🚀 go-pugleaf

In message <3m27t3$vpe@styx.uwa.edu.au> - dcheong@student.ecel.uwa.edu.au (Deni
s Cheong - undergrad) writes:
>Joe Gilvary (jgilvary@info.census.gov) wrote:
>
>> Run chkdsk /f:2 (or /f:3, for really messed up volumes) over and over 
>> until it reports 0 errors. Works for me. I don't get any error messages 
>> until I do something really intelligent again (like kicking the power 
>> switch on the UPS <g>)
>
>Like everybody else, when I used to have my Maxtor SCSI 330MB drive as 
>my root OS/2 HPFS partition, I used to get allocation errors for 
>*MANY* files.  Whenever OS/2 died, and it rebooted and ran chkdsk, it 
>would take half an hour, and it it would find these allocation errors, 
>supposedly fix them, and then continue to boot ... next time it 
>died, it would find the same allocation errors (and more) ... there 
>just started being more and more!
>
>Other than the errors at CHKDSK, I had no other problems - I could read 
>the files, write to the disk and never actually lost a file because of 
>the allocation errors.
>
>When I finally bought a new drive, I found several files that I couldn't 
>copy off the drive, but other than that it was fine.
>
>I do have a feeling that the errors may have been caused by physical 
>drive errors in portions of the directory areas ... I had no way of 
>formatting the drive and actually entering the bad sector list.
>
>When I installed, the drive was also formatted with the quick HPFS format 
>- that's when it takes 5 seconds to format the drive.  This is BAD!  When 
>the drive is formatted this way, you *cannot* do a chkdsk /f:3 on the 
>drive!!  If you want my advice, format the drive yourself at the OS/2 
>prompt (format d: /fs:hpfs /l) with the long format process, and you'll 
>be able to fdisk /f:3 ...
>

I was interested to read this. We have identical problems with our server. 
When it crashs, which is does rather more than we would like, it ends up 
doing a check disk at boot up. We have 2 IDE 1/2 gig. drives and 1 1 gig. 
SCSI drive. They all report allocation errors (except C:), which will never 
go away until you delete the files in question. Level 3 check disks will 
clear some of them but never all. I think that many of our Lan Server crashes 
are due to file system problems. We have Lan Server 3 Entry at refresh level 
7045 which is meant to be the latest. The server currently runs OS/2 2.11. We 
are going to try and put the Warp full pack onto it.

Has anybody told IBM about this?

Richard

------------------------------------------
Richard Giles (Data Systems Dept.)
Fisons Instruments Surface Systems
Tel. (+044) 342 327211  Fax. (+044) 342 315074
EMail: rgiles@fisonssurf.co.uk
Compuserve: 100065,132
Web page: http://www.fisonssurf.co.uk