Hello again,
I received 30 responses to my query, all of them very useful. The responses ranged from "This is quite normal" to "Replace the disks immediately".
Since the data on my 2 IPI disks is not critical and does not change frequently I decided to reformat the disks once again and continue using them. If the period between successive reformats drops below 6 months I will probably replace the disks with SCSI ones.
I found a patch regarding IPI drives, 101008-01 for SunOs 4.1.3 which apparently fixes a problem with false error messages related IPI drives under heavy load. My IPI drives aren't really under heavy load but I applied it the same. The other factor that may have produced the errors I was experiencing is the fact about a month ago a new piece of equipment was installed in the same room where our 4/470 is. During this installation people walked in and out of the room often leaving the door open, this caused
some temperature variation. From the messages I collected it sounds like IPI disks are quite sensitive.
We do not have a maintenance contract, most of you thought I did.....:-{
I would like to thank the following people for their fast responses, some started arriving only after 3 hours I had posted my problem !
mrs@rdcs.Kodak.COM (Mike Salehi {CPU}) reformat them & don't worry
Gary L. Berger" <gberger@eng.clemson.edu> reformat them
Ron Hall <thorn@cc.mcgill.ca> replace them
Jeff Mallory <jeff@access.digex.net> keep them in a cool room
paulo@parana.dcc.unicamp.br (Paulo Licio de Geus) don't worry
Mark Allyn (206) 860-9454 <allyn@allyn.com> replace & get SCSI
rknight@rrt.Arco.COM (Randy Knight) replace & get SCSI
ken@rocky.alaska.opensys.COM (Ken Lamoreaux) replace them
itubjs@lux.levels.unisa.edu.au (itubjs) replace them, 1.3Gb IPI drives are ok
bkk@edh.ericsson.se (Bjoern Karlsen) don't worry, watch out for overheating
bismark@alta.jpl.nasa.gov (Bismark Espinoza) asked for more info
mhr@internet.sbi.com (Michael Ringbauer) replace them
Susan M. Menig" <ddq251f@shoes.Bell-Atl.Com> replace before it dies
thys@dsy-srv3.cern.ch (Thys Antoine /CN/CE) replace them
root@wisdom.maf.nasa.gov (Mark Hargrave (504) 257-1242) don't worry
riley@stavanger.Geco-Prakla.slb.com (Mark Riley) replace with SCSI
merola@norden.com (joe merola) SCSI is quite cheap...
brw@hertz.njit.edu (Brian White) replace them
Ross.Stocks.INSDRS01@nt.com reformat & don't worry
Istvan Fekete <feksa@panix.com> asks more info
sdr@rdga3.att.com (S. D. Raffensberger 52882 (RD)) don't worry
Pell Emanuelsson <pell@lysator.liu.se> suggests use of patch
Richard Wong" <rnw@math.Princeton.EDU> replace them
davee@lightning.mitre.org (David N. Edwards) watch out for temp variations
Mike Rembis 66520 <ebumfr@ebu.ericsson.se> not worth the hassle, replace
Phil Antoine <antoine@RadOnc.Duke.EDU> the drives are pieces of .... replace or get a SS1000 :-)
Glenn.Satchell@uniq.com.au (Glenn Satchell they are getting near the end of their life
Mark.McIntosh@Engr.uvic.ca has the same problem
Thanks again...
-Martin
Martin Achilli - martin@gea.hsr.it
Consiglio Nazionale delle Ricerche - I.N.B.
c/o Medicina Nucleare, Osp. San Raffaele
20132 Milano, Italy
tel: +39/2/26433648 fax: ../26415202
********************************************************************************
>From primus.rdcs.Kodak.COM!mrs Thu Mar 2 18:43:58 1995
Date: Thu, 2 Mar 95 11:49:27 EST
From: mrs@rdcs.Kodak.COM (Mike Salehi {CPU})
To: martin@gea.hsr.it
Subject: Re: IPI disk drives - experience with
Content-Length: 2470
X-Lines: 56
Status: RO
Martin,
They are supposed to last a lot more than that mine have not failed
in 3 years. But I did run into your problem a while back.
The solution was, to format from the format program and restore
from backups, if you do not have backups take a bakup before formatting, it
will be a good backup despite error messages and the delay.
>.> From sun-managers-relay@ra.mcs.anl.gov Thu Mar 2 11:42:45 1995
>.> Sender: sun-managers-relay@ra.mcs.anl.gov
>.> Date: Thu, 2 Mar 95 14:04 MET
>.> From: martin@gea.hsr.it (Martin Achilli)
>.> Reply-To: martin@gea.hsr.it (Martin Achilli)
>.> Followup-To: junk
>.> To: sun-managers@eecs.nwu.edu
>.> Subject: IPI disk drives - experience with
>.> Cc: martin@gea.hsr.it
>.> Content-Length: 1172
>.>
>.> Hello,
>.> I have a Sun 4/470 with SunOs 4.1.3 32Mb RAM, 4 669Mb SCSI HDs and
>.> 2 911Mb Sun IPI HDs in an a/c room at 20 degrees connected to a UPS. The machine is about 5 years old. I have reformatted the IPI drives at the following intervals, I was getting the "Data Retry Performed" (see below) errors at the time:
>.> - machine new
>.> - after 3 1/2 years
>.> - after 1 year (25 august 1994)
>.>
>.> now, after only 7 months I am getting the "Data Retry Performed" errors again. Can anyone who has specifically had experience with IPI drives tell me the life expectancy of these drives ? Are they about to crash, or should I reformat again and not worry ?
>.>
>.> The drives are called id000 and id001, the errors are:
>.>
>.> Oct 31 13:44:19 zeus vmunix: id001g: block 1057606 (1350262 abs): read: Conditional Success. Data Retry Performed.
>.> Oct 31 13:44:20 zeus vmunix: id001g: block 1057666 (1350322 abs): read: Conditional Success. Data Retry Performed.
>.> ..many many of these/day..
>.>
>.> I will SUMMARIZE,
>.> thanks Martin
>.>
>.> Martin Achilli - martin@gea.hsr.it
>.> Consiglio Nazionale delle Ricerche - I.N.B.
>.> c/o Medicina Nucleare, Osp. San Raffaele
>.> 20132 Milano, Italy
>.> tel: +39/2/26433648 fax: ../26415202
>.>
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
Mike (Mehran) Salehi Still, but not for long, working for CIBER(CPU) at Kodak
Mail mrs@rdcs.kodak.com Phone (716)253-9446 Pager (716) 781-3920 USA
However, never daunted, I will cope with adversity in my traditional
manner ... sulking and nausea.
