browne@savax.UUCP (01/30/87)
I've been working on installing a set of the Fujitsu Super Eagle disks, with an Emulex controller, on a 780 (running VMS). They ran fine for a while, then one started having timing errors on track 1, and hard errors on track 7. We copied all the data from that disk to a spare, and now the entire platter #6 on the spare has gone out. We tried to backup that disk to another spare (these are all the super eagles). There was much error messaging and gnashing of teeth, and when all was said and done, we had 0 free blocks, where the old disk had over 65,000 free. So, what I'm wondering, is that since 2 out of 5 super eagles shipped to us, as of November, 1986 have gone bad, do I have more to look forward to? I've seen the articles others have submitted here, about their problems - are there still people out there having trouble with dying Super eagles? The folks out there that have had them replaced, are the replacements still running OK out there, or have you had more problems? We'd like to know, so we can figure out whether we should try to exchange these for the regular eagles, and just say to hell with the larger capacity. You can mail to me at ...decvax!savax!browne, or call me at (603) 885-1082. I'll call you back, if you're worried about the toll charges. Thanks much! Duff Browne
sherouse@godot.UUCP (02/04/87)
As previously reported in this forum our two SI9761 drives both blew out - one at 4 months and the other at ~9 months. For both the failure mode was as has been widely reported - a practically geometric increase in bad block count over a few days, appearing to afflict only recently written blocks. The fix was to replace the HDAs. We subscribed to the myth that we were seeing a fault in some early rev of the drives. This weekend the myth and one of our *replacement* HDAs failed. This HDA was replaced less than a month ago. Our replacement is en route from California. This makes three in less than a year. For what it's worth the first HDA was replaced about 6 months ago and has given no trouble since. It was one of the first that SI experienced this failure problem with. I offer these facts without editorial comment. When the big eagles work, we love them - especially in combination with SI's cache controller. Stay tuned for more thrilling updates. - George <decvax!mcnc!godot!sherouse>
dem@uwslh.UUCP (David E. Miran) (02/04/87)
*** REPLACE THIS LINE WITH YOUR MESSAGE *** Recently browne@savax.UUCP (Duff Browne) wrote >What I'm wondering, is that since 2 out of 5 super eagles shipped to us, as >of November, 1986 have gone bad, do I have more to look forward to? >We'd like to know, so we can figure out whether we should try to exchange >these for the regular eagles, and just say to hell with the larger capacity. We just replaced a super eagle that died (in the past week). This drive was acquired last summer (august 1986). The word that I was given by our Systems Industries repair person is: 1. Fujitsu is aware of the problem and working on it. 2. It is not fixed, so replacements will probably fail in the same way. 3. There is no word available on when the problems will be solved. If I had it to do over I would go with eagles, rather than super eagles. P.S. Many thanks to the net and especially Don Seeley who posted the summary of 2361 problems. This prompted me to quickly rearrange our system so that the root and /usr file system were on our eagle and not the super eagle where they had been. This made it much easier to recover when the supereagle died. I got everything switched about 1 week before the supereagle started dying. -- David E. Miran ...!{seismo,harvard,topaz,ihnp4}!uwvax!uwslh!dem Wisconsin State Hygiene Lab or uwslh!dem@rsch.wisc.edu University of Wisconsin (608) 262-0019 465 Henry Mall Madison, WI 53706
bjb@mitisft.UUCP (02/05/87)
We had similar problems with our Eagle-XP drives. After the first two both failed, I put them on an UPS (with power conditioner). I haven't had a problem with them in the last 9 months. Bruce Beare
lacasse@randvax.UUCP (02/06/87)
In article <493@savax.UUCP> browne@savax.UUCP (Duff Browne) writes: >I've been working on installing a set of the Fujitsu Super Eagle disks... >I'm wondering, is that since 2 out of 5 super eagles shipped to us, as >of November, 1986 have gone bad, do I have more to look forward to? >...The folks out there that have had them replaced, are the replacements still >running OK out there, or have you had more problems? I installed Rand's first super eagle last August. It died. Fujitsu repaired it (which took 3 months). The new one has worked fine ever since. Your mileage may vary. Hint: it is supposed to be better to get an HDA sub-assembly with revision of D4 or higher. Mark LaCasse qantel!hplabs!sdcrdcf!randvax!lacasse c/o The Rand Corporation cbosgd!ihnp4!sdcrdcf!randvax!lacasse 1700 Main Street decvax!randvax!lacasse Santa Monica, CA 90406 213/393-0411 ext. 7420 lacasse@Rand-Unix
mra@fathom.UUCP (02/16/87)
In article <735@godot.UUCP>, sherouse@godot.UUCP (George W. Sherouse) writes: > > As previously reported in this forum our two SI9761 drives both blew > out - one at 4 months and the other at ~9 months. For both the > failure mode was as has been widely reported - a practically geometric > increase in bad block count over a few days, appearing to afflict only > recently written blocks. The fix was to replace the HDAs. We > subscribed to the myth that we were seeing a fault in some early rev > of the drives. > > This weekend the myth and one of our *replacement* HDAs failed. This > HDA was replaced less than a month ago. Our replacement is en route > from California. This makes three in less than a year. > Both SI and Emulex recently submitted quotes on a 4-drive system for two microVAX's using Super-Eagles. Two weeks ago, both companies pulled their quotes and removed the Super Eagles from the quotes. SI hinted that it would no longer sell Super Eagles in the near future; Emulex quoted a high mucky-muck memo that told sales reps not to offer Super Eagles. SI suggested using their new 9744, an 8inch drive with Eagle capacity (also made by Fujitsu) and higher transfer rates; Emulex suggested staying with the Eagles...
