brk102@leah.albany.edu (Brian King) (06/14/91)
I have an IDE hard drive, a Toshiba 100MB (can't remember the model number), that seems to pause sometimes, for around 15 seconds, and then continues. When I say pause, I mean that it causes my whole system to completely stop whatever it is doing. The drive light stays on for about 10 seconds, and then goes off for the remainder of the time, until the system recovers. It usually never takes more than 30 secs. to fully recover. When it pauses, it happens in the middle of a disk read/write. It doesn't just happen for no reason. Does anyone have ANY idea what could be going on? It can happen at ANY time, however, it seems to happen more often when I am in Windows. I am guessing that a nice low level format would do the trick, (maybe the interleave is off a little?) but I've heard horror stories about low level formatting an IDE drive. Are there any companies out there that have software for low level formatting an IDE drive? Any tips, pointers, etc? Relevant system information: 386/25 Mhz, 64k cache, 4 megs RAM AMI BIOS - 4/89 Thanks for any information you can give me. Send email. I'll post a summary if interest shows. -Brian King (brk102@leah.albany.edu)
feustel@netcom.COM (David Feustel) (06/14/91)
I had a very similar problem with a Conner drive that I was doing some nonstandardd things with. I came to the conclusion that it was caused either by a programming error by me or by a problem with the auxillary power supply I was using with the drive. I never have had a problem before or since with this drive when I was using it as a standard dos drive connected to the system power supply. -- David Feustel, 1930 Curdes Ave, Fort Wayne, IN 46805, (219) 482-9631 EMAIL: feustel@netcom.com or feustel@cvax.ipfw.indiana.edu
tt@euler.jyu.fi (Tapani Tarvainen) (06/14/91)
In article <1991Jun13.193634.24680@sarah.albany.edu> brk102@leah.albany.edu (Brian King) writes: >I have an IDE hard drive, a Toshiba 100MB (can't remember the model >number), that seems to pause sometimes, for around 15 seconds, and >then continues. Does your machine have AMI bios? It seems that certain versions of AMI 386 bios have a bug that causes them to hang occasionally like that with big disks (I don't know the exact limit: it doesn't appear with 40MB drives, and does with 100MB ones). We've had this problem with several machines; after we complained enough they got as new bios eproms and the problem went away. -- Tapani Tarvainen (tarvaine@jyu.fi, tarvainen@finjyu.bitnet)
hv@uwasa.fi (Harri Valkama) (06/14/91)
In article <TT.91Jun14142123@euler.jyu.fi> tt@euler.jyu.fi (Tapani Tarvainen) writes: >In article <1991Jun13.193634.24680@sarah.albany.edu> brk102@leah.albany.edu (Brian King) writes: > >>I have an IDE hard drive, a Toshiba 100MB (can't remember the model >>number), that seems to pause sometimes, for around 15 seconds, and >>then continues. > >Does your machine have AMI bios? It seems that certain versions of >AMI 386 bios have a bug that causes them to hang occasionally like >that with big disks (I don't know the exact limit: it doesn't appear >with 40MB drives, and does with 100MB ones). We've had this problem >with several machines; after we complained enough they got as new >bios eproms and the problem went away. >-- >Tapani Tarvainen (tarvaine@jyu.fi, tarvainen@finjyu.bitnet) Perhaps you could give us some dates and version numbers out of these AMI bioses you are writing about. I have AMI bios and 135 Mb AT/IDE and I haven't noticed such problems. -- == Harri Valkama, University of Vaasa, Finland ============================ P.O. Box 700, 65101 VAASA, Finland (tel:+358 61 248426 fax:+358 61 248465) Anon ftp garbo.uwasa.fi (128.214.12.37) & nic.funet.fi (128.214.6.100) hv@uwasa.fi hv@finfiles.bitnet /s=hv/o=uwasa/prdm=inet/amdm=fumail/c=fi
feustel@netcom.COM (David Feustel) (06/14/91)
I have two machines with AMI bioses and large IDE drives. This problem has never occurred on either machine. The problem I reported earlier occurred on a 386sx (ALR) with Phoenix bios and only when I attempted to program the disk io myself. -- David Feustel, 1930 Curdes Ave, Fort Wayne, IN 46805, (219) 482-9631 EMAIL: feustel@netcom.com or feustel@cvax.ipfw.indiana.edu
