Discussion:
Occasional traps or locks on WPS start
(too old to reply)
d***@sneakemail.com
2007-08-31 19:17:02 UTC
Permalink
Raw Message
I have been running eComStation 2.0 beta 2 milestone 2 (my
subscription expired) for some months, but lately I find a strange
problem:

When the WPS starts, sometimes it gets lock repeating a loop of the
startup sound while the pointer freezes with the clock cursor.
Sometimes it stays there, sometimes it goes on to a trap:
In one of the cases, I took a snapshot of the screen. Here you have
some of the data of the trap screen:

TRAP 0008 ERRCD=0000 ERACC=**** ERLIM=********
EAX=00000090 EBX=ffdd27b4 ECX=00000000 EDX=00000000
ESI=fa290160 EDI=fdc7efb4 EBP=00000006 FLG=00012087
CS:EIP=1200:00002ca7 CSACC=009b CSLIM=000077a8
SS:ESP=0090:00000002 SSACC=0093 SSLIM=000001ff
DS=0160 DSACC=c093 DSLIM=ffffffff CR0=8001001b
ES=0160 ESACC=c093 ESLIM=ffffffff CR2=fdc7efbc
FS=0000 FSACC=**** FSLIM=********
GS=0000 GSACC=**** GSLIM=********

The system detected an internal processing error at
location ##0168:fff1da1f - 000e:ca1f.
60000, 9084

06860662
Internal revision 14.103a_W4

The system is stopped. Record all the above information
contact your service representative.

Strangely this trap screen is so similar to the one at
<news:dggr53$***@ripley.aoltw.net> (
http://groups.google.es/group/netscape.public.mozilla.os2/msg/5cadc2e9877ea1a0
) that I pasted it and rewrote the different values.
However at n.p.mozilla.os2 it is attributed to Mozilla plugins.

Other times I have got CSLIM:ffffffff.

When I hit reset, in some rare occasions the next start proceeds to a
working WPS, but it mostly keeps locking. It could be related to the
INIs, because I have find that if I go back to a certain version of
the INIs, I will have a working WPS, however running Checkini,
Cleanini, Iconomize or FPos to fix the INIs does not guarantee a
working session after I close.
I have also tried shuffling some of the icons in the Startup folder
but it does not seem to have an influence. Currently I have: TCP/IP
startup, CAD-Popup, XCenter, Scheduler Helper, MMOS2 Updater,
LSwitcher, Removable Device Monitor 1.2.

Following some similar messages on the Internet I have updated to the
latest DaniS506 (1.8.1) and I activated UltraDMA5, but I don't see any
difference.

Some idea on what is causing these errors?

(I am going to post this also in news://news.ecomstation.com,
apologies to those seen it more than once)

--
David Mediavilla
d***@sneakemail.com
2007-08-31 19:30:01 UTC
Permalink
Raw Message
Post by d***@sneakemail.com
When the WPS starts, sometimes it gets lock repeating a loop of the
startup sound while the pointer freezes with the clock cursor.
TRAP 0008 ERRCD=0000 ERACC=**** ERLIM=********
CS:EIP=1200:00002ca7 CSACC=009b CSLIM=000077a8
I forgot to say that ExeInfo gives:

ExeInfo, New Exe File Format Info, v2.0f
(C) Copyright Ngb Technologies, 1991-1993, All rights reserved.

E:\OS2KRNL
849372bytes 05-04-29 11:43:56
002 077a9 READABLE EXECUTABLE (preload) (16:16)

--
David Mediavilla
Nobody
2007-09-02 11:27:33 UTC
Permalink
Raw Message
Post by d***@sneakemail.com
Post by d***@sneakemail.com
When the WPS starts, sometimes it gets lock repeating a loop of the
startup sound while the pointer freezes with the clock cursor.
TRAP 0008 ERRCD=0000 ERACC=**** ERLIM=********
CS:EIP=1200:00002ca7 CSACC=009b CSLIM=000077a8
ExeInfo, New Exe File Format Info, v2.0f
(C) Copyright Ngb Technologies, 1991-1993, All rights reserved.
E:\OS2KRNL
849372bytes 05-04-29 11:43:56
002 077a9 READABLE EXECUTABLE (preload) (16:16)
--
David Mediavilla
My experience and answer is different than others.

I have always gotten this message when there is something wrong with a
driver file. For some reason, the startup code in Obj. #2 of OS2KRNL
doesn't re-act or interact correctly as it should. I suspect some
loading procedure or interrupt procedure.

Remove as many drivers as possible, including FLT, ADD, SYS, SNP, IFS,
DMD, etc. and see if the problem goes away.

Or ... make a different bootable source, such as a CD, diskette,
partition, with the exact same files, and see if the problem goes away.

And on ... and on.

