Original question:
We buy Seagate 18G SCSI drives and multi-disk enclosures from Anacapa to use
throughout our UNIX environment (SUN, Tru64 Unix, HP-UX, Siemens, NCR, IBM
etc.,etc.,etc. -- you get the picture). They are bought in bulk as generic
UNIX server accessories.
The Sun systems and to a lesser degree, the HP, are chewing through these
disks at an alarming rate. I just started here Jan. 31, and I have already
seen these disks fail in a major way on at least 4 Sun systems and 2 HPs. I
have been out on SunSolve reading patch descriptions and bug reports with no
really definitive conclusions. I'm really beginning to believe we have a
firmware compatibility issue here.
Anybody got a clue why we might be experiencing such a high rate of failure?
We use Veritas to manage the disks. Most of the volumes we create are
striped but not mirrored.
Thanks to:
jrg@blodwen.demon.co.uk
Caparrosso, Nelson T. [Nelson.T.Caparrosso@Mail.AAS.ameritech.com]
Bill Armand [barmand@flash.net]
Craig Anderson [craiga@ggise.com]
Chad Price [cprice@molbio.unmc.edu]
Mark Baldwin [mark.baldwin@makesys.com]
Al Hopper [al@logical-approach.com]
vogelke@c17mis.region2.wpafb.af.mil
Litwin, Gary [gary.litwin@fsbti.com]
Damir Delija [ddelija@srce.hr]
Phil Antoine [antoine@radonc.duke.edu]
Brooke King [bking@yipes.com]
Buddy Lumpkin [BLumpkin@ijapan.com]
Dan Sweet [sweet@qiclab.scn.rain.com]
Krongard, Aaron [akrongard@aztectech.com]
Coffindaffer, Virginia [Virginia.Coffindaffer@wang.com]
Walter Reed [walt@stockmaster.com]
The most compelling opinion seems to be that operating temperature is the
main reason these drives fail. Indeed, equipment is stacked willy-nilly in
our cabinets with little regard for air-flow needs. So we have decided that
we will take the first downtime opportunity we have to do some slight
rearrangment to allow for maximum ventilation. We will also be
investigating the purchase of fans for our cabinets (currently, none.) We
will also check the suggestions of the other kind respondees.
I will include all the messages following this.
Annette Lee
BMC Software, IS/UNIX
San Jose CA
(408) 546-1030
===========================================
From: jrg@blodwen.demon.co.uk
Sent: Monday, May 15, 2000 9:15 AM
To: Lee, Annette
Subject: Re: Seagate 18G drives
by far and away the most usual thing with disks is lack of adequate
cooling in whatever disk cabinets they're in. and I'm sure 18GB drives
are no better than 7200rpm 4-9GB drives in that respect.
(I'm not familiar with the Anacapa enclosures, I'm afraid)
James.
===========================================
From: Caparrosso, Nelson T. [Nelson.T.Caparrosso@Mail.AAS.ameritech.com]
Sent: Tuesday, May 09, 2000 8:02 AM
To: 'Lee, Annette'
Subject: RE: Seagate 18G drives
If it start with SCSI transport errors -- then you most likely have a
cabling problem - either (1) faulty cabling/termination or (2) you've maxed
out your TOTAL SCSI CABLE length. I usually don't go beyond 2 SCSI cable
lengths or if I must, use a high quality SCSI cable. This is most true if
you 're using just plain SCSI controllers.
Hope this helps.
===========================================
From: Caparrosso, Nelson T.
[mailto:Nelson.T.Caparrosso@Mail.AAS.ameritech.com]
Sent: Monday, May 08, 2000 1:22 PM
To: 'Lee, Annette'
Subject: RE: Seagate 18G drives
Strange.... you should not have any problems with your non-SUN drives.
Possibilities:
- make sure the disks you buy are *new*, change vendors if you can or
you were just unlucky with the batch... but they should have warranty..
- make sure your enclosures are properly ventilated and stable
- make sure your connectors (cables) etc, do'nt get easily disturbed.
You did not describe how your disks have been failing???