-- Tom K. Ryan
Mar 2 11:49:28
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
>From eng.clemson.edu!Gary.Berger Thu Mar 2 20:06:57 1995
Date: Thu, 2 Mar 1995 14:03:41 -0500 (EST)
From: "Gary L. Berger" <gberger@eng.clemson.edu>
X-Sender: gberger@wildcat
To: Martin Achilli <martin@gea.hsr.it>
Subject: Re: IPI disk drives - experience with
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Content-Length: 1640
X-Lines: 39
Status: RO
We have one as well and I am experiencing the same problem. I just reformat
and backup frequently. Kind of a pain but I think the oxide on the platters
is showing age.
--------------------------------
Gary L. Berger
Senior Systems Administrator
Advanced Systems Support
--------------------------------
Riggs Hall, Room 10
803-656-1228
gberger@eng.clemson.edu
On Thu, 2 Mar 1995, Martin Achilli wrote:
> Hello,
> I have a Sun 4/470 with SunOs 4.1.3 32Mb RAM, 4 669Mb SCSI HDs and
> 2 911Mb Sun IPI HDs in an a/c room at 20 degrees connected to a UPS. The machine is about 5 years old. I have reformatted the IPI drives at the following intervals, I was getting the "Data Retry Performed" (see below) errors at the time:
> - machine new
> - after 3 1/2 years
> - after 1 year (25 august 1994)
>
> now, after only 7 months I am getting the "Data Retry Performed" errors again. Can anyone who has specifically had experience with IPI drives tell me the life expectancy of these drives ? Are they about to crash, or should I reformat again and not worry ?
>
> The drives are called id000 and id001, the errors are:
>
> Oct 31 13:44:19 zeus vmunix: id001g: block 1057606 (1350262 abs): read: Conditional Success. Data Retry Performed.
> Oct 31 13:44:20 zeus vmunix: id001g: block 1057666 (1350322 abs): read: Conditional Success. Data Retry Performed.
> ..many many of these/day..
>
> I will SUMMARIZE,
> thanks Martin
>
> Martin Achilli - martin@gea.hsr.it
> Consiglio Nazionale delle Ricerche - I.N.B.
> c/o Medicina Nucleare, Osp. San Raffaele
> 20132 Milano, Italy
> tel: +39/2/26433648 fax: ../26415202
>
>From amaretto.CC.McGill.CA!thorn Thu Mar 2 20:17:51 1995
Date: Thu, 2 Mar 1995 14:21:02 -0500 (EST)
From: Ron Hall <thorn@cc.mcgill.ca>
To: Martin Achilli <martin@gea.hsr.it>
Subject: Re: IPI disk drives - experience with
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Content-Length: 842
X-Lines: 18
Status: RO
On Thu, 2 Mar 1995, Martin Achilli wrote:
> Hello,
> I have a Sun 4/470 with SunOs 4.1.3 32Mb RAM, 4 669Mb SCSI HDs and
> 2 911Mb Sun IPI HDs in an a/c room at 20 degrees connected to a UPS. The machine is about 5 years old. I have reformatted the IPI drives at the following intervals, I was getting the "Data Retry Performed" (see below) errors at the time:
> - machine new
> - after 3 1/2 years
> - after 1 year (25 august 1994)
Toast`em. We had them for 9 months and many replacements. Finally
told Sun, replace them with something else. They are real dogs.
Ron Hall | DISCLAIMER: I said it. I must've meant it. There is
Analyst | nobody else to blame.
<thorn@cc.mcgill.ca> | Clown:n: A man who acts too natural.
+1 514 398 3718 | "Whenever doubt can be called into question, do so."
>From ra.mcs.anl.gov!sun-managers-relay Thu Mar 2 20:50:46 1995
Sender: sun-managers-relay@ra.mcs.anl.gov
Date: Thu, 2 Mar 95 14:04 MET
From: martin@gea.hsr.it (Martin Achilli)
Reply-to: martin@gea.hsr.it (Martin Achilli)
Followup-to: junk
To: sun-managers@eecs.nwu.edu
Subject: IPI disk drives - experience with
Cc: martin@gea.hsr.it
Content-Length: 1172
X-Lines: 23
Status: RO
Hello,
I have a Sun 4/470 with SunOs 4.1.3 32Mb RAM, 4 669Mb SCSI HDs and
2 911Mb Sun IPI HDs in an a/c room at 20 degrees connected to a UPS. The machine is about 5 years old. I have reformatted the IPI drives at the following intervals, I was getting the "Data Retry Performed" (see below) errors at the time:
- machine new
- after 3 1/2 years
- after 1 year (25 august 1994)
now, after only 7 months I am getting the "Data Retry Performed" errors again. Can anyone who has specifically had experience with IPI drives tell me the life expectancy of these drives ? Are they about to crash, or should I reformat again and not worry ?
The drives are called id000 and id001, the errors are:
Oct 31 13:44:19 zeus vmunix: id001g: block 1057606 (1350262 abs): read: Conditional Success. Data Retry Performed.
Oct 31 13:44:20 zeus vmunix: id001g: block 1057666 (1350322 abs): read: Conditional Success. Data Retry Performed.
..many many of these/day..
I will SUMMARIZE,
thanks Martin
Martin Achilli - martin@gea.hsr.it
Consiglio Nazionale delle Ricerche - I.N.B.
c/o Medicina Nucleare, Osp. San Raffaele
20132 Milano, Italy
tel: +39/2/26433648 fax: ../26415202
>From access.digex.net!jeff Thu Mar 2 23:10:42 1995
Date: Thu, 2 Mar 1995 17:07:34 -0500
From: Jeff Mallory <jeff@access.digex.net>
To: martin@gea.hsr.it
Subject: IPI drives
Content-Length: 640
X-Lines: 13
Status: RO
The cooler you can keep the drives, the longer they avoid the data retry
syndrome, though amount of use seems to be a factor, too. Our IPI's will
start steadily increasing the number of data retry errors they get,
until we re-format them, then they;re just fine for a while. I don't
think the periodic resurrection of the retry errors are an indication
of a bad drive by themselves, but if there are other errors, drive failure
should be looked in to. We got some newer IPI drives just before Sun
discontinued them and they have never shown the retry errors. If you have
a Sun service contract you can ask that they be replaced.
JJM
>From parana.dcc.unicamp.br!paulo Fri Mar 3 00:26:08 1995
Date: Thu, 2 Mar 1995 20:21:18 -0300
From: paulo@parana.dcc.unicamp.br (Paulo Licio de Geus)
To: martin@gea.hsr.it (Martin Achilli)
Subject: IPI disk drives - experience with
Content-Length: 745
X-Lines: 16
Status: RO
My 4/390 is 4.5 years old and the 1 GB IPI is still holding fine, I've
never had to reformat it because of error msgs, only occasionally
during a OS upgrade. Anyway, it's been a while since I last did this.
Well, I hope it doesn't need this in the near future, otherwise I'll
have to retire it before it's "scheduled" demise in a few months time:
the console serial interface fried together with a PC interface that
was its console. Since then I'm running it headless with a high level in
it's RXD pin.
