Re: 5.0.6 grinds to a complete halt
From: Bela Lubkin (belal_at_sco.com)
Date: 05/25/04
- Next message: pablo hernandez: "only one program per login"
- Previous message: Pineapple: "Open Server 5.0.6 - problem in login"
- In reply to: Barry Swane: "Re: 5.0.6 grinds to a complete halt"
- Next in thread: Bela Lubkin: "using scodb for various purposes, Re: 5.0.6 grinds to a complete halt"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Date: Tue, 25 May 2004 19:48:04 GMT To: scomsc@xenitec.ca
Barry Swane wrote:
> I think we are getting darned close to the root of my problem now.
> I found the following on Seagate Web site:(these drives are ST336607)
>
> -------------------------------------------------------------
> Ultra 320 Time-Out Firmware Upgrade
>
> The Ultra 320 firmware update applies to the following Seagate hard
> drives:
> Cheetah 10K.6: ST3146807; ST373307; ST336607
> Cheetah 15K.3: ST373453; ST336753; ST318453
> Problem description
>
> Some Seagate Cheetah 10K.6 hard drives with OEM firmware up to 0006
> and Cheetah 15K.3 hard drives with OEM firmware up to 0005 on Ultra
> 320 SCSI host adapters are experiencing time out issues when running
> RAID 0, 1 and 5 with some host adapters. This issue has been observed
> using U320 Adaptec and LSI SCSI controllers, but may not limited to
> these host adapter manufacturers.
>
> Root cause
>
> The Cheetah 10K.6 and 15K.3 drives (models listed above) will
> sometimes hang due to an issue in the firmware when reading and
> writing at U320 packetized mode.
>
> Corrective Action
>
> Seagate has modified the firmware and added an additional register
> check while in U320 packetized mode thus preventing the time out issue
> due to system hang.
>
> Contact your host adapter manufacturer for the latest BIOS revision
> for your U320 SCSI controller.
>
> To obtain the unique firmware download certificate number for your
> hard drive, contact Seagate Technical Support via phone or by email at
> discsupport@seagate.com.
>
> Please have your drive part number (Example: 9U8006-001) and the
> current firmware level available when contacting Technical Support.
>
> For the 10K.6 models the new code will be OEM 0007
> For the 15K.3 models the new code will be OEM 0006
>
> Please backup any important files before upgrading the firmware.
> Seagate is not responsible for any data loss.
>
> Copyright )2004, Seagate Technology LLC | About Seagate | Privacy
> Policy | Legal
> --------------------------------------------------
>
> That last comment is a little chilling-- the implication being it
> might blow out the entire RAID 5 hard drive?
You should act as if it is guaranteed to blow away all data on the
drive. Make and verify backups and a bootable crash recovery disk
(`mkdev fd` with native OSR5 tools, or use a supertar to get much
friendlier whole-system-restore media). If the firmware upgrade turns
out to leave your RAID intact, great! You lost an hour or two of time
making the recovery media (which you should have anyway). No comparison
to losing all your data.
The problem Seagate describes might be causing the whole thing. It's in
the right general area. As you say, this is a good place to start.
>Bela<
- Next message: pablo hernandez: "only one program per login"
- Previous message: Pineapple: "Open Server 5.0.6 - problem in login"
- In reply to: Barry Swane: "Re: 5.0.6 grinds to a complete halt"
- Next in thread: Bela Lubkin: "using scodb for various purposes, Re: 5.0.6 grinds to a complete halt"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Relevant Pages
|