Nelson
===========================================
From: Bill Armand [barmand@flash.net]
Sent: Monday, May 08, 2000 2:23 PM
To: Lee, Annette
Subject: Re: Seagate 18G drives
What is the temperature in your lab??
From: Craig Anderson [craiga@ggise.com]
Sent: Monday, May 08, 2000 1:31 PM
To: 'Lee, Annette'
Subject: RE: Seagate 18G drives
We have had great luck with the Seagate 18GBs. We do however do a full
format and surface analysis on each drive before putting it into
surface.
It may be anal, it may take a long time, but it does put our minds at
ease, and it HAS caught a few bad block, and disks over the last
11-years
I've been doing it.
Also, when you order a drive or any other part CPU/NIC/etc. from a VAR,
make sure they order an individually boxed one from the distributor
instead of a bagged one that came out of a box of 20/30/etc.
Craig
----------------------------------------------------------------------
| Craig H. Anderson | craiga@ggise.com |
| Systems Administrator-------------------------+----------------------|
| Genesis Group / | www.ggise.com |
| Biological Research Associates | www.biolresearch.com |
|----------------------------------------------------------------------|
| Voice/(813) 620-4500 FAX/(813) 620-4980 Tampa, FL 33619 |
----------------------------------------------------------------------
===========================================
From: Chad Price [cprice@molbio.unmc.edu]
Sent: Tuesday, May 09, 2000 7:00 AM
To: Lee, Annette
Subject: Re: Seagate 18G drives
I ran a couple of Seagate 18's in an external enclosure for over a year
with no problems. I'd guess that you have just had bad luck on the
Suns/HPs.
(I moved them over the a DEC Alpha system recently because I needed the
space there. No problems have surfaced.)
cp
Chad Price
Systems Manager, Genetic Sequence Analysis Facility
University of Nebraska Medical Center
986495 Nebraska Medical Center
Omaha, NE 68506-6495
cprice@molbio.unmc.edu
(402) 559-9527
(402) 559-4077 (FAX)
===========================================
From: Mark Baldwin [mark.baldwin@makesys.com]
Sent: Tuesday, May 09, 2000 6:21 AM
To: Lee, Annette
Subject: Re: Seagate 18G drives
If these are LVD drives and you are attaching them to a single-ended
controller, be sure to set the jumper on the drive to forced
single-ended. I have had problems with these same drives (LVD) when
attached to a single-ended controller. The drives are suppose to sense
what type of controller they are attched to and automatically adjust. I
found that after setting the jumper to force the drives to single-ended,
they worked fine. There is more information about this at
http://www.seagate.com:80/support/kb/disc/ultra2_scsi_lvd.html
Regards, Mark...
-- Mark A. Baldwin System and Network Administrator Make Systems, Inc. Cary, NC 27511 mark.baldwin@makesys.com www.makesys.com=========================================== From: Al Hopper [al@logical-approach.com] Sent: Tuesday, May 09, 2000 2:01 AM To: Lee, Annette Subject: Re: Seagate 18G drives
Your drives *must* be running too hot. I don't know anything about Anacapa, but measure the temp of the disk drives while they are operating and you'll have your answer.
PS: Fluke make a nice IR non-contact temp probe that can be used in conjunction with their digital multimeters that is ideal for this task.
Regards,
Al Hopper Logical Approach Inc, Plano, TX. al@logical-approach.com Voice: 972-379-2133 Fax: 972-379-2134 aka The Project From Hell Person From: vogelke@c17mis.region2.wpafb.af.mil Sent: Monday, May 08, 2000 5:27 PM To: Annette_Lee@bmc.com Subject: Re: Seagate 18G drives ===========================================
It might be firmware, but I've seen or heard of drives going to hell in a hurry for two other reasons.
a. We had several drives in a Sun Storage Array crap out, one right after another. We'd pull one, and a new one would fail. It turned out to be the power supply in the Storage Array, so I'd look for power problems somewhere.
b. Drives sitting on an extremely hard surface (marble floor, etc.) can crap out due to vibration, even with rubber pads on the boxes. It's rare, but it did take out a bunch of drives at the New York Public Library some years ago.
