhost (localhost [127.0.0.1]) by harmony.bsdimp.com (8.14.2/8.14.1) with ESMTP id n073bTlZ043282 for ; Tue, 6 Jan 2009 20:37:29 -0700 (MST) (envelope-from imp@bsdimp.com) Date: Tue, 06 Jan 2009 20:37:36 -0700 (MST) Message-Id: <20090106.203736.-1300543249.imp@bsdimp.com> To: scsi@FreeBSD.org From: "M. Warner Losh" X-Mailer: Mew version 5.2 on Emacs 21.3 / Mule 5.0 (SAKAKI) Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Content-Transfer-Encoding: 7bit Cc: Subject: Patch review X-BeenThere: freebsd-scsi@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: SCSI subsystem List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 07 Jan 2009 03:38:33 -0000 What do people think of this? It is very useful when trying to recover a CD/DVD that's been scratched with ddrescue which handles all the retry logic. This patch is modeled after the kern.cam.da.retry_count sysctl and friends. This makes ddrescue about between 5x and 20x faster. The reason for the large delta is that it can now 'stream' past errors rather than doing a lot of reseeking when reading a large block size. I've also observed fewer hangs in one of the cheap DVD drives I tried this with, likely due to fewer forced retries frying its brain... Of course, all of this was with kern.cam.cd.retry_count = 0. I've not changed the default behavior. Comments? Warner P.S. I have a recollection that Scott Long sent me something akin to this, but I couldn't find it so I just recreated this trivial patch from scratch, cheating off of scsi_da.c. Index: cam/scsi/scsi_cd.c =================================================================== --- cam/scsi/scsi_cd.c (revision 186807) +++ cam/scsi/scsi_cd.c (working copy) @@ -292,6 +292,9 @@ PERIPHDRIVER_DECLARE(cd, cddriver); +#ifndef CD_DEFAULT_RETRY +#define CD_DEFAULT_RETRY 4 +#endif #ifndef CHANGER_MIN_BUSY_SECONDS #define CHANGER_MIN_BUSY_SECONDS 5 #endif @@ -299,11 +302,15 @@ #define CHANGER_MAX_BUSY_SECONDS 15 #endif +static int cd_retry_count = CD_DEFAULT_RETRY; static int changer_min_busy_seconds = CHANGER_MIN_BUSY_SECONDS; static int changer_max_busy_seconds = CHANGER_MAX_BUSY_SECONDS; SYSCTL_NODE(_kern_cam, OID_AUTO, cd, CTLFLAG_RD, 0, "CAM CDROM driver"); SYSCTL_NODE(_kern_cam_cd, OID_AUTO, changer, CTLFLAG_RD, 0, "CD Changer"); +SYSCTL_INT(_kern_cam_cd, OID_AUTO, retry_count, CTLFLAG_RW, + &cd_retry_count, 0, "Normal I/O retry count"); +TUNABLE_INT("kern.cam.cd.retry_count", &cd_retry_count); SYSCTL_INT(_kern_cam_cd_changer, OID_AUTO, min_busy_seconds, CTLFLAG_RW, &changer_min_busy_seconds, 0, "Minimum changer scheduling quantum"); TUNABLE_INT("kern.cam.cd.changer.min_busy_seconds", &changer_min_busy_seconds); @@ -1454,7 +1461,7 @@ devstat_start_transaction_bio(softc->disk->d_devstat, bp); scsi_read_write(&start_ccb->csio, - /*retries*/4, + /*retries*/cd_retry_count, /* cbfcnp */ cddone, MSG_SIMPLE_Q_TAG, /* read */bp->bio_cmd == BIO_READ, From owner-freebsd-scsi@FreeBSD.ORG Wed Jan 7 09:33:27 2009 Return-Path: Delivered-To: freebsd-scsi@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id F33C1106566C for ; Wed, 7 Jan 2009 09:33:26 +0000 (UTC) (envelope-from danny@cs.huji.ac.il) Received: from kabab.cs.huji.ac.il (kabab.cs.huji.ac.il [132.65.16.84]) by mx1.freebsd.org (Postfix) with ESMTP id AE17E8FC13 for ; Wed, 7 Jan 2009 09:33:26 +0000 (UTC) (envelope-from danny@cs.huji.ac.il) Received: from pampa.cs.huji.ac.il ([132.65.80.32]) by kabab.cs.huji.ac.il with esmtp id 1LKUmz-000O18-Kp; Wed, 07 Jan 2009 11:33:25 +0200 X-Mailer: exmh version 2.7.2 01/07/2005 with nmh-1.2 To: Angelo In-reply-to: Your message of Tue, 6 Jan 2009 18:38:27 +0100 . Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Wed, 07 Jan 2009 11:33:25 +0200 From: Danny Braniss Message-ID: Cc: freebsd-scsi@freebsd.org Subject: Re: iscsi client hangs performing I/O on a dead target X-BeenThere: freebsd-scsi@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: SCSI subsystem List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 07 Jan 2009 09:33:27 -0000 > Hi, > > I'm experimenting on iscsi; both the target and the initiator are vmware > virtual machines and are running 7.0-RELEASE. > So far everything works fine except that the client hangs forever (state is > PHYSWR) if it is accessing (dd if=/dev/zero of=/dev/da0) the device while it > "dies" (shutdown of the target machine). > what happens when you bring back the target? > Is this the expected behaviour? > yes. but maybe it can be 'fixed', I'm concidering adding some kind of timeout flag/option. > I would like to be able to configure a timeout so the command that is trying > to access the unavailable device fails and I can handle the problem but I > can't find anything in iscontrol man pages. > > Thanks, > Angelo > _______________________________________________ > freebsd-scsi@freebsd.org mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-scsi > To unsubscribe, send any mail to "freebsd-scsi-unsubscribe@freebsd.org" > From owner-freebsd-scsi@FreeBSD.ORG Wed Jan 7 10:47:18 2009 Return-Path: Delivered-To: freebsd-scsi@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 1D47F106564A for ; Wed, 7 Jan 2009 10:47:18 +0000 (UTC) (envelope-from nglrossi@gmail.com) Received: from yw-out-2324.google.com (yw-out-2324.google.com [74.125.46.28]) by mx1.freebsd.org (Postfix) with ESMTP id C84E28FC18 for ; Wed, 7 Jan 2009 10:47:17 +0000 (UTC) (envelope-from nglrossi@gmail.com) Received: by yw-out-2324.google.com with SMTP id 9so3552392ywe.13 for ; Wed, 07 Jan 2009 02:47:17 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:message-id:date:from:to :subject:in-reply-to:mime-version:content-type:references; bh=OjY+IGB42UPemRHvAI/Jz6fVjX1nwzIbcxR9YylLUGE=; b=xAzmORVv7ZpdsXAmzFGyLDb2ncMposJ5oNn1VK8HLsZXlsXu0BHKXofWEZvd/ljTwo ISATHBX+Cvm4TisHX5WpxkQWoeo50Ct/zSAcdqmhk0ZSRJ39phddp/YfhrabaQALXYGe jb7hwoo+tH1UiqZ2THeMMiDD3iDc+jkmvymn8= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=message-id:date:from:to:subject:in-reply-to:mime-version :content-type:references; b=T2MiGBaj7dFN2QVpJd+XnPw2cusi1gWQzOLzCW1cUp51mOT8lY/o2MIxeB8gY0La4Q 5jbUQhnrg6Pp08tGOHHnaODwRzrg7kdAiIyEwfqG0miLl8N/Ulqd3bZ9ezuKe97Mjl5s i5ljHPRf4To01ASVBI8SWIxLQYmoXyJTopHtU= Received: by 10.100.127.15 with SMTP id z15mr12318808anc.10.1231325237123; Wed, 07 Jan 2009 02:47:17 -0800 (PST) Received: by 10.100.136.13 with HTTP; Wed, 7 Jan 2009 02:47:17 -0800 (PST) Message-ID: <6c1e076a0901070247l7c006efajda8fddee84c337a@mail.gmail.com> Date: Wed, 7 Jan 2009 11:47:17 +0100 From: Angelo To: "Danny Braniss" , freebsd-scsi@freebsd.org In-Reply-To: MIME-Version: 1.0 References: Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline X-Content-Filtered-By: Mailman/MimeDel 2.1.5 Cc: Subject: Re: iscsi client hangs performing I/O on a dead target X-BeenThere: freebsd-scsi@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: SCSI subsystem List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 07 Jan 2009 10:47:18 -0000 On Wed, Jan 7, 2009 at 10:33 AM, Danny Braniss wrote: > > Hi, > > > > I'm experimenting on iscsi; both the target and the initiator are vmware > > virtual machines and are running 7.0-RELEASE. > > So far everything works fine except that the client hangs forever (state > is > > PHYSWR) if it is accessing (dd if=/dev/zero of=/dev/da0) the device while > it > > "dies" (shutdown of the target machine). > > > what happens when you bring back the target? > > > Is this the expected behaviour? > > > yes. > but maybe it can be 'fixed', I'm concidering adding some kind of timeout > flag/option. > > > I would like to be able to configure a timeout so the command that is > trying > > to access the unavailable device fails and I can handle the problem but I > > can't find anything in iscontrol man pages. > > > > Thanks, > > Angelo > > _______________________________________________ > > freebsd-scsi@freebsd.org mailing list > > http://lists.freebsd.org/mailman/listinfo/freebsd-scsi > > To unsubscribe, send any mail to "freebsd-scsi-unsubscribe@freebsd.org" > > > > Hi Danny, thanks for your reply. when the target comes back it is reconnected properly by the initiator. The usage of iscsi I had in mind was an initiator connecting to two targets and using the two devices as parts of a raid 1 array. I was expecting that a failure on one of the disks (network down, target machine down) would be eventually detected and the raid turned to degraded but still usable. What happens now to me is that the whole raid device gets unusable for all the time the target is down. As I said before I am just experimenting. I don't know how other implementations of iscsi initiator work and maybe my idea is just insane :) Thanks, Angelo From owner-freebsd-scsi@FreeBSD.ORG Wed Jan 7 13:27:59 2009 Return-Path: Delivered-To: freebsd-scsi@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id A6AC7106566B for ; Wed, 7 Jan 2009 13:27:59 +0000 (UTC) (envelope-from danny@cs.huji.ac.il) Received: from kabab.cs.huji.ac.il (kabab.cs.huji.ac.il [132.65.16.84]) by mx1.freebsd.org (Postfix) with ESMTP id 5D9118FC14 for ; Wed, 7 Jan 2009 13:27:59 +0000 (UTC) (envelope-from danny@cs.huji.ac.il) Received: from pampa.cs.huji.ac.il ([132.65.80.32]) by kabab.cs.huji.ac.il with esmtp id 1LKYRx-0000na-VV; Wed, 07 Jan 2009 15:27:58 +0200 X-Mailer: exmh version 2.7.2 01/07/2005 with nmh-1.2 To: Angelo In-reply-to: <6c1e076a0901070247l7c006efajda8fddee84c337a@mail.gmail.com> References: <6c1e076a0901070247l7c006efajda8fddee84c337a@mail.gmail.com> Comments: In-reply-to Angelo message dated "Wed, 07 Jan 2009 11:47:17 +0100." Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Wed, 07 Jan 2009 15:27:57 +0200 From: Danny Braniss Message-ID: Cc: freebsd-scsi@freebsd.org Subject: Re: iscsi client hangs performing I/O on a dead target X-BeenThere: freebsd-scsi@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: SCSI subsystem List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 07 Jan 2009 13:28:00 -0000 > ------=_Part_16620_23870833.1231325237102 > Content-Type: text/plain; charset=ISO-8859-1 > Content-Transfer-Encoding: 7bit > Content-Disposition: inline > > On Wed, Jan 7, 2009 at 10:33 AM, Danny Braniss wrote: > > > > Hi, > > > > > > I'm experimenting on iscsi; both the target and the initiator are vmware > > > virtual machines and are running 7.0-RELEASE. > > > So far everything works fine except that the client hangs forever (state > > is > > > PHYSWR) if it is accessing (dd if=/dev/zero of=/dev/da0) the device while > > it > > > "dies" (shutdown of the target machine). > > > > > what happens when you bring back the target? > > > > > Is this the expected behaviour? > > > > > yes. > > but maybe it can be 'fixed', I'm concidering adding some kind of timeout > > flag/option. > > > > > I would like to be able to configure a timeout so the command that is > > trying > > > to access the unavailable device fails and I can handle the problem but I > > > can't find anything in iscontrol man pages. > > > > > > Thanks, > > > Angelo > > > _______________________________________________ > > > freebsd-scsi@freebsd.org mailing list > > > http://lists.freebsd.org/mailman/listinfo/freebsd-scsi > > > To unsubscribe, send any mail to "freebsd-scsi-unsubscribe@freebsd.org" > > > > > > > Hi Danny, > > > thanks for your reply. > > when the target comes back it is reconnected properly by the initiator. > > The usage of iscsi I had in mind was an initiator connecting to two targets > and using the two devices as parts of a raid 1 array. > I was expecting that a failure on one of the disks (network down, target > machine down) would be eventually detected and the raid turned to degraded > but still usable. > What happens now to me is that the whole raid device gets unusable for all > the time the target is down. > > As I said before I am just experimenting. I don't know how other > implementations of iscsi initiator work and maybe my idea is just insane :) > I don't know either, the RFC does not mention this case - or maybe I missed it :-) But others have raised the issue, and so it got me thinking, but freebsd likes to panic if a disc goes away, so don't hold your breath for a solution soon. danny > Thanks, > Angelo From owner-freebsd-scsi@FreeBSD.ORG Wed Jan 7 14:23:36 2009 Return-Path: Delivered-To: freebsd-scsi@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 84EA21065673 for ; Wed, 7 Jan 2009 14:23:36 +0000 (UTC) (envelope-from asossi@dotcom.ts.it) Received: from mail.dotcom.ts.it (89-96-242-204.ip14.fastwebnet.it [89.96.242.204]) by mx1.freebsd.org (Postfix) with SMTP id 9ACB78FC29 for ; Wed, 7 Jan 2009 14:23:34 +0000 (UTC) (envelope-from asossi@dotcom.ts.it) Received: (qmail 80666 invoked by uid 89); 7 Jan 2009 14:23:33 -0000 Received: by simscan 1.1.0 ppid: 80648, pid: 80662, t: 0.3329s scanners:none X-Spam-Checker-Version: SpamAssassin 3.2.3 (2007-08-08) on web.dotcom.ts.it X-Spam-Level: X-Spam-Status: No, hits=-4.1 required=3.0 tests=ALL_TRUSTED,AWL,BAYES_00 autolearn=ham version=3.2.3 Received: from unknown (HELO ?192.168.0.2?) (asossi@dotcom.ts.it@192.168.254.2) by test.dotcom.ts.it with SMTP; 7 Jan 2009 14:23:33 -0000 Message-ID: <4964BA40.5080307@dotcom.ts.it> Date: Wed, 07 Jan 2009 15:20:48 +0100 From: Sossi Andrej Organization: DotCom Information technology User-Agent: Thunderbird 2.0.0.18 (X11/20081105) MIME-Version: 1.0 To: Danny Braniss References: In-Reply-To: Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Cc: freebsd-scsi@freebsd.org Subject: Re: ISCSI slow to read from DELL MD3000i X-BeenThere: freebsd-scsi@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: SCSI subsystem List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 07 Jan 2009 14:23:36 -0000 Danny Braniss wrote: > wups, didn't read all your mail, sorry, > try increasing tht tag opening, in the config file add > tag = 64 > for example, or via camcontrol > > danny > I modify my iscsi.conf by adding tags = 64 (and no tag). Now works fine, about 43MB/s. Thank You very match. But what means this parameter? -- Sossi Andrej ------------------------- DotCom Information technology Via Biancospino, 9 34151 - Opicina (TS) Italy tel: +39 040 2158191 fax: +39 040 0641954 E-mail: asossi@dotcom.ts.it ---------------------------- Ai sensi del D.lgs n. 196 del 30.06.03 (Codice Privacy) si precisa che le informazioni contenute in questo messaggio sono riservate e ad uso esclusivo del destinatario. Qualora il messaggio in parola Le fosse pervenuto per errore, La preghiamo di eliminarlo senza copiarlo e di non inoltrarlo a terzi, dandocene gentilmente comunicazione. Grazie This message, for the D.lgs n. 196 / 30.06.03 (Privacy Code), may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation. From owner-freebsd-scsi@FreeBSD.ORG Wed Jan 7 17:17:34 2009 Return-Path: Delivered-To: freebsd-scsi@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 3B6251065687 for ; Wed, 7 Jan 2009 17:17:34 +0000 (UTC) (envelope-from nglrossi@gmail.com) Received: from yx-out-2324.google.com (yx-out-2324.google.com [74.125.44.29]) by mx1.freebsd.org (Postfix) with ESMTP id E40518FC34 for ; Wed, 7 Jan 2009 17:17:33 +0000 (UTC) (envelope-from nglrossi@gmail.com) Received: by yx-out-2324.google.com with SMTP id 8so3614851yxb.13 for ; Wed, 07 Jan 2009 09:17:33 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:message-id:date:from:to :subject:cc:in-reply-to:mime-version:content-type:references; bh=DR8x4+8givpaBJFxKzfibXPHsruFSAowlEPpDI9OucY=; b=PRuOiN3doGbqD7pwxl/0PrG6+aIwTTavhJvVXWJ9lD1V2iDH8Vj4/e9IH8zGwjtMqu R5IfRfUtuueoxOuBxWdgHd/tRJqKFkop0Qy2D0d9lUdBAleAYGCao+F5pzyAMXtKVOA4 zlKg6rxUmixkPSIcNoUUaE+AbJUukb2T5/GN0= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=message-id:date:from:to:subject:cc:in-reply-to:mime-version :content-type:references; b=Dl/lIovtsHO/JfDES6A0FtC7j6Avqfj3LX7CZzb1Hjl3UGPLFQwbCaeplCtxtr+fje RKb7vaM7CxUPYS3BcHmTN7Yak8MW4b6R5u0G6iptyTIOwYOAPY4+AFL1v4QoZVS3i/h1 3wGYaWVpU5QPzHa9jrDiNiguTH5Vp4rC54ZO4= Received: by 10.100.9.19 with SMTP id 19mr12653515ani.8.1231348653389; Wed, 07 Jan 2009 09:17:33 -0800 (PST) Received: by 10.100.136.13 with HTTP; Wed, 7 Jan 2009 09:17:33 -0800 (PST) Message-ID: <6c1e076a0901070917p52d32078qff616449c963f6fc@mail.gmail.com> Date: Wed, 7 Jan 2009 18:17:33 +0100 From: Angelo To: "Danny Braniss" In-Reply-To: MIME-Version: 1.0 References: <6c1e076a0901070247l7c006efajda8fddee84c337a@mail.gmail.com> Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline X-Content-Filtered-By: Mailman/MimeDel 2.1.5 Cc: freebsd-scsi@freebsd.org Subject: Re: iscsi client hangs performing I/O on a dead target X-BeenThere: freebsd-scsi@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: SCSI subsystem List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 07 Jan 2009 17:17:35 -0000 Danny, this sounds great (except for the panic, but I guess this would at least be the same behaviour of two scsi disks configured in raid1 with one failing). I'll keep an eye on the changes and be glad to help testing if you implement the timeout setting! thanks, Angelo On Wed, Jan 7, 2009 at 2:27 PM, Danny Braniss wrote: > > ------=_Part_16620_23870833.1231325237102 > > Content-Type: text/plain; charset=ISO-8859-1 > > Content-Transfer-Encoding: 7bit > > Content-Disposition: inline > > > > On Wed, Jan 7, 2009 at 10:33 AM, Danny Braniss > wrote: > > > > > > Hi, > > > > > > > > I'm experimenting on iscsi; both the target and the initiator are > vmware > > > > virtual machines and are running 7.0-RELEASE. > > > > So far everything works fine except that the client hangs forever > (state > > > is > > > > PHYSWR) if it is accessing (dd if=/dev/zero of=/dev/da0) the device > while > > > it > > > > "dies" (shutdown of the target machine). > > > > > > > what happens when you bring back the target? > > > > > > > Is this the expected behaviour? > > > > > > > yes. > > > but maybe it can be 'fixed', I'm concidering adding some kind of > timeout > > > flag/option. > > > > > > > I would like to be able to configure a timeout so the command that is > > > trying > > > > to access the unavailable device fails and I can handle the problem > but I > > > > can't find anything in iscontrol man pages. > > > > > > > > > Thanks, > > > > Angelo > > > > _______________________________________________ > > > > freebsd-scsi@freebsd.org mailing list > > > > http://lists.freebsd.org/mailman/listinfo/freebsd-scsi > > > > To unsubscribe, send any mail to " > freebsd-scsi-unsubscribe@freebsd.org" > > > > > > > > > > Hi Danny, > > > > > > thanks for your reply. > > > > when the target comes back it is reconnected properly by the initiator. > > > > The usage of iscsi I had in mind was an initiator connecting to two > targets > > and using the two devices as parts of a raid 1 array. > > I was expecting that a failure on one of the disks (network down, target > > machine down) would be eventually detected and the raid turned to > degraded > > but still usable. > > What happens now to me is that the whole raid device gets unusable for > all > > the time the target is down. > > > > As I said before I am just experimenting. I don't know how other > > implementations of iscsi initiator work and maybe my idea is just insane > :) > > > I don't know either, the RFC does not mention this case - or maybe I missed > it > :-) > But others have raised the issue, and so it got me thinking, but freebsd > likes > to > panic if a disc goes away, so don't hold your breath for a solution soon. > > danny > > > Thanks, > > Angelo > > From owner-freebsd-scsi@FreeBSD.ORG Wed Jan 7 23:45:32 2009 Return-Path: Delivered-To: freebsd-scsi@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 2FB0910656CB for ; Wed, 7 Jan 2009 23:45:32 +0000 (UTC) (envelope-from andrew@modulus.org) Received: from email.octopus.com.au (email.octopus.com.au [122.100.2.232]) by mx1.freebsd.org (Postfix) with ESMTP id E3E328FC08 for ; Wed, 7 Jan 2009 23:45:31 +0000 (UTC) (envelope-from andrew@modulus.org) Received: by email.octopus.com.au (Postfix, from userid 1002) id 1B7CC17D8D; Thu, 8 Jan 2009 10:28:18 +1100 (EST) X-Spam-Checker-Version: SpamAssassin 3.2.3 (2007-08-08) on email.octopus.com.au X-Spam-Level: X-Spam-Status: No, score=-1.4 required=10.0 tests=ALL_TRUSTED autolearn=failed version=3.2.3 Received: from [10.1.50.60] (ppp121-44-0-132.lns10.syd7.internode.on.net [121.44.0.132]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) (Authenticated sender: admin@email.octopus.com.au) by email.octopus.com.au (Postfix) with ESMTP id 14A83172D8; Thu, 8 Jan 2009 10:28:10 +1100 (EST) Message-ID: <49653A7D.30906@modulus.org> Date: Thu, 08 Jan 2009 10:27:57 +1100 From: Andrew Snow User-Agent: Thunderbird 2.0.0.14 (X11/20080523) MIME-Version: 1.0 To: Danny Braniss References: <6c1e076a0901070247l7c006efajda8fddee84c337a@mail.gmail.com> In-Reply-To: Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Cc: freebsd-fs@freebsd.org, freebsd-scsi@freebsd.org Subject: Re: iscsi client hangs performing I/O on a dead target X-BeenThere: freebsd-scsi@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: SCSI subsystem List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 07 Jan 2009 23:45:32 -0000 Danny Braniss wrote: > freebsd likes to > panic if a disc goes away, so don't hold your breath for a solution soon. You hit the nail on the head. This is going to be a increasingly bad problem as we continue pushing in this of networked storage. It would be REALLY nice, if an I/O fails, to simply have the process attempting the I/O to be killed, instead of bringing the machine to its knees and locking up or panic the kernel. - Andrew From owner-freebsd-scsi@FreeBSD.ORG Thu Jan 8 02:40:35 2009 Return-Path: Delivered-To: freebsd-scsi@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 7E24F1065670 for ; Thu, 8 Jan 2009 02:40:35 +0000 (UTC) (envelope-from andrew@modulus.org) Received: from email.octopus.com.au (email.octopus.com.au [122.100.2.232]) by mx1.freebsd.org (Postfix) with ESMTP id 3DC6A8FC1B for ; Thu, 8 Jan 2009 02:40:35 +0000 (UTC) (envelope-from andrew@modulus.org) Received: by email.octopus.com.au (Postfix, from userid 1002) id 2D36B17D7E; Thu, 8 Jan 2009 13:40:35 +1100 (EST) X-Spam-Checker-Version: SpamAssassin 3.2.3 (2007-08-08) on email.octopus.com.au X-Spam-Level: X-Spam-Status: No, score=-1.4 required=10.0 tests=ALL_TRUSTED autolearn=failed version=3.2.3 Received: from [10.1.50.60] (ppp121-44-0-132.lns10.syd7.internode.on.net [121.44.0.132]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) (Authenticated sender: admin@email.octopus.com.au) by email.octopus.com.au (Postfix) with ESMTP id 0B3F2172CC; Thu, 8 Jan 2009 13:40:31 +1100 (EST) Message-ID: <49656791.1000809@modulus.org> Date: Thu, 08 Jan 2009 13:40:17 +1100 From: Andrew Snow User-Agent: Thunderbird 2.0.0.14 (X11/20080523) MIME-Version: 1.0 To: Angelo , freebsd-scsi@freebsd.org References: <6c1e076a0901070247l7c006efajda8fddee84c337a@mail.gmail.com> In-Reply-To: <6c1e076a0901070247l7c006efajda8fddee84c337a@mail.gmail.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Cc: Subject: Re: iscsi client hangs performing I/O on a dead target X-BeenThere: freebsd-scsi@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: SCSI subsystem List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 08 Jan 2009 02:40:35 -0000 > I was expecting that a failure on one of the disks (network down, target > machine down) would be eventually detected and the raid turned to degraded > but still usable. > What happens now to me is that the whole raid device gets unusable for all > the time the target is down. If you're using gmirror, you might want to experiment with this setting: kern.geom.mirror.timeout Set it to something reasonably low like 10 seconds, and it will ignore an iSCSI device that is not responding more rapidly. From owner-freebsd-scsi@FreeBSD.ORG Thu Jan 8 09:37:54 2009 Return-Path: Delivered-To: freebsd-scsi@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 7C7EF106564A; Thu, 8 Jan 2009 09:37:53 +0000 (UTC) (envelope-from peterjeremy@optushome.com.au) Received: from fallbackmx08.syd.optusnet.com.au (fallbackmx08.syd.optusnet.com.au [211.29.132.10]) by mx1.freebsd.org (Postfix) with ESMTP id D8A998FC23; Thu, 8 Jan 2009 09:36:08 +0000 (UTC) (envelope-from peterjeremy@optushome.com.au) Received: from mail17.syd.optusnet.com.au (mail17.syd.optusnet.com.au [211.29.132.198]) by fallbackmx08.syd.optusnet.com.au (8.13.1/8.13.1) with ESMTP id n087HreT018567; Thu, 8 Jan 2009 18:17:53 +1100 Received: from server.vk2pj.dyndns.org (c122-106-215-175.belrs3.nsw.optusnet.com.au [122.106.215.175]) by mail17.syd.optusnet.com.au (8.13.1/8.13.1) with ESMTP id n087HoAS004649 