mmshah@athena.mit.edu (Milan M Shah) (06/14/91)
In article <1991Jun13.193634.24680@sarah.albany.edu> brk102@leah.albany.edu (Brian King) writes: >I have an IDE hard drive, a Toshiba 100MB (can't remember the model >number), that seems to pause sometimes, for around 15 seconds, and >then continues. When I say pause, I mean that it causes my whole >system to completely stop whatever it is doing. The drive light stays >on for about 10 seconds, and then goes off for the remainder of the >time, until the system recovers. >... >Relevant system information: > 386/25 Mhz, 64k cache, 4 megs RAM > AMI BIOS - 4/89 > >-Brian King (brk102@leah.albany.edu) > I have the exact same problem with a completely different setup! I have a 20MHz Mylex MB machine with 64K cache, a Conner 3104, and Phoenix BIOS v 1.10.10 (copyrighted 1985-1989). Problem is not software specific: so far, it has happened while I am in Windows (which is most of the time), while I was using MicroEmacs (from plain old DOS), while I was using a shareware DOS shell, and while using FoxBase. I thought the drive was bad, and was almost ready to send it back to the vendor. Anyone know what the problem might be? It seems that my Conner and Brian's Toshiba have a cache on the drive. Could it be that the cache is just being flushed? Thanks! Milan .
<H90SI@CUNYVM.BITNET> (06/15/91)
In article <1991Jun14.144313.3504@athena.mit.edu>, mmshah@athena.mit.edu (Milan M Shah) says: > >In article <1991Jun13.193634.24680@sarah.albany.edu> brk102@leah.albany.edu >(Brian King) writes: >>I have an IDE hard drive, a Toshiba 100MB (can't remember the model >>number), that seems to pause sometimes, for around 15 seconds, and >>then continues. When I say pause, I mean that it causes my whole >>system to completely stop whatever it is doing. The drive light stays >>on for about 10 seconds, and then goes off for the remainder of the >>time, until the system recovers. > >I have the exact same problem with a completely different setup! I have a >20MHz Mylex MB machine with 64K cache, a Conner 3104, and Phoenix BIOS >v 1.10.10 (copyrighted 1985-1989). Problem is not software specific: so far, I have the same problem with my ST157A 40 Mb IDE drive. I have 386sx with DTK BIOS. It happens more often when I switch from 16Mhz to 8Mhz. Usually it pauses during heavy disk activity. I think it has something to do with timing. +----------------------------------+--------------------------------+ | Dmitry Teytelman (718)252-7708 | Design a system that even | | Internet: h90si@cunyvm.cuny.edu | fool can use and only fool | | Bitnet: h90si@cunyvm | will use it. | +----------------------------------+--------------------------------+
sichermn@beach.csulb.edu (Jeff Sicherman) (06/16/91)
In article <91166.100432H90SI@CUNYVM.BITNET> H90SI@CUNYVM.BITNET writes: >In article <1991Jun14.144313.3504@athena.mit.edu>, mmshah@athena.mit.edu (Milan >M Shah) says: >> >>In article <1991Jun13.193634.24680@sarah.albany.edu> brk102@leah.albany.edu >>(Brian King) writes: >>>I have an IDE hard drive, a Toshiba 100MB (can't remember the model >>>number), that seems to pause sometimes, for around 15 seconds, and >>>then continues. When I say pause, I mean that it causes my whole >>>system to completely stop whatever it is doing. The drive light stays >>>on for about 10 seconds, and then goes off for the remainder of the >>>time, until the system recovers. >> >>I have the exact same problem with a completely different setup! I have a >>20MHz Mylex MB machine with 64K cache, a Conner 3104, and Phoenix BIOS >>v 1.10.10 (copyrighted 1985-1989). Problem is not software specific: so far, > >I have the same problem with my ST157A 40 Mb IDE drive. I have 386sx with >DTK BIOS. It happens more often when I switch from 16Mhz to 8Mhz. Usually >it pauses during heavy disk activity. I think it has something to do with >timing. > It sounds like it might be I-O retries due to disk read errors. Of course they could be media failure or weakness which is recoverable after some number of attempts or random errors due to noise or the speed change.