-- Karl Vogel ASC/YCOA, Wright-Patterson AFB, OH 45433, USA vogelke@c17mis.region2.wpafb.af.mil or kvogel@sumaria.com
"How do you feel?" "Me?" "I believe it's customary to ask after the health of one recently plugged three times." "Thanks for asking." "Don't get sentimental." --Agents Rosenfield and Cooper, "Twin Peaks" =========================================== From: Litwin, Gary [gary.litwin@fsbti.com] Sent: Monday, May 08, 2000 8:39 PM To: 'Lee, Annette' Subject: RE: Seagate 18G drives
Annette - I purchased a bunch of 9GB Seagate drives in enclosures from WEX about 2 years ago, and had what I felt was a rather high initial failure rate. My failures were mostly disks, but I found a couple of intermittent enclosures too. Are you finding as well that the problems are mostly disks?
When I got replacement disk drives under warranty from Seagate, they all worked...
- Gary L ===========================================
From: Damir Delija [ddelija@srce.hr] Sent: Monday, May 08, 2000 10:11 PM To: Lee, Annette Subject: Re: Seagate 18G drives
Enviromental conditions ? Power ? Cabling ?
I've got impression that problem started suddelnly can you corelate this with some other event ?
Was there any change in load profile for this disks ? If there is are firmware problems it usually comes out when diffrent types of disks are on the same scsi at high utilisations ... Also it is not nesecarry to be disks once we have badly pllugged controler on E250 which caused that disk failures at hish scsi utilisations
As far as I now such disks need very stable enviroment I've seen once that system failure (not disks but it is same idea) was caused by hot air generated from other system SUN E450 was grilling the Alpha server, because they were too close
I hope this helps
Damir Delija
Damir Delija
e-mail: ddelija@srce.hr home page: http://jagor.srce.hr/~ddelija
=========================================== From: Phil Antoine [antoine@radonc.duke.edu] Sent: Monday, May 08, 2000 5:16 PM To: Lee, Annette Subject: Re: Seagate 18G drives
Firmware will likely not cause disks to fail (irreparably as your message implies). Generally if they work flawlessly for any reasonable period (a few hours to a couple days) then you're home free. Heat, on the other hand, will cause irreparable damage. You'll likely find that you may have inadequate ventilation for those hotplates disguised as disks. Those things are too hot to handle without generous ventilation. If you're putting them in older chassis' that don't list the 18Gb disks as options at the Sun Store, its likely the media bay ventilation isn't spec'd for that much heat.
FYI, Phil Antoine
Phil Antoine Department of Radiation Oncology Duke University Medical Center Room 04216 Red Zone Hospital South Durham, NC 27710 USA (919)660-2186(Office) (919)970-4461(Pager) =========================================== From: Brooke King [bking@yipes.com] Sent: Monday, May 08, 2000 4:45 PM To: Annette_Lee@bmc.com Subject: RE: Seagate 18G drives
I don't recall what kind of drives you specified, but if they are 10000 RPM they'll definitely be sensitive to heat, as in they need to dissipate a lot of heat.
--
Brooke King direct: +1.415.901.2207 fax: +1.415.901.2201 brooke.king@yipes.com Yipes Communications 114 Sansome Street, San Francisco, CA 94104 USA =========================================== From: Buddy Lumpkin [BLumpkin@ijapan.com] Sent: Monday, May 08, 2000 1:44 PM To: 'Lee, Annette' Subject: RE: Seagate 18G drives
striped but not mirrored? That's begging for a problem you should at least opt for RAID 5, but a mirror is ideal.
Buddy Lumpkin Sr. Systems/Network Administrator Ijapan.com Inc. (425) 709-2900 Ext. 13 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Contrary to popular opinion, UNIX is user friendly... It just happens to be highly selective about who it makes friends with ===========================================
From: Brooke King [bking@yipes.com] Sent: Monday, May 08, 2000 1:19 PM To: Annette_Lee@bmc.com Subject: Re: Seagate 18G drives
Maybe Anacapa doesn't perform very good quality control on its drives. Some vendors, e.g. Metastor and NetApp, claim to perform better quality control on incoming drives than do some of the computer OEMs. If true, this could be your issue.