-- postmaster/manager Paulo Licio de Geus INTERNET: paulo@dcc.unicamp.br Depto de Ciencia da Computacao voice: +55 192 39-3115/8695/8442 DCC - IMECC - UNICAMP fax: +55 192 39-7470/5808 caixa postal: 6065 13081-970 Campinas SP Brazil>From allyn.com!allyn Fri Mar 3 02:16:51 1995 From: Mark Allyn (206) 860-9454 <allyn@allyn.com> Subject: Re: IPI disk drives - experience with To: martin@gea.hsr.it Date: Thu, 2 Mar 1995 17:20:14 -0800 (PST) X-Mailer: ELM [version 2.4 PL23] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 863 X-Lines: 22 Status: RO
Hello!
Guess what! I just had an IPI drive replaced because it was getting these stupid errors about once every second and it was doeing horrors to the performance to the machine.
The machine is an Oracle database server which is used to store Sunnet Manager MIB maps for a large private internet; these maps have to be constructed and updated every night off shift so that the NOC's can have them by the time prime shift begins. These data retry's were slowing the system down and causing the map builds to creep into the morning hours.
This is the second time this has happened within two years. I think that there might be some sore of media related problem with the drives. It allways starts out very gradually and gets progressively worse as time goes on.
The scsi drives on this same machine have never had such a problem.
Mark Allyn allyn@allyn.com
>From rrt.Arco.COM!rknight Fri Mar 3 03:47:58 1995 From: rknight@rrt.Arco.COM (Randy Knight) Subject: Re: IPI disk drives - experience with To: martin@gea.hsr.it Date: Thu, 2 Mar 1995 20:44:44 -0600 (CST) X-Mailer: ELM [version 2.4 PL24] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 300 X-Lines: 7 Status: RO
I recently went thru the same exercise with an IPI drive on a 470, I hoped that the problem was with the controller since the errors were recoverable. But to my surprise, a new drive fixed the problem If you have had the drives for 5 years, you have been lucky. Replace the drives.(with scsi's)
RK
>From rambone.psi.net!denali!rocky.alaska.opensys.COM!ken Fri Mar 3 04:04:45 1995 Date: Thu, 2 Mar 1995 09:30:56 -0900 From: ken@rocky.alaska.opensys.COM (Ken Lamoreaux) To: martin@gea.hsr.it Subject: Re: IPI disk drives - experience with X-Sun-Charset: US-ASCII Content-Length: 1030 X-Lines: 25 Status: RO
Martin,
The IPI drives, because of their size (they are huge!) have a known problem with the heads going out of alignment. It gets prograssivly worse as your experience shows. The only thing to do is to replace the drive or have it refurbished.
The length of the head "arm" was just to long. Gravity and movement takes its tole.
If you've got the funds, have pity on guys like me that have to replace these when they are defective (I am a Sun Field Eng.) Replace the drives with SCSI disks! :) ------------------------------------------------- _/_/_/_/_/ _/_/_/_/_/ _/_/_/_/_/ _/ _/ _/ _/ _/ _/ _/_/_/_/_/ _/ _/ _/ _/ _/ _/_/_/_/_/ _/ _/_/_/_/_/ _/ _/_/_/_/_/ _/ ------------------------------------------------- Kenneth P. Lamoreaux ken@alaska.opensys.COM OpenSystems, Inc. Ph: (907) 261-8700 3111 C Street #400 Fax: (907) 261-8710 Anchorage, AK 99503 Direct: (907) 261-8716 -------------------------------------------------
>From lux.levels.unisa.edu.au!itubjs Fri Mar 3 04:59:35 1995 From: itubjs@lux.levels.unisa.edu.au (itubjs) Subject: Re: IPI disk drives - experience with To: martin@gea.hsr.it Date: Fri, 3 Mar 1995 14:26:30 +1030 (CST) X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 1361 X-Lines: 28 Status: RO
Hello, In Sun Managers you wrote: > I have a Sun 4/470 with SunOs 4.1.3 32Mb RAM, 4 669Mb SCSI HDs and >2 911Mb Sun IPI HDs in an a/c room at 20 degrees connected to a UPS. >[...] >now, after only 7 months I am getting the "Data Retry Performed" errors >again. Can anyone who has specifically had experience with IPI drives tell >me the life expectancy of these drives ? Are they about to crash, or should >I reformat again and not worry ? We have 2 900Mb IPI drives and 3 1.3Gb drives.
Our experience was that we had to replace the early 900Mb drives after a few years due to one type of failure or another. The 1.3Gb drives have been fine.
>The drives are called id000 and id001, the errors are: >Oct 31 13:44:19 zeus vmunix: id001g: block 1057606 (1350262 abs): read: > Conditional Success. Data Retry Performed. Get format to map out those blocks. If the problem surfaces the next day repeat the mapping out operation. If the problem comes again in a day or two then things are looking flaky, the drive may need replacing, call an engineer :-)
Bevin Steer, Systems & Networks Team, ------------ ------------ Bevin.Steer@unisa.edu.au Phone: +61-8-302-3271 Fax: 302 3577 Information Technology Unit, University of SA., The Levels, SA 5095, Australia
>From edh.ericsson.se!bkk Fri Mar 3 08:37:58 1995 Date: Fri, 3 Mar 1995 08:35:59 --100 From: bkk@edh.ericsson.se (Bjoern Karlsen) To: martin@gea.hsr.it Subject: Re: IPI disk drives - experience with X-Sun-Charset: US-ASCII Content-Length: 698 X-Lines: 15 Status: RO
[snip]
> now, after only 7 months I am getting the "Data Retry Performed" errors again
I got the same error message on 2 of my 4 911 MB IPI disks. They came after I moved the cabinet under a table, and close to a wall. It was approx. 30 cm (1 foot) between the disk fans and the wall. I tried to reformat the disks, but the problem returned in about 2-3 weeks.
I replaced the disks with 2 new SCSI disks, and moved the cabinet to another location. The IPI disks generates a lot of heat. Make sure that the exhaust from the disk fans does not blow into a wall or something. The 2 remaining IPI-disks are still running happily. All 4 are about the same age -- 4 years.
-kkc- bkk@edh.ericsson.se
>From alta.jpl.nasa.gov!bismark Thu Mar 2 18:12:44 1995 Date: Thu, 2 Mar 95 09:11:11 PST From: bismark@alta.jpl.nasa.gov (Bismark Espinoza) To: martin@gea.hsr.it Subject: Re: IPI disk drives - experience with Cc: bismark@alta.jpl.nasa.gov Content-Length: 159 Status: RO X-Lines: 5
How fragementized is id001g ? How full is it? What is the Company and model number? I would just run fsck on that one and avoid reformatting unless necessary.