berger@datacube.UUCP (02/18/87)
This is very scary. We recently purchased a Super Eagle and a
Xylogics controler to connect to our Suns. It basically
works fine. We have had some crashs but we have presumed
that it was because we had been using a 3/75 as the host for a while
and there were some mechanical problems with the 3/75 and the
Xylogics/Multibus adapter.
Since we have put it back into our 3/160 things seem fine.
The only wierd thing is that the Super Eagle beeps every once in a while.
Is this normal?
What should we look out for in terms of the Super Eagle disease?
Bob Berger
Datacube Inc. Systems / Software Group 4 Dearborn Rd. Peabody, Ma 01960
VOICE: 617-535-6644; FAX: (617) 535-5643; TWX: (710) 347-0125
UUCP: ihnp4!datacube!berger
{seismo,cbosgd,cuae2,mit-eddie}!mirror!datacube!bergermangler@cit-vax.UUCP (02/22/87)
In article <268@fathom.UUCP>, mra@fathom.UUCP (Mark R. Abbott) writes: > SI suggested using their new 9744, an 8inch drive with Eagle capacity I assume this is the Fujitsu 2344. I called Fujitsu, and they said the drive is still in beta test, they couldn't even send me a spec sheet. Don Speck speck@vlsi.caltech.edu {seismo,rutgers,ames}!cit-vax!speck
mra@fathom.UUCP (03/03/87)
In article <1849@cit-vax.Caltech.Edu>, mangler@cit-vax.Caltech.Edu (System Mangler) writes: > In article <268@fathom.UUCP>, mra@fathom.UUCP (Mark R. Abbott) writes: > > SI suggested using their new 9744, an 8inch drive with Eagle capacity > > I assume this is the Fujitsu 2344. I called Fujitsu, and they said the > drive is still in beta test, they couldn't even send me a spec sheet. > > Don Speck speck@vlsi.caltech.edu {seismo,rutgers,ames}!cit-vax!speck Yes, these are the Fujitsu 2344's. SI seemed to be willing to pass out info. Capacity is the same as the 2351 (Eagle) but seek time is faster (altho latency is worse - slower motor). They are still in beta test, but SI plans to start shipping in April time frame. I don't know how to calibrate this but Emulex rep was very skeptical. His claim is that Super Eagles now are having defective media problems (original problem was HDA contaminating heads he claimed). Since the Super Eagle and the 2344 both use thin film technology, he thinks the same problem will crop up in 2344. Could be a bit of sour grapes towards SI.
lrj@batcomputer.tn.cornell.edu (Lewis R. Jansen) (03/06/87)
In article <269@fathom.UUCP> mra@fathom.UUCP (Mark R. Abbott) writes: >[... long discussion of bad Super Eagles ...] >heads he claimed). Since the Super Eagle and the 2344 both use thin film >technology, he thinks the same problem will crop up in 2344. Could be >a bit of sour grapes towards SI. The story i hear from Secure Data (another marketting company) is thus: The problems with the Super Eagles is due to the head interlock not doing its job well enough. Seems that Fuji uses a magnetic lock on the SEs, which wasn't strong enough to hold in particularly bad shipping conditions. The heads would pop off the landing zone, do a dance on the platters, and pop back. The symptoms were an increasing number of bad blocks showing up everywhere. When a new HDA is shipped to replace the bad one, it gets an even worse treatment as it's all alone in its box; none of the mountings to take any of the shock. So the problem is suddenly on the new HDA too. In any case, this was supposedly only the problem in something like 2% of the disks. Has anyone here had problems with Super Eagles that were NOT purchased from SI? Fuji has supposedly fixed this problem, and it shouldn't happen again. In addition, the 2344s are supposed to be physically almost identical to the 2333s, except there are 27 R/W heads, and fewer cylinders. Please note that the above information was obtained from a source who certainly has (a) an interest in selling us disks, and (b) an interest in NOT selling us bum disks, as we're buying an extra four years worth of service contract in the purchase price (we're getting 1 year free). I also have no connection with Secure Data except as a happy customer (we bought a pair of 2333s a little while back). -- -- Lewis R. Jansen, LASSP Systems Grunt lrj@lasspvax.tn.cornell.edu ...!cornell!lasspvax!lrj The above opinions are for sale or rent. Inquire within.
berger@datacube.UUCP (03/08/87)
We also have a Super Eagle that has the disease. We purchased it thru
a Fuji distributer. We are having it repaired by Fuji and are "renting"
a 2351 for a flat fee of $1k from Fuji until our Super Eagle is back and
flying.