I have gotten nearly all my drives for Suns from Sun. The only time I had a rash of failures was in a batch of workstations that took their Seagates from the same lot that later came to be known by Sun to have problems. Every one of them failed within a few months of acquisition. Happily, all the workstations had no local data other than some network and application customizations that were easy to reproduce. I wonder if all or most of the failing Anacapa-supplied drives were from the same Seagate lot.
Certainly, if this is a problem you should be mirroring or using RAID 5 or something to make individual disk failures less likely to cause loss of access to data. Simple stripes are probably the worst thing BMC could be doing.
Brooke King direct: +1.415.901.2207 fax: +1.415.901.2201 brooke.king@yipes.com Yipes Communications 114 Sansome Street, San Francisco, CA 94104 USA =========================================== From: Dan Sweet [sweet@qiclab.scn.rain.com] Sent: Monday, May 08, 2000 2:43 PM To: Lee, Annette Subject: Re: Seagate 18G drives
Bad as it sounds.... Seagate sucks. I have moved to IBM Ultrastar series drives only.
I was bit hard by seagate cheetaha's. Had TWO of five in a raid 5 strip set fail at the same time. Cost over $7000 to have ontrack recover the data not to mention the down time.
Another one of the seagate failed on an NT box.
All three had the same problem. Track 0 died.
Over the last few years I have run the range on drives. Quantum 100% Failure rate, Seagate about 40% failure rate,
I have yet to have an IBM fail...
good luck.
dan =========================================== From: Krongard, Aaron [akrongard@aztectech.com] Sent: Monday, May 08, 2000 1:21 PM To: 'Lee, Annette' Subject: RE: Seagate 18G drives
Annette, Overheating or poor power supplies are the two prominant reasons that come to mind. Make sure the drives are not selected to supply term power to the bus as well.
Good luck Aaron Krongard
=========================================== From: Coffindaffer, Virginia [Virginia.Coffindaffer@wang.com] Sent: Monday, May 08, 2000 1:05 PM To: 'Lee, Annette' Subject: RE: Seagate 18G drives
Did you reseat these disks after you received them ?
Virginia Coffindaffer Program Manager Wang Government Services Tel: 703-481-4804 FAX: 703 689-4759 =========================================== From: Walter Reed [walt@stockmaster.com] Well, you didn't provide any details on the disk enclosures from anacapa, but I would suspect that they don't have good enough ventilation. Seagate drives are VERY sensitive to heat. I have stopped using Seagate because of this and now only use IBM (IMHO, IBM is about two years ahead of any other drive manufacturer.)
Here is a paper from IBM that talks about this (keep in mind that it also has a marketing spin on it.) http://www.storage.ibm.com/hardsoft/diskdrdl/technolo/drivetemp/drivetemp.ht m
You may want to actually stick a temp probe near the drives inside the case (you will need to have the case cover on to make sure that you are getting the correct airflow.) and watch it for a day. See if the anacapa enclosures have a higher temp than a normal PC for example (or another drive enclosure that has NOT had failures.) If they do, complain to anacapa that their enclosures are designed like crap, and then find a new vendor that has designed their systems correctly.
From: jrg@blodwen.demon.co.uk Sent: Monday, May 15, 2000 9:15 AM To: Lee, Annette Subject: Re: Seagate 18G drives
by far and away the most usual thing with disks is lack of adequate cooling in whatever disk cabinets they're in. and I'm sure 18GB drives are no better than 7200rpm 4-9GB drives in that respect.
(I'm not familiar with the Anacapa enclosures, I'm afraid)
James. =========================================== From: Caparrosso, Nelson T. [Nelson.T.Caparrosso@Mail.AAS.ameritech.com] Sent: Tuesday, May 09, 2000 8:02 AM To: 'Lee, Annette' Subject: RE: Seagate 18G drives
If it start with SCSI transport errors -- then you most likely have a cabling problem - either (1) faulty cabling/termination or (2) you've maxed out your TOTAL SCSI CABLE length. I usually don't go beyond 2 SCSI cable lengths or if I must, use a high quality SCSI cable. This is most true if you 're using just plain SCSI controllers.
