|
Home > Archive > Data Storage > May 2005 > RAID-5 of 4x IBM IC35L073UCDY10 incompatible with SEAGATE ST373207LC
You are viewing an archived Text-only version of the thread.
To view this thread in it's original format and/or if you want to reply to
this thread please [click here]
| Author |
RAID-5 of 4x IBM IC35L073UCDY10 incompatible with SEAGATE ST373207LC
|
|
| Victor Mataré 2005-05-03, 5:47 pm |
| Hi,
I'm running a RAID-5 of 4x IBM IC35L073UCDY10 on an ICP Vortex GDT8514RZ
SCSI/RAID-Controller, firmware 2.44.02-RC7B (current). It is to be
expanded by 2 SEAGATE ST373207LC, but the expand command in the
controller's console fails with the message "no suitable disks found".
This seems to be due to a compatibility issue of the IBM drives. The IBM
support doesn't seem to be of much help - the firmware update (S29A)
they sent me made no difference.
If anyone has experience with expanding RAID arrays of IBM disks or
similar situations, please tell me...
thx in advance
--
Victor Mataré
* Server- & Network-Administration *
Lehrstuhl für Ingenieurgeologie und Hydrogeologie der RWTH-Aachen
Lochnerstr. 4-20
52064 Aachen
Germany
http://www.lih.rwth-aachen.de
Ph: +49-0241-80-97130
Fx: +49-0241-80-92280
| |
| Victor Matare 2005-05-03, 5:47 pm |
| John-Paul Stewart wrote:
> Victor Mataré wrote:
>
>
>
> Some RAID controllers simply refuse to work unless all drives are
> identical brand and model. I don't think it's a compatability issue
> with the IBM drives so much as it's an issue with the controller. Check
> with the controller manufacturer for information; there's a chance they
> might have a workaround (if this is indeed your problem).
The controller definitely does support using different drives in a
single array. It even leaves a "tolerance gap" of about 10 MB of unused
space at the end of the disks to ensure that drives with a slightly
smaller capacity can still be added to the array. I'm pretty sure it's
an incompatibility issue with either the IBM- or the Seagate-disks.
If this can't be resolved, I'd appreciate some tips on how to find out
who (IBM, Seagate or maybe ICP Vortex) has been messing with the
standards there, in order to find someone to "blame" for this ;)
thx4help,
--
Victor Mataré
* Server- & Network-Administration *
Lehrstuhl für Ingenieurgeologie und Hydrogeologie der RWTH-Aachen
Lochnerstr. 4-20
52064 Aachen
Germany
http://www.lih.rwth-aachen.de
Ph: +49-0241-80-97130
Fx: +49-0241-80-92280
| |
| Victor Matare 2005-05-04, 7:45 am |
| Victor Matare wrote:
> John-Paul Stewart wrote:
>
>
>
> The controller definitely does support using different drives in a
> single array. It even leaves a "tolerance gap" of about 10 MB of unused
> space at the end of the disks to ensure that drives with a slightly
> smaller capacity can still be added to the array. I'm pretty sure it's
> an incompatibility issue with either the IBM- or the Seagate-disks.
> If this can't be resolved, I'd appreciate some tips on how to find out
> who (IBM, Seagate or maybe ICP Vortex) has been messing with the
> standards there, in order to find someone to "blame" for this ;)
>
> thx4help,
>
Update:
The guy who told me about that "tolerance gap" was wrong. I called ICP
support and they said that if the disks being added are only a single
sector smaller, the controller won't accept them. This seems to be the
reason for my problem, as the Seagate drive are actually a few sectors
shorter than the IBM drives. Tonight I'm gonna rebuild the whole array
with slightly reduced sizes for the individual disks. If that doesn't
work out, I'll post another call for help 
bye,
--
Victor Mataré
* Server- & Network-Administration *
Lehrstuhl für Ingenieurgeologie und Hydrogeologie der RWTH-Aachen
Lochnerstr. 4-20
52064 Aachen
Germany
http://www.lih.rwth-aachen.de
Ph: +49-0241-80-97130
Fx: +49-0241-80-92280
| |
| Victor Mataré 2005-05-04, 7:45 am |
| Victor Matare wrote:
> John-Paul Stewart wrote:
>
>
>
> The controller definitely does support using different drives in a
> single array. It even leaves a "tolerance gap" of about 10 MB of unused
> space at the end of the disks to ensure that drives with a slightly
> smaller capacity can still be added to the array. I'm pretty sure it's
> an incompatibility issue with either the IBM- or the Seagate-disks.
> If this can't be resolved, I'd appreciate some tips on how to find out
> who (IBM, Seagate or maybe ICP Vortex) has been messing with the
> standards there, in order to find someone to "blame" for this ;)
>
> thx4help,
>
Update:
The guy who told me about that "tolerance gap" was wrong. I called ICP
support and they said that if the disks being added are only a single
sector smaller, the controller won't accept them. This seems to be the
reason for my problem, as the Seagate drive are actually a few sectors
shorter than the IBM drives. Tonight I'm gonna rebuild the whole array
with slightly reduced sizes for the individual disks. If that doesn't
work out, I'll post another call for help
bye,
--
Victor Mataré
* Server- & Network-Administration *
Lehrstuhl für Ingenieurgeologie und Hydrogeologie der RWTH-Aachen
Lochnerstr. 4-20
52064 Aachen
Germany
http://www.lih.rwth-aachen.de
Ph: +49-0241-80-97130
Fx: +49-0241-80-92280
| |
| Kasper Dupont 2005-05-04, 7:45 am |
| Victor Matare wrote:
>
> It even leaves a "tolerance gap" of about 10 MB of unused
> space at the end of the disks to ensure that drives with a slightly
> smaller capacity can still be added to the array.
But 10MB is not nearly enough. I have seen a difference
of about 3.6GB between two drives both sold as 160GB.
--
Kasper Dupont -- der bruger for meget tid på usenet.
Note to self: Don't try to allocate 256000 pages
with GFP_KERNEL on x86.
| |
| Arno Wagner 2005-05-04, 5:46 pm |
| In comp.sys.ibm.pc.hardware.storage Victor Mataré <v.matare@lih.rwth-aachen.de> wrote:
> Victor Matare wrote:
[vbcol=seagreen]
> Update:
> The guy who told me about that "tolerance gap" was wrong. I called ICP
> support and they said that if the disks being added are only a single
> sector smaller, the controller won't accept them. This seems to be the
> reason for my problem, as the Seagate drive are actually a few sectors
> shorter than the IBM drives. Tonight I'm gonna rebuild the whole array
> with slightly reduced sizes for the individual disks. If that doesn't
> work out, I'll post another call for help
Having at least one of the seagates in there when you rebuild should
reduce the size automatically.
Arno
|
|
|
|
|