>From internet.sbi.com!mhr Thu Mar 2 17:41:18 1995 Date: Thu, 2 Mar 95 11:36:32 EST From: mhr@internet.sbi.com (Michael Ringbauer) To: martin@gea.hsr.it Subject: Re: IPI disk drives - experience with Content-Length: 3259 Status: RO X-Lines: 70
Hi....
If your IPIs are 3-5 years old you may want to have Sun replace them with new ones if you've reformatted them a few times. The heads gradually start to misalign after a while and that's where the error messages start coming from. As long as you're getting the Conditional Success messages and they're intermittent you'll be okay. Eventually they'll increase in number probably based on how heavily your system is used. If you start seeing retry failed messages or a large number occurring you should replace the drive.
We have (or had) a large number of 4/490s with IPIs that were installed approximately four years ago and in the last year they've experienced a higher incidence of those messages. Depending on how critical the system was (most are in production as Sybase database servers for financial data) we tended to have Sun replace the IPIs with brand new ones. It wasn't worth the time or effort to re-format the drive and reload the data only to have it re-appear a few months later. I've spent a number of weekends in the last couple of years having to do this and putting in a new drive meant I didn't have to worry about it again. Of course have 20-30 of them spread among 7 servers meant the errors started appearing on other drives once I replaced the bad ones...:-)
I don't know if you have a maintenance agreement with Sun (or other vendor) and what it says about handling these types of replacements but it's worth a shot. Or retire them and get SCSI drives.... I don't know if you're planning to retire the 4/470 soon but the SCSIs and their data will be easier to migrate to another machine.
Good luck.....
Mike
> From sun-managers-relay@ra.mcs.anl.gov Thu Mar 2 10:37:52 1995 > Sender: sun-managers-relay@ra.mcs.anl.gov > Date: Thu, 2 Mar 95 14:04 MET > From: martin@gea.hsr.it (Martin Achilli) > Reply-To: martin@gea.hsr.it (Martin Achilli) > Followup-To: junk > To: sun-managers@eecs.nwu.edu > Subject: IPI disk drives - experience with > Cc: martin@gea.hsr.it > Content-Length: 1172 > > Hello, > I have a Sun 4/470 with SunOs 4.1.3 32Mb RAM, 4 669Mb SCSI HDs and > 2 911Mb Sun IPI HDs in an a/c room at 20 degrees connected to a UPS. The machine is about 5 years old. I have reformatted the IPI drives at the following intervals, I was getting the "Data Retry Performed" (see below) errors at the time: > - machine new > - after 3 1/2 years > - after 1 year (25 august 1994) > > now, after only 7 months I am getting the "Data Retry Performed" errors again. Can anyone who has specifically had experience with IPI drives tell me the life expectancy of these drives ? Are they about to crash, or should I reformat again and not worry ? > > The drives are called id000 and id001, the errors are: > > Oct 31 13:44:19 zeus vmunix: id001g: block 1057606 (1350262 abs): read: Conditional Success. Data Retry Performed. > Oct 31 13:44:20 zeus vmunix: id001g: block 1057666 (1350322 abs): read: Conditional Success. Data Retry Performed. > ..many many of these/day.. > > I will SUMMARIZE, > thanks Martin > > Martin Achilli - martin@gea.hsr.it > Consiglio Nazionale delle Ricerche - I.N.B. > c/o Medicina Nucleare, Osp. San Raffaele > 20132 Milano, Italy > tel: +39/2/26433648 fax: ../26415202 >
>From bagate.BELL-ATL.COM!bagate!shoes!ddq251f Thu Mar 2 17:36:05 1995 Subject: Re: IPI disk drives - experience with To: martin@gea.hsr.it Date: Thu, 2 Mar 1995 11:32:44 -0500 (EST) From: "Susan M. Menig" <ddq251f@shoes.Bell-Atl.Com> Organization: Bell Atlantic - Open Systems Integration and Support Reply-To: Susan Menig <Susan.M.Menig@Bell-Atl.Com> X-Mailer: ELM [version 2.4 PL22] MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=US-ASCII Content-Length: 2240 Status: RO X-Lines: 46
> Martin Achilli wrote: > >Hello, > I have a Sun 4/470 with SunOs 4.1.3 32Mb RAM, 4 669Mb SCSI HDs and >2 911Mb Sun IPI HDs in an a/c room at 20 degrees connected to a UPS. The machine is about 5 years old. I have reformatted the IPI drives at the following intervals, I was getting the "Data Retry Performed" (see below) errors at the time: >- machine new >- after 3 1/2 years >- after 1 year (25 august 1994) > >now, after only 7 months I am getting the "Data Retry Performed" errors again. Can anyone who has specifically had experience with IPI drives tell me the life expectancy of these drives ? Are they about to crash, or should I reformat again and not worry ?> > >The drives are called id000 and id001, the errors are: > >Oct 31 13:44:19 zeus vmunix: id001g: block 1057606 (1350262 abs): read: Conditional Success. Data Retry Performed. >Oct 31 13:44:20 zeus vmunix: id001g: block 1057666 (1350322 abs): read: Conditional Success. Data Retry Performed. >..many many of these/day.. > >I will SUMMARIZE, > thanks Martin > >Martin Achilli - martin@gea.hsr.it >Consiglio Nazionale delle Ricerche - I.N.B. >c/o Medicina Nucleare, Osp. San Raffaele >20132 Milano, Italy >tel: +39/2/26433648 fax: ../26415202 > Martin,
I've had those kind of errors with the same drives on a 690 and I would recommend you replace the drive while you have a chance - before it dies. I honestly don't know how long it would last. It could be all right for a day, a year, ...? It depends on how critical the availability of the machine is. The site the 690 was at wasn't willing to risk putting the engineers whose stuff was on that machine out of business for a day or so (it was the system disk that was complaining). They preferred that I order a replacement (they had a maintenance contract) and schedule down time to install it. Of course, you'll want to make sure you're getting good backups off of it. I had to go back a month to get a clean backup of the OS. The system files were backed up once a month and the latest had problems because of all the read errors.
Sue -- =**= - =*= - =*= - =*= - =*= - =*= - =*= - =*= - =*= - =*= - =*= - =*= - =**= Susan M. Menig Advanced Systems Consulting, Inc. Susan.M.Menig@Bell-Atl.Com
>From dsy-srv3.cern.ch!thys Fri Mar 3 11:55:47 1995 From: thys@dsy-srv3.cern.ch (Thys Antoine /CN/CE) Date: Fri, 3 Mar 95 11:52:58 +0100 To: martin@gea.hsr.it Subject: Re: IPI disk drives - experience with Content-Length: 1143 X-Lines: 25 Status: RO
Hello,
we have also a 4/470 with 4 911Mb Sun IPI disks, the environment is also a/c room at about 20 degrees and UPS. We get up to now about every year one to two disks that start to give the kind of errors you mention. It starts with one or two errors a day and climbs over weeks to several errors an hour. When the error rate goes higher (tens of errors in an hour) we call Sun to replace the disk (we have a maintenance contract). Reformatting makes it better but does not remove the cause of the errors. The story told to me says that the diskcoating itself starts to lose it properties (oxydation? hence the strong request to put the disks in a cool environment). Even if I still feel that the IPI speed is very good we will give up those disk because they are so unreliable.
regards, Thys A.