Hope this helps. ===========================================
From: Caparrosso, Nelson T. [mailto:Nelson.T.Caparrosso@Mail.AAS.ameritech.com] Sent: Monday, May 08, 2000 1:22 PM To: 'Lee, Annette' Subject: RE: Seagate 18G drives
Strange.... you should not have any problems with your non-SUN drives. Possibilities:
- make sure the disks you buy are *new*, change vendors if you can or you were just unlucky with the batch... but they should have warranty.. - make sure your enclosures are properly ventilated and stable - make sure your connectors (cables) etc, do'nt get easily disturbed.
You did not describe how your disks have been failing???
Nelson ===========================================
From: Bill Armand [barmand@flash.net] Sent: Monday, May 08, 2000 2:23 PM To: Lee, Annette Subject: Re: Seagate 18G drives
What is the temperature in your lab??
From: Craig Anderson [craiga@ggise.com] Sent: Monday, May 08, 2000 1:31 PM To: 'Lee, Annette' Subject: RE: Seagate 18G drives
We have had great luck with the Seagate 18GBs. We do however do a full format and surface analysis on each drive before putting it into surface. It may be anal, it may take a long time, but it does put our minds at ease, and it HAS caught a few bad block, and disks over the last 11-years I've been doing it.
Also, when you order a drive or any other part CPU/NIC/etc. from a VAR, make sure they order an individually boxed one from the distributor instead of a bagged one that came out of a box of 20/30/etc.
Craig ---------------------------------------------------------------------- | Craig H. Anderson | craiga@ggise.com | | Systems Administrator-------------------------+----------------------| | Genesis Group / | www.ggise.com | | Biological Research Associates | www.biolresearch.com | |----------------------------------------------------------------------| | Voice/(813) 620-4500 FAX/(813) 620-4980 Tampa, FL 33619 | ---------------------------------------------------------------------- ===========================================
From: Chad Price [cprice@molbio.unmc.edu] Sent: Tuesday, May 09, 2000 7:00 AM To: Lee, Annette Subject: Re: Seagate 18G drives
I ran a couple of Seagate 18's in an external enclosure for over a year with no problems. I'd guess that you have just had bad luck on the Suns/HPs.
(I moved them over the a DEC Alpha system recently because I needed the space there. No problems have surfaced.)
cp
Chad Price Systems Manager, Genetic Sequence Analysis Facility University of Nebraska Medical Center 986495 Nebraska Medical Center Omaha, NE 68506-6495 cprice@molbio.unmc.edu (402) 559-9527 (402) 559-4077 (FAX) =========================================== From: Mark Baldwin [mark.baldwin@makesys.com] Sent: Tuesday, May 09, 2000 6:21 AM To: Lee, Annette Subject: Re: Seagate 18G drives
If these are LVD drives and you are attaching them to a single-ended controller, be sure to set the jumper on the drive to forced single-ended. I have had problems with these same drives (LVD) when attached to a single-ended controller. The drives are suppose to sense what type of controller they are attched to and automatically adjust. I found that after setting the jumper to force the drives to single-ended, they worked fine. There is more information about this at http://www.seagate.com:80/support/kb/disc/ultra2_scsi_lvd.html
Regards, Mark...
-- Mark A. Baldwin System and Network Administrator Make Systems, Inc. Cary, NC 27511 mark.baldwin@makesys.com www.makesys.com
=========================================== From: Al Hopper [al@logical-approach.com] Sent: Tuesday, May 09, 2000 2:01 AM To: Lee, Annette Subject: Re: Seagate 18G drives
Your drives *must* be running too hot. I don't know anything about Anacapa, but measure the temp of the disk drives while they are operating and you'll have your answer.
PS: Fluke make a nice IR non-contact temp probe that can be used in conjunction with their digital multimeters that is ideal for this task.