quimby@madoka.its.rpi.edu (Quimby Pipple) (06/16/91)
I've seen this slowdown thing too. We were unable to determine the cause, the tech people we called seemed to think it indicated a bad drive, so we returned it. Since then we've been doing much more intensive disk timing tests on the IDE systems we sell. Quimby -- quimby@mts.rpi.edu, quimby@rpitsmts.bitnet
tt@euler.jyu.fi (Tapani Tarvainen) (06/17/91)
In article <1991Jun14.124305.8021@uwasa.fi> hv@uwasa.fi (Harri Valkama) writes: >In article <TT.91Jun14142123@euler.jyu.fi> tt@euler.jyu.fi (Tapani Tarvainen) writes: >>In article <1991Jun13.193634.24680@sarah.albany.edu> brk102@leah.albany.edu (Brian King) writes: >> >>>I have an IDE hard drive, a Toshiba 100MB (can't remember the model >>>number), that seems to pause sometimes, for around 15 seconds, and >>>then continues. >>It seems that certain versions of >>AMI 386 bios have a bug that causes them to hang occasionally like >>that with big disks (I don't know the exact limit: it doesn't appear >>with 40MB drives, and does with 100MB ones). We've had this problem >>with several machines; after we complained enough they got as new >>bios eproms and the problem went away. >Perhaps you could give us some dates and version numbers out of these >AMI bioses you are writing about. I have AMI bios and 135 Mb AT/IDE >and I haven't noticed such problems. The problematic bios date was 4/30/89. Actually it appears it may not be the size of the disk (or at least not only that); the speed of the disk may be more important: it does look like it could be a timing problem. Anyway, our experience was that machines with the AMI bios had the problem, when it was changed (in one case to Phoenix, in another to newer AMI) it went away. -- Tapani Tarvainen (tarvaine@jyu.fi, tarvainen@finjyu.bitnet)
hv@uwasa.fi (Harri Valkama) (06/18/91)
In article <TT.91Jun17141709@euler.jyu.fi> tt@euler.jyu.fi (Tapani Tarvainen) writes: >In article <1991Jun14.124305.8021@uwasa.fi> hv@uwasa.fi (Harri Valkama) writes: >>Perhaps you could give us some dates and version numbers out of these >>AMI bioses you are writing about. I have AMI bios and 135 Mb AT/IDE >>and I haven't noticed such problems. > >The problematic bios date was 4/30/89. Actually it appears it may >not be the size of the disk (or at least not only that); the speed >of the disk may be more important: it does look like it could be a >timing problem. Anyway, our experience was that machines with the >AMI bios had the problem, when it was changed (in one case to Phoenix, >in another to newer AMI) it went away. >-- >Tapani Tarvainen (tarvaine@jyu.fi, tarvainen@finjyu.bitnet) Yes, that is what I wanted to know. Newer AMI is alright. Phew. -- == Harri Valkama, University of Vaasa, Finland ============================ P.O. Box 700, 65101 VAASA, Finland (tel:+358 61 248426 fax:+358 61 248465) Anon ftp garbo.uwasa.fi (128.214.12.37) & nic.funet.fi (128.214.6.100) hv@uwasa.fi hv@finfiles.bitnet /s=hv/o=uwasa/prdm=inet/amdm=fumail/c=fi