---------------------------------------------------------------------- Antoine Thys e_mail : thys@dxcern.cern.ch CERN CN/CE tel. : +41 22 767 49 18 CH-1211 Geneva 23 fax : +41 22 767 71 55 ----------------------------------------------------------------------
>From wisdom.maf.nasa.gov!root Fri Mar 3 13:45:11 1995 Date: Fri, 3 Mar 95 06:42:01 CST From: root@wisdom.maf.nasa.gov (Mark Hargrave (504) 257-1242) To: martin@gea.hsr.it Subject: Re: IPI disk drives - experience with Content-Length: 477 X-Lines: 16 Status: RO
We have had our IPI drives for approx. 5 years now. We get these errors all the time. Our local FE said that these errors are normal and does not indicate any hardware problems.
Thanks, Mark
------------------------------------------ Mark Hargrave, Unix Systems Manager Martin Marietta Manned Space Systems PO Box 29304 Mail Stop: DPI/Bin 41 New Orleans, LA 70189
Phone: 504-257-1242 E-Mail: meh@wisdom.maf.nasa.gov ------------------------------------------
>From wisdom.maf.nasa.gov!root Fri Mar 3 13:46:13 1995 Date: Fri, 3 Mar 95 06:43:10 CST From: root@wisdom.maf.nasa.gov (Mark Hargrave (504) 257-1242) To: martin@gea.hsr.it Subject: Re: IPI disk drives - experience with Content-Length: 79 X-Lines: 5 Status: RO
Martin,
I've also tried formatting the disk. The errors still continue.
Mark
>From stavanger.Geco-Prakla.slb.com!riley Fri Mar 3 14:44:30 1995 Date: Fri, 3 Mar 95 12:25:55 +0100 From: riley@stavanger.Geco-Prakla.slb.com (Mark Riley) To: martin@gea.hsr.it Subject: Re: IPI disk drives - experience with Content-Length: 276 X-Lines: 12 Status: RO
Hi there,
We are sick and tired of our 4 IPI drives failing - we are replacing them with SCSI disks. My general opion is YICK!
Mark Riley Senior System Programmer Processing, Geco-Prakla, Stavanger, Norway email: riley@stavanger.Geco-Prakla.slb.com. phone: (NOR) 5150 6437
>From norden.com!merola Fri Mar 3 15:13:57 1995 Date: Fri, 3 Mar 1995 09:09:31 -0500 From: merola@norden.com (joe merola) To: martin@gea.hsr.it Subject: Re: IPI disk drives - experience with X-Sun-Charset: US-ASCII Content-Length: 1112 X-Lines: 27 Status: RO
We had about 3.8 GB of IPI drives which on our 470 which began failing after about 4 years (roughly late 93). We started to see precisely the same problem. In time, the warning went from 'conditional success' to 'uncorrectable error'. We had to reformat every 4 to 6 months. Another danger is that a failure could occur during a tape dump, giving you a trash backup tape (this happened to us also.) We had a maintenance contract, called our service, who replaced the drives with "refurbished" drives, but these failed also. As this was a mission-critical system, we replaced them with a BoxHill RAID stack, and *now* use the IPI's ONLY for nfs-mounted swap space on the network.
The bottom line is the drive surface media has an early end of life, in our experience. Don't beat yourself up trying to save them... you'll get bit. I've spoken to other owners of similar systems - they laughed - in sympathy - having experienced the same troubles. I don't know how much disk you have, but scsi's pretty cheap these days.
Hope this helps.
Joe Merola merola@norden.com Norden Systems Inc. Norwalk, CT
>From hertz.njit.edu!brw Fri Mar 3 15:24:31 1995 From: brw@hertz.njit.edu (Brian White) Date: Fri, 3 Mar 1995 09:20:14 -0500 X-Mailer: Mail User's Shell (7.2.5 10/14/92) To: martin@gea.hsr.it (Martin Achilli) Subject: Re: IPI disk drives - experience with Cc: brw@hertz.njit.edu Content-Length: 2064 X-Lines: 44 Status: RO
On Mar 2, 2:04pm, Martin Achilli wrote: } Subject: IPI disk drives - experience with } Hello, } I have a Sun 4/470 with SunOs 4.1.3 32Mb RAM, 4 669Mb SCSI HDs and } 2 911Mb Sun IPI HDs in an a/c room at 20 degrees connected to a UPS. The machine is about 5 years old. I have reformatted the IPI drives at the following intervals, I was getting the "Data Retry Performed" (see below) errors at the time: } - machine new } - after 3 1/2 years } - after 1 year (25 august 1994) } } now, after only 7 months I am getting the "Data Retry Performed" errors } again. Can anyone who has specifically had experience with IPI drives tell } me the life expectancy of these drives ? Are they about to crash, or should } I reformat again and not worry ? } } The drives are called id000 and id001, the errors are: } } Oct 31 13:44:19 zeus vmunix: id001g: block 1057606 (1350262 abs): read: Conditional Success. Data Retry Performed. } Oct 31 13:44:20 zeus vmunix: id001g: block 1057666 (1350322 abs): read: Conditional Success. Data Retry Performed. } ..many many of these/day.. } } I will SUMMARIZE, } thanks Martin
We have IPIs on both Sun systems and Ultrix systems. The 4/690 that has them was kicking out errors like these for months without a problem. We've had problems with the IPIs on both types of hardware - the only thing that ultimately fixed them was replacement; reformats (our hardware repair service insists we do this first) gave us a few error-free weeks, then the messages started up again.
Note that even with write errors, we were able to run for a while without system crashes and data losses, but we were lucky. Make sure you have good backups!
Given the time you've had the systems (MTBF starts to drop at 5 years), it appears to be new drive time, Martin. If it will cost you real money (i.e., you have no maintenance), I suggest you buy SCSI drives. The access differential is minimal and you'll get a better financial deal.
