elliott@b11.ingr.com (Jeff Elliott) (10/17/89)
Subject: problem with WD2010 and Miniscribe 3085 Newsgroups: unix-pc.general,comp.sys.att Summary: surface test errors galore! References: <22964@cup.portal.com> <2221@ethz.UUCP> <23093@cup.portal.com> In article <23093@cup.portal.com>, thad@cup.portal.com (Thad P Floryan) writes: > ...................................................... As I mentioned, I've > been "away" for awhile and upon return discovered email indicating that seven > (7) of the WD2010 chips are bad, and everyone has reported the exact same > symptoms. I'll be sending private email to those people and be working on > getting replacement chips; I'm curious as to the date code on the bad chips. I *finally* got around to installing my MS3085 and the WD2010 this weekend. The WD2010 seemed to work fine with the stock 20MB drive I was using, so I then put in the 3085. Formatting with the new s4diags was no problem (except for the missing loader on the diag floppy :-) but the surface test started reporting write data errors for 60-70% of all sectors. I stopped the test after about 10 cylinders, removed the sectors from the bad block table (by the way, is there a *easy* way to do this???), powered down and reseated the WD2010 and reconnected the 3085 to make sure I had everything right, and tried it again; same results. So I'm now using the 3085 with the WD1010 with no problems (bad block table now has only 12 entries after formatting with the WD1010 and the original diag floppy). So, is this a 1) 3085 problem (don't think so, because it works with the WD1010), 2) WD2010 problem (don't think so, because it works with the 20MB drive), or 3) a 3085/WD2010 interfacing problem (I think I remember Jan Bagend discussing this at one time...are these the symptoms you saw?). BTW, my system is a P3..P5 7300 with 1/2MB and an 1/2MB expansion memory board. Thanks in advance for any input, -- ------------------------------------------ Jeff Elliott b11!elvis!elliott@ingr.com uunet!ingr!b11!elvis!elliott
dold@mitisft.Convergent.COM (Clarence Dold) (10/20/89)
in article <6216@b11.ingr.com>, elliott@b11.ingr.com (Jeff Elliott) says: > reporting write data errors for 60-70% of all sectors. I stopped the test > after about 10 cylinders, removed the sectors from the bad block table (by the > way, is there a *easy* way to do this???), powered down and reseated the Try: iv -d /dev/rfp000 >d0 vi d0, changing "badblocktable 1" to "badblocktable 1 empty" and deleting all of the unwanted bad spots. Then iv -uv /dev/rfp000 d0 should do it for you. This assumes that there was no data that you wanted mapped to any of those bad blocks, either eliminated or remaining. They're all gone. -- --- Clarence A Dold - dold@tsmiti.Convergent.COM (408) 434-5293 ...pyramid!ctnews!tsmiti!dold P.O.Box 6685, San Jose, CA 95150-6685 MS#10-007
thad@cup.portal.com (Thad P Floryan) (10/21/89)
Jeff Elliott comments about problems with a WD2010 and a Miniscribe 3085. Sigh. The combination DOES work perfectly WHEN ALL PARTS ARE OK. In Jeff's case, I suspect the WD2010. I've heard from another person (in email that I've just started to be able to look at today) that there appears to be a higher than expected failure rate of the chips. I'm working on getting replacements for those who participated in the WD2010 group buy of 6 weeks ago. And I will personally test each and every chip on one of my systems with a Miniscribe 3085. I'm sorry for the delays; circumstances beyond my control have prevented me correcting this problem earlier. Thad Floryan [ thad@cup.portal.com (OR) ..!sun!portal!cup.portal.com!thad ]