Date: Tue, 22 Aug 2000 15:29:59 +0930 From: Greg Lehey <grog@lemis.com> To: John Lengeling <johnl@raccoon.com> Cc: freebsd-scsi@FreeBSD.ORG Subject: Re: Different init speeds of raid5 plex subdisks under vinum Message-ID: <20000822152959.D46852@wantadilla.lemis.com> In-Reply-To: <39A21314.1107B6AD@raccoon.com>; from johnl@raccoon.com on Tue, Aug 22, 2000 at 12:43:48AM -0500 References: <39A21314.1107B6AD@raccoon.com>
next in thread | previous in thread | raw e-mail | index | archive | help
[Format recovered--see http://www.lemis.com/email/email-format.html] Please limit your text lines to 80 characters or less. Don't limit log output or similar texts. On Tuesday, 22 August 2000 at 0:43:48 -0500, John Lengeling wrote: > I created a raid5 plex under vinum using 3 drives. These are > supposed to be identical drives. They are slightly different in > size. > da2 at ahc0 bus 0 target 2 lun 0 > da2: <IBM DNES-318350W SA30> Fixed Direct Access SCSI-3 device > da2: 80.000MB/s transfers (40.000MHz, offset 31, 16bit), Tagged Queueing Enabled > da2: 17501MB (35843670 512 byte sectors: 255H 63S/T 2231C) > da1 at ahc0 bus 0 target 1 lun 0 > da1: <IBM DNES-318350W SA60> Fixed Direct Access SCSI-3 device > da1: 80.000MB/s transfers (40.000MHz, offset 31, 16bit), Tagged Queueing Enabled > da1: 17366MB (35566501 512 byte sectors: 255H 63S/T 2213C) > da3 at ahc0 bus 0 target 3 lun 0 > da3: <IBM DNES-318350W SA60> Fixed Direct Access SCSI-3 device > da3: 80.000MB/s transfers (40.000MHz, offset 31, 16bit), Tagged Queueing Enabled > da3: 17366MB (35566501 512 byte sectors: 255H 63S/T 2213C) > > When I ran the vinum init command to initialize the plex, one of the > drives da2 completed the initialization 3-4 times faster than da1 and > da3. Something like 20 minutes for da2 versus 60 minutes for > da1/da3. Both da1 and da3 initialized at the same rate. Is this > weird? Yes. > Did da3/da1 get transfers negotiated down to slower speeds? Probably not. > Bad cabling? No. > Or is this normal? For a certain definition of "normal", yes. I've seen this too. It seems to have something to do with the scheduler. Since we've seen it both on IDE and SCSI drives, and normally the request queue is short on such operations, it doesn't appear to be the disk driver. Greg -- When replying to this message, please take care not to mutilate the original text. For more information, see http://www.lemis.com/email.html Finger grog@lemis.com for PGP public key See complete headers for address and phone numbers To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-scsi" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20000822152959.D46852>