Brian
Brian White, System Manager New Jersey Institute of Technology brw@hertz.njit.edu brw@njit.edu
>From nt.com!/DD.ID=INSDRS01/G=Ross/S=Stocks/ Fri Mar 3 15:35:43 1995 X400-Received: by mta NT.COM in /PRMD=NT/ADMD=MCI/C=US/; Relayed; Fri, 3 Mar 1995 14:31:25 +0000 X400-Received: by /PRMD=NT/ADMD=MCI/C=US/; Relayed; Fri, 3 Mar 1995 14:31:12 +0000 X400-Received: by /PRMD=nt/ADMD=mci/C=us/; Relayed; Fri, 3 Mar 1995 14:33:08 +0000 Date: Fri, 3 Mar 1995 14:33:08 +0000 X400-Originator: Ross.Stocks.INSDRS01@nt.com X400-Recipients: non-disclosure:; X400-MTS-Identifier: [/PRMD=NT/ADMD=MCI/C=US/;<9503031433.AA05110@npdcs2.DDPSP] X400-Content-Type: P2-1984 (2) Content-Identifier: Re: IPI disk ... From: Ross.Stocks.INSDRS01@nt.com To: martin <martin%gea.hsr.it@NT400.nt.com> Subject: Re: IPI disk drives - experience with Content-Length: 2279 X-Lines: 54 Status: RO
Martin:
I have the same problems. I am told it is just a limitation of the IPI drives. I would just reformat and not worry about it. Also I would not reformat (unless you have time to burn) until the number of messages becomes objectionable or until you encounter an uncorrectable error. If after reformatting, the errors return immediately, then it is time to replace the drive.
I keep of log of the messages. They ususally start out slowly and build geometrically, sort of. Mine usually reach a threshold after which the number increases dramatically.
I have 21 IPI drives and see a wide variance of performance. Some last a very long time, some not long at all.
You might be inclined to try fixing the block, something you can do from the format command, but I have found that to be only marginally effective and not really worth the effort.
If by "many many / day" you mean more than a couple dozen, I would plan to reformat fairly soon. But I don't think you are in any danger of a meltdown.
Hope this helps, Ross ross.stocks@nt.com
> Hello, > I have a Sun 4/470 with SunOs 4.1.3 32Mb RAM, 4 669Mb SCSI HDs and > 2 911Mb Sun IPI HDs in an a/c room at 20 degrees connected to a UPS. The machine is about 5 years old. I have reformatted the IPI drives at the following intervals, I was getting the "Data Retry Performed" (see below) errors at the time: > - machine new > - after 3 1/2 years > - after 1 year (25 august 1994) > > now, after only 7 months I am getting the "Data Retry Performed" errors again. Can anyone who has specifically had experience with IPI drives tell me the life expectancy of these drives ? Are they about to crash, or should I reformat again and not worry ? > > The drives are called id000 and id001, the errors are: > > Oct 31 13:44:19 zeus vmunix: id001g: block 1057606 (1350262 abs): read: Conditional Success. Data Retry Performed. > Oct 31 13:44:20 zeus vmunix: id001g: block 1057666 (1350322 abs): read: Conditional Success. Data Retry Performed. > ..many many of these/day.. > > I will SUMMARIZE, > thanks Martin > > Martin Achilli - martin@gea.hsr.it > Consiglio Nazionale delle Ricerche - I.N.B. > c/o Medicina Nucleare, Osp. San Raffaele > 20132 Milano, Italy > tel: +39/2/26433648 fax: ../26415202 > >
>From panix.com!feksa Fri Mar 3 15:38:13 1995 Date: Fri, 3 Mar 1995 09:33:56 -0500 From: Istvan Fekete <feksa@panix.com> To: martin@gea.hsr.it Subject: Re: IPI disk drives - experience with Newsgroups: info.sun-managers Content-Length: 48 X-Lines: 4 Status: RO
How many bad blocks you have per drive?
Feksa
>From rdga3.att.com!sdr Fri Mar 3 16:01:40 1995 From: sdr@rdga3.att.com (S. D. Raffensberger 52882 (RD)) To: martin@gea.hsr.it Date: Thu, 2 Mar 95 09:48:44 EST Original-From: rdga3!sdr (S. D. Raffensberger 52882 (RD)) Original-To: gea.hsr.it!martin Subject: Re: IPI disk drives - experience with Content-Length: 734 X-Lines: 21 Status: RO
Hi Martin,
I run a 690MP with four 1.3G Seagate Wren IPI drives. These drives were available as both IPI and SCSI drives. I don't believe reliability is as dependent on the interface as it is on the basic drive technology/construction.
To answer your question, these have been running at about 6% overall usage based on the data I collect from "iostat". This means that they are being used (read or write) 6% of their uptime. They are three years old. I have never re- formatted them. I have seen the errors you see, but only two or three a year.
You probably can only replace your drives with used ones. I would look for a nice, cheap 3G SCSI drive and sell the IPI's to one of the refurbishing houses.
Steve Raffensberger AT&T
>From lysator.liu.se!pell Fri Mar 3 16:22:44 1995 Date: Fri, 3 Mar 1995 16:19:25 +0100 (MET) From: Pell Emanuelsson <pell@lysator.liu.se> X-Sender: pell@godot To: Martin Achilli <martin@gea.hsr.it> Subject: Re: IPI disk drives - experience with MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Content-Length: 2210 X-Lines: 53 Status: RO
It could be a problem with the disks, but it's also a known symptom from a problem with the firmware in the IPI-controller. Under heavy load, and especially with more than one disk per controller, the OS will get these error messages from the drive (and sometimes more serious ones as well). We are trying to get new proms from Sun for our controllers, but we haven't gotten them yet so we can't confirm if that will fix the problem, although the following notes so suggest.
Cheers - Pell === >From bugid 1036367:
Synopsis: Conditional success, motion retry messages Summary: On heavy disk activity following messages keep popping up on the console: Apr 9 13:43:46 muishu vmunix: id003h: block 849694 (849694 abs): write: Conditional Success. Data Retry Performed. === >From bugid 1139738:
Description: Under load the customer machine is failing with IPI controller faults. Customer is running 4.1.3, and has patch 101008-01 installed. Patch 101008-01 installed. This was supposed to have been the fix... See bugid 1119466 (now closed)..
Error messages: Jul 12 16:31:19 gp_golden vmunix: id023d: block 1273486 (1280116 abs): read: Conditional Success. Data Retry Performed. Jul 13 09:55:13 gp_golden vmunix: id001e: block 481182 (1657266 abs): read: Conditional Success. Data Retry Performed. Aug 4 10:22:26 gp_golden vmunix: id002f: block 700692 (1723896 abs): read: Alternate Port Exception. Facility switched to other port. Summary: With extreme I/O load on a system with many IPI controllers and IPI disk drives, you may notice some I/O requests getting aborted with VME timeout error. Work around: The following will help in reducing the problem: 1. Distribution of disks and load across the controllers. 2. Use of new IPI firmware prom (level 10, 1855 Rev 4, with FW revision dated = 2/10/93 ') 3. Use of Segate IPI ZBR Elite drives instead of CDC 9720/9722 drives. ===
Lysator Computer Society | email: pell@lysator.liu.se Linkoping University | WWW home at: http://www.lysator.liu.se/~pell/ Sweden | or: http://www.cs.tu-berlin.de/~pell/
>From math.Princeton.EDU!rnw Fri Mar 3 16:41:37 1995 From: "Richard Wong" <rnw@math.Princeton.EDU> Date: Fri, 3 Mar 1995 10:19:32 -0500 To: sun-managers@eecs.nwu.edu, martin@gea.hsr.it Subject: Re: IPI disk drives - experience with Content-Length: 1935 X-Lines: 52 Status: RO
Hi Martin-
The prognosis is not good. After about 4 ot 5 years of service IPI and SMD drives start to have problems.