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NO); Thu, 8 Jan 2009 18:17:51 +1100 X-Bogosity: Ham, spamicity=0.000000 Received: from server.vk2pj.dyndns.org (localhost.vk2pj.dyndns.org [127.0.0.1]) by server.vk2pj.dyndns.org (8.14.3/8.14.3) with ESMTP id n087Hofv057629; Thu, 8 Jan 2009 18:17:50 +1100 (EST) (envelope-from peter@server.vk2pj.dyndns.org) Received: (from peter@localhost) by server.vk2pj.dyndns.org (8.14.3/8.14.3/Submit) id n087HnBw057628; Thu, 8 Jan 2009 18:17:49 +1100 (EST) (envelope-from peter) Date: Thu, 8 Jan 2009 18:17:49 +1100 From: Peter Jeremy To: Andrew Snow Message-ID: <20090108071749.GM87057@server.vk2pj.dyndns.org> References: <6c1e076a0901070247l7c006efajda8fddee84c337a@mail.gmail.com> <49653A7D.30906@modulus.org> MIME-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="p7qwJlK53pWzbayA" Content-Disposition: inline In-Reply-To: <49653A7D.30906@modulus.org> X-PGP-Key: http://members.optusnet.com.au/peterjeremy/pubkey.asc User-Agent: Mutt/1.5.18 (2008-05-17) Cc: freebsd-fs@freebsd.org, freebsd-scsi@freebsd.org Subject: Re: iscsi client hangs performing I/O on a dead target X-BeenThere: freebsd-scsi@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: SCSI subsystem List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 08 Jan 2009 09:37:54 -0000 --p7qwJlK53pWzbayA Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On 2009-Jan-08 10:27:57 +1100, Andrew Snow wrote: >It would be REALLY nice, if an I/O fails, to simply have the process=20 >attempting the I/O to be killed, instead of bringing the machine to its=20 >knees and locking up or panic the kernel. And if it was that easy, it would have been done about a decade ago. --=20 Peter Jeremy Please excuse any delays as the result of my ISP's inability to implement an MTA that is either RFC2821-compliant or matches their claimed behaviour. --p7qwJlK53pWzbayA Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.9 (FreeBSD) iEYEARECAAYFAkllqJ0ACgkQ/opHv/APuIf2xACfcDeN/QQNyVPTS5Y8wiQGpnEX cGMAnA8baEtNGyTZOGaaq5Qo5ecuQsH3 =fMRS -----END PGP SIGNATURE----- --p7qwJlK53pWzbayA-- From owner-freebsd-scsi@FreeBSD.ORG Thu Jan 8 15:39:01 2009 Return-Path: Delivered-To: freebsd-scsi@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id EAE3B1065675 for ; Thu, 8 Jan 2009 15:39:01 +0000 (UTC) (envelope-from asossi@dotcom.ts.it) Received: from mail.dotcom.ts.it (89-96-242-204.ip14.fastwebnet.it [89.96.242.204]) by mx1.freebsd.org (Postfix) with SMTP id 47E1B8FC12 for ; Thu, 8 Jan 2009 15:39:00 +0000 (UTC) (envelope-from asossi@dotcom.ts.it) Received: (qmail 37896 invoked by uid 89); 8 Jan 2009 15:38:59 -0000 Received: by simscan 1.1.0 ppid: 37866, pid: 37886, t: 0.6552s scanners:none X-Spam-Checker-Version: SpamAssassin 3.2.3 (2007-08-08) on web.dotcom.ts.it X-Spam-Level: X-Spam-Status: No, hits=-4.1 required=3.0 tests=ALL_TRUSTED,AWL,BAYES_00 autolearn=ham version=3.2.3 Received: from unknown (HELO ?192.168.0.2?) (asossi@dotcom.ts.it@192.168.254.2) by test.dotcom.ts.it with SMTP; 8 Jan 2009 15:38:58 -0000 Message-ID: <49661D6D.1090602@dotcom.ts.it> Date: Thu, 08 Jan 2009 16:36:13 +0100 From: Sossi Andrej Organization: DotCom Information technology User-Agent: Thunderbird 2.0.0.18 (X11/20081105) MIME-Version: 1.0 To: Danny Braniss References: In-Reply-To: Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Cc: freebsd-scsi@freebsd.org Subject: Re: ISCSI slow to read from DELL MD3000i X-BeenThere: freebsd-scsi@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: SCSI subsystem List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 08 Jan 2009 15:39:02 -0000 Danny Braniss wrote: >> Danny Braniss wrote: >> >>> wups, didn't read all your mail, sorry, >>> try increasing tht tag opening, in the config file add >>> tag = 64 >>> for example, or via camcontrol >>> >>> danny >>> >>> >> I modify my iscsi.conf by adding tags = 64 (and no tag). Now works fine, >> about 43MB/s. Thank You very match. >> >> But what means this parameter? >> > > is the max number of commands that can be sent without waiting for ack. I set > it > too concervative :-( > I made many tests with different tags values and checked the writing speed with dd. 0 (default) -> ~850kB/s 1 -> ~850kB/s 2 -> ~6.5MB/s 3 -> ~40MB/s 4 -> ~49MB/s 8 -> ~49MB/s ... I think that NAS server sends ACK after 4 commands. Is it correct? Is it possible that the NAS server modifies this behavior? Can I set tags at 4 or is it smarter to keep it higher? Thanks. -- Sossi Andrej ------------------------- DotCom Information technology Via Biancospino, 9 34151 - Opicina (TS) Italy tel: +39 040 2158191 fax: +39 040 0641954 E-mail: asossi@dotcom.ts.it ---------------------------- Ai sensi del D.lgs n. 196 del 30.06.03 (Codice Privacy) si precisa che le informazioni contenute in questo messaggio sono riservate e ad uso esclusivo del destinatario. Qualora il messaggio in parola Le fosse pervenuto per errore, La preghiamo di eliminarlo