Regards,
Al Hopper Logical Approach Inc, Plano, TX. al@logical-approach.com Voice: 972-379-2133 Fax: 972-379-2134 aka The Project From Hell Person From: vogelke@c17mis.region2.wpafb.af.mil Sent: Monday, May 08, 2000 5:27 PM To: Annette_Lee@bmc.com Subject: Re: Seagate 18G drives ===========================================
It might be firmware, but I've seen or heard of drives going to hell in a hurry for two other reasons.
a. We had several drives in a Sun Storage Array crap out, one right after another. We'd pull one, and a new one would fail. It turned out to be the power supply in the Storage Array, so I'd look for power problems somewhere.
b. Drives sitting on an extremely hard surface (marble floor, etc.) can crap out due to vibration, even with rubber pads on the boxes. It's rare, but it did take out a bunch of drives at the New York Public Library some years ago.
-- Karl Vogel ASC/YCOA, Wright-Patterson AFB, OH 45433, USA vogelke@c17mis.region2.wpafb.af.mil or kvogel@sumaria.com
"How do you feel?" "Me?" "I believe it's customary to ask after the health of one recently plugged three times." "Thanks for asking." "Don't get sentimental." --Agents Rosenfield and Cooper, "Twin Peaks" =========================================== From: Litwin, Gary [gary.litwin@fsbti.com] Sent: Monday, May 08, 2000 8:39 PM To: 'Lee, Annette' Subject: RE: Seagate 18G drives
Annette - I purchased a bunch of 9GB Seagate drives in enclosures from WEX about 2 years ago, and had what I felt was a rather high initial failure rate. My failures were mostly disks, but I found a couple of intermittent enclosures too. Are you finding as well that the problems are mostly disks?
When I got replacement disk drives under warranty from Seagate, they all worked...
- Gary L ===========================================
From: Damir Delija [ddelija@srce.hr] Sent: Monday, May 08, 2000 10:11 PM To: Lee, Annette Subject: Re: Seagate 18G drives
Enviromental conditions ? Power ? Cabling ?
I've got impression that problem started suddelnly can you corelate this with some other event ?
Was there any change in load profile for this disks ? If there is are firmware problems it usually comes out when diffrent types of disks are on the same scsi at high utilisations ... Also it is not nesecarry to be disks once we have badly pllugged controler on E250 which caused that disk failures at hish scsi utilisations
As far as I now such disks need very stable enviroment I've seen once that system failure (not disks but it is same idea) was caused by hot air generated from other system SUN E450 was grilling the Alpha server, because they were too close
I hope this helps
Damir Delija
Damir Delija
e-mail: ddelija@srce.hr home page: http://jagor.srce.hr/~ddelija
=========================================== From: Phil Antoine [antoine@radonc.duke.edu] Sent: Monday, May 08, 2000 5:16 PM To: Lee, Annette Subject: Re: Seagate 18G drives
Firmware will likely not cause disks to fail (irreparably as your message implies). Generally if they work flawlessly for any reasonable period (a few hours to a couple days) then you're home free. Heat, on the other hand, will cause irreparable damage. You'll likely find that you may have inadequate ventilation for those hotplates disguised as disks. Those things are too hot to handle without generous ventilation. If you're putting them in older chassis' that don't list the 18Gb disks as options at the Sun Store, its likely the media bay ventilation isn't spec'd for that much heat.
FYI, Phil Antoine
Phil Antoine Department of Radiation Oncology Duke University Medical Center Room 04216 Red Zone Hospital South Durham, NC 27710 USA (919)660-2186(Office) (919)970-4461(Pager) =========================================== From: Brooke King [bking@yipes.com] Sent: Monday, May 08, 2000 4:45 PM To: Annette_Lee@bmc.com Subject: RE: Seagate 18G drives
I don't recall what kind of drives you specified, but if they are 10000 RPM they'll definitely be sensitive to heat, as in they need to dissipate a lot of heat.
--
Brooke King direct: +1.415.901.2207 fax: +1.415.901.2201 brooke.king@yipes.com Yipes Communications 114 Sansome Street, San Francisco, CA 94104 USA =========================================== From: Buddy Lumpkin [BLumpkin@ijapan.com] Sent: Monday, May 08, 2000 1:44 PM To: 'Lee, Annette' Subject: RE: Seagate 18G drives
striped but not mirrored? That's begging for a problem you should at least opt for RAID 5, but a mirror is ideal.