This is now the time to look at a RAID rack or a chain of FAST SCSI disks.
The most inexpensive way to go is get SCSI. Most SCSI disks have a 5 year warranty. This will get you out of the IPI business and into something that will be stable.
Richard Wong Princeton Math
> From sun-managers-request@Princeton.EDU Thu Mar 2 20:54:00 1995 > Sender: sun-managers-relay@ra.mcs.anl.gov > Reply-To: martin@gea.hsr.it (Martin Achilli) > Followup-To: junk > To: sun-managers@eecs.nwu.edu > Subject: IPI disk drives - experience with > Cc: martin@gea.hsr.it > Content-Length: 1172 > X-Lines: 23 > > Hello, > I have a Sun 4/470 with SunOs 4.1.3 32Mb RAM, 4 669Mb SCSI HDs and > 2 911Mb Sun IPI HDs in an a/c room at 20 degrees connected to a UPS. The machine is about 5 years old. I have reformatted the IPI drives at the following intervals, I was getting the "Data Retry Performed" (see below) errors at the time: > - machine new > - after 3 1/2 years > - after 1 year (25 august 1994) > > now, after only 7 months I am getting the "Data Retry Performed" errors again. Can anyone who has specifically had experience with IPI drives tell me the life expectancy of these drives ? Are they about to crash, or should I reformat again and not worry ? > > The drives are called id000 and id001, the errors are: > > Oct 31 13:44:19 zeus vmunix: id001g: block 1057606 (1350262 abs): read: Conditional Success. Data Retry Performed. > Oct 31 13:44:20 zeus vmunix: id001g: block 1057666 (1350322 abs): read: Conditional Success. Data Retry Performed. > ..many many of these/day.. > > I will SUMMARIZE, > thanks Martin > > Martin Achilli - martin@gea.hsr.it > Consiglio Nazionale delle Ricerche - I.N.B. > c/o Medicina Nucleare, Osp. San Raffaele > 20132 Milano, Italy > tel: +39/2/26433648 fax: ../26415202 >
>From lightning.mitre.org!davee Fri Mar 3 17:09:48 1995 Date: Fri, 3 Mar 95 11:06:09 EST From: davee@lightning.mitre.org (David N. Edwards) To: martin@gea.hsr.it Subject: Re: IPI disk drives - experience with Content-Length: 1552 X-Lines: 30 Status: RO
Greetings Martin,
I have a 690-514 that started life (with 4 IPI drives) as a 490 in around 1990 or 1989. The original 4 spindles are the 4.5MB/sec. ~1GB drives. We also have 1 6MB/sec. 911MB drive, and several 1.3GB IPI's (5" form factor, vs. the older 8" drives). So far, we've had one of the original 4 fail completely (say, about 1 year ago). All the others are still running fine! I have noticed a pattern similar to the one you've seen, however. When the drives were new, they recorded almost no "read retries." When they hit about 4 years old, they started logging several. We started a routine of formatting one spindle a month (on our system this means each drive gets formatted about once a year). This used to be a normal part of the routine on our older Vax machines, too. So far, no other drives have failed, but I do see the read retries returning more often. This could well be due to temperature, (I hope the 20 degrees you mentioned is Celcius! :-> ), as about twice per year we have to re-adjust the A/C in the machine room due to widely varying outside temps. This winter has had an unusual number of widely varying days (I think we hit 70F in January, and also -1F !).
At any rate, I've been happy with the IPIs we have. By the way, our read retries usually show up during a "whole disk" operation such as a level-0 dump to tape. I will be grateful to see a copy of your summary, should you get more people responding!
Regards,
Dave Edwards MITRE Corp. J-Lab Asst. Mgr. davee@mitre.org
>From ebu.ericsson.se!ebumfr Fri Mar 3 19:15:30 1995 From: Mike Rembis 66520 <ebumfr@ebu.ericsson.se> Date: Fri, 3 Mar 1995 10:12:23 -0800 To: martin@gea.hsr.it Subject: Re: IPI disk drives - experience with Reply-To: Mike.Rembis@ebu.ericsson.com Content-Length: 536 X-Lines: 15 Status: RO
After about 3 years, IPI drive heads start to drift more frequently. We observered, and later correlated with others on the net, that the frequency of reformatting was every 6 months.
This was enough hassle for us to replace ALL of our IPI drives with SCSI drives.
Since you have a 470 - you have a built in SCSI bus. Use it.
Mike Rembis | Internet: mike.rembis@ebu.ericsson.se Ericsson, Inc. (EUS/BT) | MEMO: ERI.EUS.EUSBMFR ````````````````Reality is in the Eye of the Beholder'''''''''''''''
>From IMIHSRA.HSR.IT!roentgen.RadOnc.Duke.EDU!antoine Fri Mar 3 20:34:45 1995 To: martin@gea.hsr.it (Martin Achilli) Subject: Re: IPI disk drives - experience with Date: Fri, 03 Mar 1995 14:26:39 -0500 From: Phil Antoine <antoine@RadOnc.Duke.EDU> Content-Length: 1785 X-Lines: 39 Status: RO
> Date: Thu, 2 Mar 95 14:04 MET > From: martin@gea.hsr.it (Martin Achilli) > Subject: IPI disk drives - experience with > > Hello, > I have a Sun 4/470 with SunOs 4.1.3 32Mb RAM, 4 669Mb SCSI HDs and > 2 911Mb Sun IPI HDs in an a/c room at 20 degrees connected to a UPS. The machine is about 5 years old. I have reformatted the IPI drives at the following intervals, I was getting the "Data Retry Performed" (see below) errors at the time: > - machine new > - after 3 1/2 years > - after 1 year (25 august 1994) > > now, after only 7 months I am getting the "Data Retry Performed" errors again. Can anyone who has specifically had experience with IPI drives tell me the life expectancy of these drives ? Are they about to crash, or should I reformat again and not worry ? > > The drives are called id000 and id001, the errors are: > > Oct 31 13:44:19 zeus vmunix: id001g: block 1057606 (1350262 abs): read: Conditional Success. Data Retry Performed. > Oct 31 13:44:20 zeus vmunix: id001g: block 1057666 (1350322 abs): read: Conditional Success. Data Retry Performed. > ..many many of these/day.. > > I will SUMMARIZE, > thanks Martin
Those drives were unequivocally, the worst peices of crap we ever bought from Sun. We finally ditched the problem when we upgraded from the 4/470 to a SS1000. Of course, then we got Solaris...