senza copiarlo e di non inoltrarlo a terzi, dandocene gentilmente comunicazione. Grazie This message, for the D.lgs n. 196 / 30.06.03 (Privacy Code), may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation. From owner-freebsd-scsi@FreeBSD.ORG Thu Jan 8 17:18:27 2009 Return-Path: Delivered-To: freebsd-scsi@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 3195B1065861 for ; Thu, 8 Jan 2009 17:18:27 +0000 (UTC) (envelope-from asossi@dotcom.ts.it) Received: from mail.dotcom.ts.it (89-96-242-204.ip14.fastwebnet.it [89.96.242.204]) by mx1.freebsd.org (Postfix) with SMTP id 69F1F8FC12 for ; Thu, 8 Jan 2009 17:18:25 +0000 (UTC) (envelope-from asossi@dotcom.ts.it) Received: (qmail 51559 invoked by uid 89); 8 Jan 2009 17:18:25 -0000 Received: by simscan 1.1.0 ppid: 51539, pid: 51551, t: 0.4032s scanners:none X-Spam-Checker-Version: SpamAssassin 3.2.3 (2007-08-08) on web.dotcom.ts.it X-Spam-Level: X-Spam-Status: No, hits=-4.1 required=3.0 tests=ALL_TRUSTED,AWL,BAYES_00 autolearn=ham version=3.2.3 Received: from unknown (HELO ?192.168.0.2?) (asossi@dotcom.ts.it@192.168.254.2) by test.dotcom.ts.it with SMTP; 8 Jan 2009 17:18:25 -0000 Message-ID: <496634BB.9050902@dotcom.ts.it> Date: Thu, 08 Jan 2009 18:15:39 +0100 From: Sossi Andrej Organization: DotCom Information technology User-Agent: Thunderbird 2.0.0.18 (X11/20081105) MIME-Version: 1.0 To: freebsd-scsi@freebsd.org Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit Subject: ISCSI and multipath X-BeenThere: freebsd-scsi@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: SCSI subsystem List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 08 Jan 2009 17:18:28 -0000 Another question about MD3000i. My MD3000i have 2 controller and every controller have 2 net interface. My FreeBSD server is connect on NAS server with 2 network cards throw out 2 switch on th 2 networks. (192.168.253.X and 192.168.252.X). In the official manual is written if the primary controller (or net) have a problem the NAS automatically switch disks drive on second controller. For this is required MPIO on Windows or MPP on Linux. It is supported also on FreeBSD? Thanks in advance. -- Sossi Andrej ------------------------- DotCom Information technology Via Biancospino, 9 34151 - Opicina (TS) Italy tel: +39 040 2158191 fax: +39 040 0641954 E-mail: asossi@dotcom.ts.it ---------------------------- Ai sensi del D.lgs n. 196 del 30.06.03 (Codice Privacy) si precisa che le informazioni contenute in questo messaggio sono riservate e ad uso esclusivo del destinatario. Qualora il messaggio in parola Le fosse pervenuto per errore, La preghiamo di eliminarlo senza copiarlo e di non inoltrarlo a terzi, dandocene gentilmente comunicazione. Grazie This message, for the D.lgs n. 196 / 30.06.03 (Privacy Code), may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation. From owner-freebsd-scsi@FreeBSD.ORG Fri Jan 9 01:03:47 2009 Return-Path: Delivered-To: freebsd-scsi@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 8605E106566C for ; Fri, 9 Jan 2009 01:03:47 +0000 (UTC) (envelope-from andrew@modulus.org) Received: from email.octopus.com.au (email.octopus.com.au [122.100.2.232]) by mx1.freebsd.org (Postfix) with ESMTP id 47E048FC12 for ; Fri, 9 Jan 2009 01:03:47 +0000 (UTC) (envelope-from andrew@modulus.org) Received: by email.octopus.com.au (Postfix, from userid 1002) id 8E70D17D8D; Fri, 9 Jan 2009 12:03:46 +1100 (EST) X-Spam-Checker-Version: SpamAssassin 3.2.3 (2007-08-08) on email.octopus.com.au X-Spam-Level: X-Spam-Status: No, score=-1.4 required=10.0 tests=ALL_TRUSTED autolearn=failed version=3.2.3 Received: from [10.1.50.60] (ppp121-44-0-132.lns10.syd7.internode.on.net [121.44.0.132]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) (Authenticated sender: admin@email.octopus.com.au) by email.octopus.com.au (Postfix) with ESMTP id 733CB1732D; Fri, 9 Jan 2009 12:03:42 +1100 (EST) Message-ID: <4966A25E.4070803@modulus.org> Date: Fri, 09 Jan 2009 12:03:26 +1100 From: Andrew Snow User-Agent: Thunderbird 2.0.0.14 (X11/20080523) MIME-Version: 1.0 To: Sossi Andrej References: <496634BB.9050902@dotcom.ts.it> In-Reply-To: <496634BB.9050902@dotcom.ts.it> Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit Cc: freebsd-scsi@freebsd.org Subject: Re: ISCSI and multipath X-BeenThere: freebsd-scsi@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: SCSI subsystem List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 09 Jan 2009 01:03:47 -0000 Sossi Andrej wrote: > Another question about MD3000i. > My MD3000i have 2 controller and every controller have 2 net interface. > My FreeBSD server is connect on NAS server with 2 network cards throw > out 2 switch on th 2 networks. (192.168.253.X and 192.168.252.X). > In the official manual is written if the primary controller (or net) > have a problem the NAS automatically switch disks drive on second > controller. For this is required MPIO on Windows or MPP on