1.OSR5.0.7 LSIL BTLD problem Reply-To: email@example.com
| Has anyone EVER gotten lsil to work?
| I had the same problem a year and a half ago with an IBM X-Series
| server with an on-board LSI controller that required "lsil". We ended
| up disabling the on-board controller and installing an Adaptec
| controller (ad320).
I've an IBM x225 happily running on 5.0.7 with the lsil driver and
mirrored 73 gig hard drives. I left my installation notes with
the machine but I'm pretty sure the driver version ended in .27
After giving up on 5.0.6 and smaller hard drives, the installation
wasn't difficult except IBM had to send me 4 additional hard drives before
I got a pair to work but that wasn't an OS/driver issue.
Would I do it again, no. I'd insist on spending the additional money
for the IBM ServerRAID bits (which would have been overkill on this
If you stuck an Adaptec host adapter in an x225 I'd be interested in
where you found the cable/adapter to connect it to the backplane.
Tom Parsons XXXX@XXXXX.COM
2.Tape verify/restore problem (OSR5.06)
running SCO Unix OSR5.06a on PIV 2.8, 256Mb ram, U160 Adaptec 29160 SCSI
controler, 36gb hdd on 68pin port (external terminator on cable), SLR5
tandberg data tape drive on 50pins port (internal default tandberg
making backups with cpio, or tar, or Lone-Tar, when trying to
verify/restore tape, I always get errors at anywhere on then tape.
(telling that "lone tar, directory not in proper format", "cpio bad
So I put the SLR5 on another machine, it works fine.
I tried another scsi cable, same results ==> errors.
I tried another tape drive (SLR4), same results ==> errors.
So could it be the SCSI controler, or an SCO bug, or any chipset or
3.Form feed problem with Xerox PE120i (SCO Unix OSR5.0x)
running SCO Unix OSR5.05 and 5.06 boxes, Progress V8.3C11 sgbdr.
we recently install on some of our distant sites, Xerox PE120i
multi-function printers (including ethernet port with LPD server)
I configure it as remote printers.
If I send from Unix a file, it works well.
Now from Progress, if I send a file where I put a "page-size" command in
program code, the file prints, and then a blank form feed page is
printed at the end of job.
(the "page-size size-number" command puts ^L in file when line number is
equal to size-number, and puts one ^L at the end of the file).
We've got a Xerox DC535 running under SCO Unix OSR5.06 in BSD remote
format, and this problem doesn't appear.
Is the problem coming from the LPD server on Xerox PE120i ?
Enlever ".remove-me" dans l'adresse.
Remove ".remove-me" from e-mail.
4.iknt panic OSR5.0.4 - telnetd problem?
We are running OSR5.0.4 on a Compaq Proliant 3000 dual PII (w/ SMP), ECC
memory, RAID and Netelligent NIC. We recently started migrating our dumb
serial terminals (on a Digi Xem multiport board) to Ethernet NCD X-terminals
so the Netelligent is under increasingly heavy telnet load. An intermittent
kernel panic has started to become a once-a-day occurrence and practically
crippled us after we added a Sonicwall appliance with VPN access to the
Running _fst and crash, it appears that ikntrput (iknt) - telnetd - is
being fingered as the culprit. The panic screen shows an eip value of
# echo ts F01948ED | crash
dumpfile = /dev/mem, namelist = /unix, outfile = stdout
ikntrput + 0x12d
# _fst /unix -
main: 0x8b55 = push ebp
ikntrput+0x12d: 0x8b66 = mov ax,Word Ptr [eax+0x40]
Required patches were installed long ago - rs504c supplement, oss469c
(Networking & SMP), oss469d (Networking & Core OS), oss471c (PII microcode),
oss481a (SNMP suppl), oss485a (memory suppl), oss480a (HTFS) & oss601(Y2k).
oss605a (Netbios drivers) was not installed, but nb is not found (or used)
on the system.
I found an old thread in Google groups referring to exactly this panic
source (from 2002-12-20 to 2002-12-23) and the last post was from Bela
Lubkin who said:
> iknt does character I/O on behalf of rlogind and/or telnetd, avoiding
> the need to wake up a user process for every character of I/O. The
> panic you're seeing has been seen before; I will see if a fix is
Bela - or anyone - if this panic "has been seen before", was there ever a
fix or cause determined related to telnet, OS, drivers or NIC hardware? I
can try replacing the NIC, upgrading the OS, Compaq EFS (5.46a currently),
replacing the power supply, memory or machine, but if we suffer through
daily system crashes on this production system while I try to guess what's
wrong, I'll probably end up looking for a new job! Any more info or hints
would be greatly appreciated!
5.More info (was Tape verify/restore problem (OSR5.06))
I put the scsi controler on another machine and it works fine.
I put from another machine another U160 29160 scsi controler, and I
always get errors;
the motherboard is an Asus P4PE-X/TE, could it be the cpu speed?, the
Frederic STOCK a rit :
> running SCO Unix OSR5.06a on PIV 2.8, 256Mb ram, U160 Adaptec 29160 SCSI
> controler, 36gb hdd on 68pin port (external terminator on cable), SLR5
> tandberg data tape drive on 50pins port (internal default tandberg
> making backups with cpio, or tar, or Lone-Tar, when trying to
> verify/restore tape, I always get errors at anywhere on then tape.
> (telling that "lone tar, directory not in proper format", "cpio bad
> So I put the SLR5 on another machine, it works fine.
> I tried another scsi cable, same results ==> errors.
> I tried another tape drive (SLR4), same results ==> errors.
> So could it be the SCSI controler, or an SCO bug, or any chipset or
> motherboard ?
> F. STOCK
6. BTLD Floppy disk for Cristie CD1600
7. DOS Disk for BTLD
8. Fresh install with BTLD and no floppy drive