I had exactly the same problems, with the same declining interval between reformats. Sending them back with the promise of replacements that didn't have the flaw was a big joke. You are stuck with the problem until you upgrade (or outgrade). Just reformat and bear it...
Sorry,
Phil Antoine (antoine@RadOnc.Duke.EDU) Duke University Medical Center Radiation Oncology Physics Durham, North Carolina USA
>From uniq.com.au!glenn Fri Mar 3 23:35:10 1995 Date: Sat, 4 Mar 1995 08:51:59 --1000 From: Glenn.Satchell@uniq.com.au (Glenn Satchell - Uniq Professional Services) To: martin@gea.hsr.it Subject: Re: IPI disk drives - experience with X-Sun-Charset: US-ASCII Content-Length: 2582 X-Lines: 55 Status: RO
I have seen this at many sites using IPI drives of about the same age as yours. No doubt you can keep re-formatting them at ever decreasing frequencies, or if you have hardware maintenance Sun should replace the drives with re-conditioned ones (which will probably still need to go through the re-formatting exercise in, say, another 12 months).
Ultimately though I think the drives are nearing the end of their useful life. Replacing them with SCSI disks will certainly be cheaper than buying new IPI disks. You could keep the IPI's around as archive disk space if required.
Given th ehigh cost of maintenance on the VME boxes you might even consider upgrading to a SS5 or SS20, either of which will easily outperform the 4/470.
regards, -- Glenn Satchell glenn@uniq.com.au | There's a fine line Uniq Professional Services Pty Ltd ACN 056 279 335 | between fishing and PO Box 70, Paddington, NSW 2021, (Sydney) Australia | standing on the shore Phone 02 380 6360 Pager 016 287 000 Fax 02 380 6416 | looking like an idiot.
> From sun-managers-request@uniq.com.au Fri Mar 3 17:28 EST 1995 > Date: Thu, 2 Mar 95 14:04 MET > From: martin@gea.hsr.it (Martin Achilli) > Reply-To: martin@gea.hsr.it (Martin Achilli) > To: sun-managers@eecs.nwu.edu > Subject: IPI disk drives - experience with > Cc: martin@gea.hsr.it > X-Lines: 23 > > Hello, > I have a Sun 4/470 with SunOs 4.1.3 32Mb RAM, 4 669Mb SCSI HDs and > 2 911Mb Sun IPI HDs in an a/c room at 20 degrees connected to a UPS. The machine is about 5 years old. I have reformatted the IPI drives at the following intervals, I was getting the "Data Retry Performed" (see below) errors at the time: > - machine new > - after 3 1/2 years > - after 1 year (25 august 1994) > > now, after only 7 months I am getting the "Data Retry Performed" errors again. Can anyone who has specifically had experience with IPI drives tell me the life expectancy of these drives ? Are they about to crash, or should I reformat again and not worry ? > > The drives are called id000 and id001, the errors are: > > Oct 31 13:44:19 zeus vmunix: id001g: block 1057606 (1350262 abs): read: Conditional Success. Data Retry Performed. > Oct 31 13:44:20 zeus vmunix: id001g: block 1057666 (1350322 abs): read: Conditional Success. Data Retry Performed. > ..many many of these/day.. > > I will SUMMARIZE, > thanks Martin > > Martin Achilli - martin@gea.hsr.it > Consiglio Nazionale delle Ricerche - I.N.B. > c/o Medicina Nucleare, Osp. San Raffaele > 20132 Milano, Italy > tel: +39/2/26433648 fax: ../26415202 >
>From sirius.uvic.ca!mmcintos Mon Mar 6 20:05:57 1995 Date: Mon, 6 Mar 95 11:02:39 PST From: Mark.McIntosh@Engr.uvic.ca To: martin@gea.hsr.it (Martin Achilli) Subject: IPI disk drives - experience with Content-Length: 1310 X-Lines: 30 Status: RO
On Thu, 2 Mar 95 19:04:21 GMT, martin@gea.hsr.it (Martin Achilli) said: >The machine is about 5 years old. [...] I was getting the "Data Retry >Performed" (see below) errors at the time [...] now, after only 7 >months I am getting the "Data Retry Performed" errors again. Can >anyone who has specifically had experience with IPI drives tell me >the life expectancy of these drives ? Are they about to crash, or >should I reformat again and not worry ?
We, too, have a Sun 4/470 with 2 IPI drives that is giving these Data Retry errors regularly (on both read and write). The system is 4-5 years old, too. These errors started in Dec/94 - the first time we have seen them in such numbers. We haven't tried reformatting the drives yet.
We also see write errors like these:
Mar 6 10:57:49 active vmunix: id000h: block 993074 (1898354 abs): write: Uncorrectable Data Check.
We have had more frequent crashes with this system since the errors started. But we can't say for sure that the errors caused the crashes.
I'd appreciate a copy of any information you gather.
Thanks.
----- Mark J. McIntosh, <Mark.McIntosh@Engr.UVic.CA>, Programmer/Analyst University of Victoria, Faculty of Engineering - Dean's Office, EOW 248 Box 3055, Victoria, BC, V8W 3P6, CANADA, Voice: 604-721-6049 FAX: 604-721-8676
>From martin Thu Mar 2 14:04:05 1995 Date: Thu, 2 Mar 95 14:04 MET From: martin (Martin Achilli) To: sun-managers@eecs.nwu.edu Subject: IPI disk drives - experience with Cc: martin Content-Length: 1172 Status: RO X-Lines: 23
Hello, I have a Sun 4/470 with SunOs 4.1.3 32Mb RAM, 4 669Mb SCSI HDs and 2 911Mb Sun IPI HDs in an a/c room at 20 degrees connected to a UPS. The machine is about 5 years old. I have reformatted the IPI drives at the following intervals, I was getting the "Data Retry Performed" (see below) errors at the time: - machine new - after 3 1/2 years - after 1 year (25 august 1994)
now, after only 7 months I am getting the "Data Retry Performed" errors again. Can anyone who has specifically had experience with IPI drives tell me the life expectancy of these drives ? Are they about to crash, or should I reformat again and not worry ?
The drives are called id000 and id001, the errors are:
Oct 31 13:44:19 zeus vmunix: id001g: block 1057606 (1350262 abs): read: Conditional Success. Data Retry Performed. Oct 31 13:44:20 zeus vmunix: id001g: block 1057666 (1350322 abs): read: Conditional Success. Data Retry Performed. ..many many of these/day..
I will SUMMARIZE, thanks Martin
Martin Achilli - martin@gea.hsr.it Consiglio Nazionale delle Ricerche - I.N.B. c/o Medicina Nucleare, Osp. San Raffaele 20132 Milano, Italy tel: +39/2/26433648 fax: ../26415202
This archive was generated by hypermail 2.1.2 : Fri Sep 28 2001 - 23:10:18 CDT