Buddy Lumpkin Sr. Systems/Network Administrator Ijapan.com Inc. (425) 709-2900 Ext. 13 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Contrary to popular opinion, UNIX is user friendly... It just happens to be highly selective about who it makes friends with ===========================================
From: Brooke King [bking@yipes.com] Sent: Monday, May 08, 2000 1:19 PM To: Annette_Lee@bmc.com Subject: Re: Seagate 18G drives
Maybe Anacapa doesn't perform very good quality control on its drives. Some vendors, e.g. Metastor and NetApp, claim to perform better quality control on incoming drives than do some of the computer OEMs. If true, this could be your issue.
I have gotten nearly all my drives for Suns from Sun. The only time I had a rash of failures was in a batch of workstations that took their Seagates from the same lot that later came to be known by Sun to have problems. Every one of them failed within a few months of acquisition. Happily, all the workstations had no local data other than some network and application customizations that were easy to reproduce. I wonder if all or most of the failing Anacapa-supplied drives were from the same Seagate lot.
Certainly, if this is a problem you should be mirroring or using RAID 5 or something to make individual disk failures less likely to cause loss of access to data. Simple stripes are probably the worst thing BMC could be doing.
Brooke King direct: +1.415.901.2207 fax: +1.415.901.2201 brooke.king@yipes.com Yipes Communications 114 Sansome Street, San Francisco, CA 94104 USA =========================================== From: Dan Sweet [sweet@qiclab.scn.rain.com] Sent: Monday, May 08, 2000 2:43 PM To: Lee, Annette Subject: Re: Seagate 18G drives
Bad as it sounds.... Seagate sucks. I have moved to IBM Ultrastar series drives only.
I was bit hard by seagate cheetaha's. Had TWO of five in a raid 5 strip set fail at the same time. Cost over $7000 to have ontrack recover the data not to mention the down time.
Another one of the seagate failed on an NT box.
All three had the same problem. Track 0 died.
Over the last few years I have run the range on drives. Quantum 100% Failure rate, Seagate about 40% failure rate,
I have yet to have an IBM fail...
good luck.
dan =========================================== From: Krongard, Aaron [akrongard@aztectech.com] Sent: Monday, May 08, 2000 1:21 PM To: 'Lee, Annette' Subject: RE: Seagate 18G drives
Annette, Overheating or poor power supplies are the two prominant reasons that come to mind. Make sure the drives are not selected to supply term power to the bus as well.
Good luck Aaron Krongard
=========================================== From: Coffindaffer, Virginia [Virginia.Coffindaffer@wang.com] Sent: Monday, May 08, 2000 1:05 PM To: 'Lee, Annette' Subject: RE: Seagate 18G drives
Did you reseat these disks after you received them ?
Virginia Coffindaffer Program Manager Wang Government Services Tel: 703-481-4804 FAX: 703 689-4759 =========================================== From: Walter Reed [walt@stockmaster.com] Well, you didn't provide any details on the disk enclosures from anacapa, but I would suspect that they don't have good enough ventilation. Seagate drives are VERY sensitive to heat. I have stopped using Seagate because of this and now only use IBM (IMHO, IBM is about two years ahead of any other drive manufacturer.)
Here is a paper from IBM that talks about this (keep in mind that it also has a marketing spin on it.) http://www.storage.ibm.com/hardsoft/diskdrdl/technolo/drivetemp/drivetemp.ht m
You may want to actually stick a temp probe near the drives inside the case (you will need to have the case cover on to make sure that you are getting the correct airflow.) and watch it for a day. See if the anacapa enclosures have a higher temp than a normal PC for example (or another drive enclosure that has NOT had failures.) If they do, complain to anacapa that their enclosures are designed like crap, and then find a new vendor that has designed their systems correctly.
This archive was generated by hypermail 2.1.2 : Fri Sep 28 2001 - 23:14:08 CDT