Bob Berger
Datacube Inc. Systems / Software Group 4 Dearborn Rd. Peabody, Ma 01960
VOICE: 617-535-6644; FAX: (617) 535-5643; TWX: (710) 347-0125
UUCP: ihnp4!datacube!berger
{seismo,cbosgd,cuae2,mit-eddie}!mirror!datacube!bergersherouse@godot.UUCP (George W. Sherouse) (03/10/87)
In article <357@batcomputer.tn.cornell.edu> lrj@batcomputer.UUCP (Lewis R. Jansen) writes: > ... In any case, this was supposedly only the problem > in something like 2% of the disks. ... Hmm... In that case we must not be livin' right. We have just replaced our fourth HDA (on two drives) in 12 months. Our experience suggests that 1) they are certain to fail and 2) if they live 3 weeks they'll probably die at six months. As far as I can tell the failure rate is not 2% but 200%. Sigh... - George -- - George
carr@convex.UUCP (03/10/87)
>Yes, these are the Fujitsu 2344's. SI seemed to be willing to pass out >info. Capacity is the same as the 2351 (Eagle) but seek time is faster Wrong. The Fuji 2344 has an unformatted capacity of 690.10 Mb. The 2351's is around 474 Mb. >heads he claimed). Since the Super Eagle and the 2344 both use thin film >technology, he thinks the same problem will crop up in 2344. Could be Wrong again. Both drives use good old oxide media. Rob Carruthers {ihnp4!convex!carr}
loverso@sunybcs.UUCP (03/14/87)
In previous articles: > > > SI suggested using their new 9744, an 8inch drive with Eagle capacity > > > > I assume this is the Fujitsu 2344. I called Fujitsu, and they said the > > drive is still in beta test, they couldn't even send me a spec sheet. > > Yes, these are the Fujitsu 2344's. SI seemed to be willing to pass out > info. Capacity is the same as the 2351 (Eagle) but seek time is faster > (altho latency is worse - slower motor). They are still in beta test, but > SI plans to start shipping in April time frame. I just got off the phone with a lovely lass named Lisa at Fujitsu, Limited, in Calif at 408-434-6777, and this is what she told me... We currently have three 2361s - one of them died and was sent back to my vendor (Cameron Computers) 2 months ago. By the time it got to them, and then got shipped back to Fujitsu, it was Feb 6. The vendor was telling me that we should get back our drive "very soon now - 2 weeks". Well, Lisa just told me that Fujitsu has lengthened the turn-around time for all SuperEagle repairs and that it currently takes 90 days from their receipt! Of my other two - one arrived DOA in Oct ("fault" light on continuously) and was returned for repair - two months later it came back and arrived the same day as my third drive. Both of these drives are now beginning to pick up hard errors in the same manner as my first drive. One of these is involved in a crucial data collection project and is on maintenance contact. Cameron tells me that if that drive goes, they can swap in a new one within 24 hours. If my other drive fails, however, it will have to go in on a warranty repair (i.e., 90 days!). I described this to Lisa, and she told me of a policy that Fujitsu has established over the last month or so. For each 2361 in on repair for HDA problems, Fujitsu will loan out a good ol', reliable Eagle to the vendor for a "nominal charge". The vendor then loans this drive to the customer cost-free. When the 2361 is returned, the customer returns the Eagle, and the vendor is credited the nominal charge. This at least helps to maintain some of my faith in Fujitsu! A few other points that Lisa mentioned to me: the current "good" HDA revision is "D0". But, *all* my drives say that. Fujitsu is currently waiting on a shipment of HDAs from Japan. And finally, when I said that my next purchase wasn't going to be a SuperEagle (I'm currently buying two of the 2.5Mb/s, 337Mb 2333s) she said that Fujitsu has just released the "Swallow 4", the much talked about 2344 (see above). It holds as much as the 2361 (689Mb), is the same size (8" platter) and speed as the 2333. But, she said (contrary to previous reports) they _ARE_ currently available. ..jOhn -- John Robert LoVerso @ SUNY/Buffalo Computer Science (716-636-3190) csnet: loverso@buffalo.CSNET uucp: ..!{nike,allegra,decvax,watmath}!sunybcs!loverso bitnet: loverso@sunybcs.BITNET