mark@isi.UUCP (Mark Bailey) (06/21/89)
I have been having intermittant floppy problems on my Amiga 1000 (Commodore external drive, 2 Meg Starboard II) for a couple of years now. The problems occur on both drives. On nearly full floppies with lots of small (ProWrite) files I get an unbelievable number of corrupted files. Diskdoctor usually ends up finding "hard errors" on tracks near 40. This happens on various media, including Fuji and Sony DS/DD. I just ran the Ami Alignment S/W on the machine and found no problems. The A1000 has only had the case removed for 8520 replacement (it didn't help)! Any advice/info would be appreciated. -----Patient: Doctor, it hurts when I do that! -----Doctor: Don't do that!!!!! -- Mark Bailey (I didn't really say this.) via: ...!uunet!pyrdc!isi!mark ------Have a 8-| day!!!!!
phils@tekigm2.MEN.TEK.COM (Philip E Staub) (06/23/89)
In article <418@isi.UUCP> mark@isi.UUCP (Mark Bailey) writes: [ describes drive flakiness ] >unbelievable number of corrupted files. Diskdoctor usually ends up finding >"hard errors" on tracks near 40. This happens on various media, including >Fuji and Sony DS/DD. > >I just ran the Ami Alignment S/W on the machine and found no problems. This sounds like the old Delay(0) problem to me. By any chance are you running an old copy of the Dillon/Drew shell (about version 2.04 or earlier)? Or any other old software? If so, get rid of it and see if your problems go away. I haven't personally experienced this in a long time, but the symptoms are more or less just as you describe: semi-random trashing of disks by making track 40 unreadable. I don't remember all of the details, so maybe one of the other old timers can relate the "official" explanation. I was under the impression that the problem was fixed in 1.3, but maybe people just automatically work around it these days. Basically, the "original" fix was to avoid, at *all* costs, calling Delay() with an argument of 0. (The old "If-it-hurts-to-hit-your-thumb-with-a-hammer-don't-hit-your-thumb-with-a-hammer" approach.) Of course, another possible explanation is that with a lot of files on your disk, you're probably updating directories a lot. They therefore get re-written more often than most other things, so they are more likely to be trashed. Since track 40 is the site of the root directory block, all it takes is for a crash to occur during a directory operation and the root directory (and the entire track it's written on, thanks to full-track reads/writes) can be trashed. >Mark Bailey (I didn't really say this.) >via: ...!uunet!pyrdc!isi!mark ------Have a 8-| day!!!!! Hope this helps. Phil -- ------------------------------------------------------------------------------ Phil Staub Tektronix, Inc., Vancouver, Washington 98668 phils@tekigm2.MEN.TEK.COM
usenet@cps3xx.UUCP (Usenet file owner) (06/23/89)
In article <418@isi.UUCP> mark@isi.UUCP (Mark Bailey) writes: >I have been having intermittant floppy problems on my Amiga 1000 (Commodore >external drive, 2 Meg Starboard II) for a couple of years now. The problems >occur on both drives. I was having problems like that for a while. Then I finally discovered that disks formatted/written to in one drive would have more troubles in the other drive. After a couple of month of this, my external drive would nearly always destroy any disk it wrote to. My friendly repair man said it wasn't worth fixing. I hope you have better luck than I did. REAL NAME: Joe Porkka jap@frith.cl.msu.edu
thad@cup.portal.com (Thad P Floryan) (06/23/89)
Mary Bailey laments the floppy errors with his A1000 ... I used to have the same, identical problem. Tracked it down to bad AC power. For some reason, refrigerator(s) kicking in, turning on/off flourescent lamps or modems, etc. anywheres in the building would cause excessive corruption of floppies. I solved the proper with a proper surge protector; not one floppy error in over 3 years now. More specifically: I added "SAFE" model 500 and 1200A SPS systems to each Amiga (and UNIX) station and now I can also ride-through up to a 1 hour power outage also. These units, besides regenerating clean AC power, also protect againt over-/under-voltage, and have transient, surge and hash suppression/protection. If you don't care about power backup, at LEAST get a proper surge protector. I'm going to harp on the word "PROPER" forever. Most of the products on the market are total junk (including the crapola from Radio Shack.) If you need convincing, I'd be happy to enclose a digitized side-by-side photo of, for example, an opened Radio Shack and an opened GTE unit. You'd then quickly ask: "Why hasn't the FTC come down on Radio Shack for consumer fraud?" No kidding. SOme surge protectors actually do more HARM, because their poor design will AMPLIFY some spikes and transients. No kidding. At least you won't see more than 4,000 volts going into your Amiga, because that's the flashover point of the standard AC power plugs used in this country. And before everyone gets all uptight about this, I'm not specifically picking on Radio Shack, but just citing their product as one of many that is simply NOT worth buying. Thad Floryan [ thad@cup.portal.com (OR) ..!sun!portal!cup.portal.com!thad ]
thad@cup.portal.com (Thad P Floryan) (06/23/89)
OOPS! My message should have started: "Mark Bailey ..." and not "Mary Bailey ..." One of the "hazards" of reading/writing email while talking on the phone. :-) Thad Floryan [ thad@cup.portal.com (OR) ..!sun!portal!cup.portal.com!thad ]
mark@isi.UUCP (Mark Bailey) (06/23/89)
In article <5054@tekigm2.MEN.TEK.COM>, phils@tekigm2.MEN.TEK.COM (Philip E Staub) writes: > In article <418@isi.UUCP> mark@isi.UUCP (Mark Bailey) writes: > [ describes drive flakiness ] > >unbelievable number of corrupted files. Diskdoctor usually ends up finding > >"hard errors" on tracks near 40. > > This sounds like the old Delay(0) problem to me. > > > Phil Staub > Tektronix, Inc., Vancouver, Washington 98668 > phils@tekigm2.MEN.TEK.COM Could someone please (e-mail I guess) send me information on this Delay(0) problem. I am using WBench 1.2 and an (old?) version of ConMan. If upgrading to 1.3 will fix it, I'll do it. BTW, my WBench 1.2 disk (98% full) died last night (after ~2 years). Hard error, track 40! Thanks in advance for any help!! -- Mark Bailey (I didn't really say this.) via: ...!uunet!pyrdc!isi!mark ------Have a 8-| day!!!!!