Linux. It is > supported also on FreeBSD? If you only wanted to do failover at the network level with shared IP address, you could use lagg. However, since you have different IP subnets for each network, you probably want to look at using geom_fox. You set up two iSCSI connections, one to each controller, and geom_fox on top to make it appear as a single disk device. Hope that helps! - Andrew From owner-freebsd-scsi@FreeBSD.ORG Fri Jan 9 06:13:26 2009 Return-Path: Delivered-To: freebsd-scsi@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 9FA98106566B for ; Fri, 9 Jan 2009 06:13:26 +0000 (UTC) (envelope-from gpalmer@freebsd.org) Received: from noop.in-addr.com (mail.in-addr.com [IPv6:2001:470:8:162::1]) by mx1.freebsd.org (Postfix) with ESMTP id 7770D8FC14 for ; Fri, 9 Jan 2009 06:13:26 +0000 (UTC) (envelope-from gpalmer@freebsd.org) Received: from gjp by noop.in-addr.com with local (Exim 4.54 (FreeBSD)) id 1LLAcW-0002OP-JD; Fri, 09 Jan 2009 01:13:24 -0500 Date: Fri, 9 Jan 2009 01:13:24 -0500 From: Gary Palmer To: Andrew Snow Message-ID: <20090109061324.GD81380@in-addr.com> References: <496634BB.9050902@dotcom.ts.it> <4966A25E.4070803@modulus.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <4966A25E.4070803@modulus.org> Cc: freebsd-scsi@freebsd.org Subject: Re: ISCSI and multipath X-BeenThere: freebsd-scsi@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: SCSI subsystem List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 09 Jan 2009 06:13:27 -0000 On Fri, Jan 09, 2009 at 12:03:26PM +1100, Andrew Snow wrote: > Sossi Andrej wrote: > >Another question about MD3000i. > >My MD3000i have 2 controller and every controller have 2 net interface. > >My FreeBSD server is connect on NAS server with 2 network cards throw > >out 2 switch on th 2 networks. (192.168.253.X and 192.168.252.X). > >In the official manual is written if the primary controller (or net) > >have a problem the NAS automatically switch disks drive on second > >controller. For this is required MPIO on Windows or MPP on Linux. It is > >supported also on FreeBSD? > > If you only wanted to do failover at the network level with shared IP > address, you could use lagg. > > However, since you have different IP subnets for each network, you > probably want to look at using geom_fox. You set up two iSCSI > connections, one to each controller, and geom_fox on top to make it > appear as a single disk device. FreeBSD 7.0 and later also has geom_multipath, which I've been told (no experience) is a better approach. I suggest you review documentation for both and make your own decision Regards, Gary From owner-freebsd-scsi@FreeBSD.ORG Fri Jan 9 06:38:04 2009 Return-Path: Delivered-To: freebsd-scsi@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id B7D09106564A for ; Fri, 9 Jan 2009 06:38:04 +0000 (UTC) (envelope-from danny@cs.huji.ac.il) Received: from kabab.cs.huji.ac.il (kabab.cs.huji.ac.il [132.65.16.84]) by mx1.freebsd.org (Postfix) with ESMTP id 72D748FC0A for ; Fri, 9 Jan 2009 06:38:04 +0000 (UTC) (envelope-from danny@cs.huji.ac.il) Received: from pampa.cs.huji.ac.il ([132.65.80.32]) by kabab.cs.huji.ac.il with esmtp id 1LLB0M-000OZp-Ms; Fri, 09 Jan 2009 08:38:02 +0200 X-Mailer: exmh version 2.7.2 01/07/2005 with nmh-1.2 To: Sossi Andrej In-reply-to: Your message of Thu, 08 Jan 2009 16:36:13 +0100 . Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Fri, 09 Jan 2009 08:38:02 +0200 From: Danny Braniss Message-ID: Cc: freebsd-scsi@freebsd.org Subject: Re: ISCSI slow to read from DELL MD3000i X-BeenThere: freebsd-scsi@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: SCSI subsystem List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 09 Jan 2009 06:38:05 -0000 > Danny Braniss wrote: > >> Danny Braniss wrote: > >> > >>> wups, didn't read all your mail, sorry, > >>> try increasing tht tag opening, in the config file add > >>> tag = 64 > >>> for example, or via camcontrol > >>> > >>> danny > >>> > >>> > >> I modify my iscsi.conf by adding tags = 64 (and no tag). Now works fine, > >> about 43MB/s. Thank You very match. > >> RTFM :-) > >> But what means this parameter? > >> > > > > is the max number of commands that can be sent without waiting for ack. I set > > it > > too concervative :-( > > > I made many tests with different tags values and checked the writing > speed with dd. > 0 (default) -> ~850kB/s > 1 -> ~850kB/s > 2 -> ~6.5MB/s > 3 -> ~40MB/s > 4 -> ~49MB/s > 8 -> ~49MB/s > ... > > I think that NAS server sends ACK after 4 commands. Is it correct? in a simplistic way yes, it depends on how much memory the target has, the speed of its disks, and of course, the software :-) > Is it > possible that the NAS server modifies this behavior? Can I set tags at 4 > or is it smarter to keep it higher? > among many other things, it depends on how much memory you have, but I would set it somewhat higher, 8 or 16. try running several dd's simultaneously. danny