ajmyrvold@violet.waterloo.edu (Alan Myrvold) (02/16/88)
Mmmmm ... I'm having a tough time with my Seagate ST 238 hard
drive. I hope my situation will ring some bells with someone
who has experienced the same problem.
The disk seems to have trouble seeking to sectors ... It always
reads and writes my data correctly, but sometimes it makes these
weird noises while doing so :
BEEEDEEEDEEEP click BEEEDEEEDEEP click BEEEDEEEDEEEP click etc.
or
Whiiiiirrrrrrrrrhhhhhhhhhhhhhh. BEEEDEEEDEEP.
The end result is that it takes 5 seconds or so to seek a single
sector!!!
Now here is what really confuses me, It doesn't do these things
consistently, nor is it getting any worse. Some days it can't
read or write a single sector properly ... other days it works
like a charm. No sectors seem to be any worse than others -
It does this when I'm using all sorts of software. Sometimes it happens
right when I turn the machine on, sometimes after an hour or so.
I've figured out the Whirring noise, I can duplicate it by doing
a BIOS call to "recalibrate" the drive.
What I've tried to eliminate the problem (all unsuccessful so far)
1) Changing controller cards
2) Low level FORMATting
3) DOS FDISKing
4) DOS FORMATting
5) Grounding the machine to prevent static buildup
6) Different temperatures
7) Different humidities
8) Three different versions of DOS (All 3.x)
9) Unplugging assorted peripherals
I've also tried all the diagnostic programs I could get my hands
on. Even when the disk is misbehaving, none report error messages,
the only noticeable difference is in reported access times.
What I haven't tried (because of $$$)
1) Changing the actual hard disk
I've managed to write a diagnostic program of my own, which generates
random sector numbers to seek to, does a DOS ABSREAD (int 25h)
checks for a non-zero return code - immediately does a BIOS disk
service 1 (int 13h) get status of last disk operation AND whammo
even when the disk is misbehaving - no error messages.
So I modified it to time how long the ABSREAD operation took -
and print out a little error message when it is more than 300
milliseconds. It shows me when I have a problem - but still
no pattern to the sectors that trigger the message. The longest
it ever seems to take is about 7 seconds - but seven seconds is
a loooong time for me to wait for it to seek a single sector.
I also have a version of the program which does the sectors
sequentially rather than randomly - it shows far less problems.
Anyone got ANY ideas???
Please EMAIL replies to ajmyrvold@violet.waterloo.edu
Thanx.
---------------------------------------------------------------
Something short and pithy belongs here.
---------------------------------------------------------------aja@i.cc.purdue.edu (Miek Rowan) (02/17/88)
In article <5170@watdragon.waterloo.edu>, ajmyrvold@violet.waterloo.edu (Alan Myrvold) writes: > > weird noises while doing so : > BEEEDEEEDEEEP click BEEEDEEEDEEP click BEEEDEEEDEEEP click etc. > or > Whiiiiirrrrrrrrrhhhhhhhhhhhhhh. BEEEDEEEDEEP. > The end result is that it takes 5 seconds or so to seek a single Hmmm, could you tape this and send it to me? Id really like to hear these particualer sounds :-) miek
kluksdah@hpesoc1.HP.COM (Keith Kluksdahl) (02/18/88)
/ hpesoc1:comp.sys.ibm.pc / ajmyrvold@violet.waterloo.edu (Alan Myrvold) / 9:19 pm Feb 15, 1988 / > Mmmmm ... I'm having a tough time with my Seagate ST 238 hard > drive. I hope my situation will ring some bells with someone > who has experienced the same problem. > Anyone got ANY ideas??? > Please EMAIL replies to ajmyrvold@violet.waterloo.edu I am having the same sort of problems with my ST 238 drive. How about posting any responses here. If my memory doesn't fail me (unlike my 238), there was a discussion of Seagate disk problems a few months ago.
sp202-ad@garnet.berkeley.edu (Celso Alvarez) (02/19/88)
In article <2624@i.cc.purdue.edu> aja@i.cc.purdue.edu (Miek Rowan) writes: >In article <5170@watdragon.waterloo.edu>, ajmyrvold@violet.waterloo.edu (Alan Myrvold) writes: >> >> [my hard disk makes] weird noises: >> BEEEDEEEDEEEP click BEEEDEEEDEEP click BEEEDEEEDEEEP click etc. >> or >> Whiiiiirrrrrrrrrhhhhhhhhhhhhhh. BEEEDEEEDEEP. Really? You better watch out! -- those are the *wrong* noises. MY Seagate hard disk makes BEEEPEEEDEEPEEBEEP click click. BEEEPEEEDEEPEEBEEEP..... cleek. WHIIIIIIIIIIRRRLLLHH WHIIIIIIIIIIRRLLLHH, click click. Tic. AHH-FFFFFFFFFFFFFF AHH-FFFFFFFFFFFFFF....... but only when I'm using it. I'm plain lucky to have got such varied noises for the same price. Specially the "AHH-FFFFFFFF". C. Alvarez (sp202-ad@garnet.berkeley.edu.UUCP)
timothym@tekigm2.TEK.COM (Timothy D Margeson) (02/23/88)
In article <5170@watdragon.waterloo.edu> ajmyrvold@violet.waterloo.edu (Alan Myrvold) writes: >Mmmmm ... I'm having a tough time with my Seagate ST 238 hard drive >........ but sometimes it makes these weird noises while doing so : > BEEEDEEEDEEEP click BEEEDEEEDEEP click BEEEDEEEDEEEP click etc. >or Whiiiiirrrrrrrrrhhhhhhhhhhhhhh. BEEEDEEEDEEP. >The end result is that it takes 5 seconds or so to seek a single sector!!! > Hi, This is symptomatic of a hard disk recalibrate in process. So, are you running a background disk parking program (one that parks the disk every five minutes or so)? If you are, what happens is the disk is being sent out past track 615, and any time this happens, the drive itself performs a reset. All of the whirring is the drive slowwwwly seeking track 0 to regain control of itself. I had an ST225 and had to wait for the drive alot myself. I was using a program called TIMEPARK.COM. Although it was a pain when I wasn't using the disk alot, it was comforting to know TIMEPARK was doing it's job. Moral #1: Know the effects of all the software you are using. You can avoid many headaches this way. Another source of the problem could be that you have told your controller the drive has more cylinders than it really has (>615). If so, then and bad sectors could have been remapped to the last cylinders (thus causing the reset when seeking the next cylinder). To avoid this conflict, BE SURE the controller knows the true limits of the drive. That is, 615 usable cylinders, with > 650 total cylinders. These extra tracks are for head parking purposes only. If a seek is completed to one of these tracks, I believe you can do reads and writes but, the next seek will force a drive recalibrate to be performed. Moral #2: Know the limits of the hardware you have. Don't try to get more than you paid for. Use what the manuals say, no more. Good luck..... -- Tim Margeson (206)253-5240 PO Box 3500 d/s C1-937 @@ 'Who said that?' Vancouver, WA. 98668 {amd..hplabs}cae780!tektronix!tekigm2!timothym (this changes daily)