Of course there is always going back to a previous setup that did not
have this problem. But, be prepared for it to happen again.
Uwe Pilgram
2007-09-01 18:10:12 UTC
Permalink
Raw Message
Post by d***@sneakemail.com
I have been running eComStation 2.0 beta 2 milestone 2 (my
subscription expired) for some months, but lately I find a strange
When the WPS starts, sometimes it gets lock repeating a loop of the
startup sound while the pointer freezes with the clock cursor.
In one of the cases, I took a snapshot of the screen. Here you have
TRAP 0008 ERRCD=0000 ERACC=**** ERLIM=********
EAX=00000090 EBX=ffdd27b4 ECX=00000000 EDX=00000000
ESI=fa290160 EDI=fdc7efb4 EBP=00000006 FLG=00012087
CS:EIP=1200:00002ca7 CSACC=009b CSLIM=000077a8
SS:ESP=0090:00000002 SSACC=0093 SSLIM=000001ff
DS=0160 DSACC=c093 DSLIM=ffffffff CR0=8001001b
ES=0160 ESACC=c093 ESLIM=ffffffff CR2=fdc7efbc
FS=0000 FSACC=**** FSLIM=********
GS=0000 GSACC=**** GSLIM=********
The system detected an internal processing error at
location ##0168:fff1da1f - 000e:ca1f.
60000, 9084
06860662
Internal revision 14.103a_W4
The system is stopped. Record all the above information
contact your service representative.
Strangely this trap screen is so similar to the one at
http://groups.google.es/group/netscape.public.mozilla.os2/msg/5cadc2e9877ea1a0
) that I pasted it and rewrote the different values.
However at n.p.mozilla.os2 it is attributed to Mozilla plugins.
Other times I have got CSLIM:ffffffff.
When I hit reset, in some rare occasions the next start proceeds to a
working WPS, but it mostly keeps locking. It could be related to the
INIs, because I have find that if I go back to a certain version of
the INIs, I will have a working WPS, however running Checkini,
Cleanini, Iconomize or FPos to fix the INIs does not guarantee a
working session after I close.
I have also tried shuffling some of the icons in the Startup folder
but it does not seem to have an influence. Currently I have: TCP/IP
startup, CAD-Popup, XCenter, Scheduler Helper, MMOS2 Updater,
LSwitcher, Removable Device Monitor 1.2.
Following some similar messages on the Internet I have updated to the
latest DaniS506 (1.8.1) and I activated UltraDMA5, but I don't see any
difference.
Some idea on what is causing these errors?
(I am going to post this also in news://news.ecomstation.com,
apologies to those seen it more than once)
--
David Mediavilla
Hi David,

if your system can be started to a command shell by using ALT-F1 during
the boot blob then you most likely have corrupted INI-files. Try
Checkini.exe (see hobbes) to fix it. Because this requires a complete
boot to WPS you might be forced to create new INI files in a different
directory using MakeIni.exe updating config.sys variables USER.INI and
SYSTEM.INI, booting and then using checkini.exe pointing to your
originals for fixing.

Hope this helps.
Uwe
Uwe Pilgram
2007-09-01 19:07:00 UTC
Permalink
Raw Message
Post by Uwe Pilgram
Post by d***@sneakemail.com
I have been running eComStation 2.0 beta 2 milestone 2 (my
subscription expired) for some months, but lately I find a strange
When the WPS starts, sometimes it gets lock repeating a loop of the
startup sound while the pointer freezes with the clock cursor.
In one of the cases, I took a snapshot of the screen. Here you have
TRAP 0008 ERRCD=0000 ERACC=**** ERLIM=********
EAX=00000090 EBX=ffdd27b4 ECX=00000000 EDX=00000000
ESI=fa290160 EDI=fdc7efb4 EBP=00000006 FLG=00012087
CS:EIP=1200:00002ca7 CSACC=009b CSLIM=000077a8
SS:ESP=0090:00000002 SSACC=0093 SSLIM=000001ff
DS=0160 DSACC=c093 DSLIM=ffffffff CR0=8001001b
ES=0160 ESACC=c093 ESLIM=ffffffff CR2=fdc7efbc
FS=0000 FSACC=**** FSLIM=********
GS=0000 GSACC=**** GSLIM=********
The system detected an internal processing error at
location ##0168:fff1da1f - 000e:ca1f.
60000, 9084
06860662
Internal revision 14.103a_W4
The system is stopped. Record all the above information
contact your service representative.
Strangely this trap screen is so similar to the one at
http://groups.google.es/group/netscape.public.mozilla.os2/msg/5cadc2e9877ea1a0
) that I pasted it and rewrote the different values.
However at n.p.mozilla.os2 it is attributed to Mozilla plugins.
Other times I have got CSLIM:ffffffff.
When I hit reset, in some rare occasions the next start proceeds to a
working WPS, but it mostly keeps locking. It could be related to the
INIs, because I have find that if I go back to a certain version of
the INIs, I will have a working WPS, however running Checkini,
Cleanini, Iconomize or FPos to fix the INIs does not guarantee a
working session after I close.
I have also tried shuffling some of the icons in the Startup folder
but it does not seem to have an influence. Currently I have: TCP/IP
startup, CAD-Popup, XCenter, Scheduler Helper, MMOS2 Updater,
LSwitcher, Removable Device Monitor 1.2.
Following some similar messages on the Internet I have updated to the
latest DaniS506 (1.8.1) and I activated UltraDMA5, but I don't see any
difference.
Some idea on what is causing these errors?
(I am going to post this also in news://news.ecomstation.com,
apologies to those seen it more than once)
--
David Mediavilla
Hi David,
if your system can be started to a command shell by using ALT-F1 during
the boot blob then you most likely have corrupted INI-files. Try
Checkini.exe (see hobbes) to fix it. Because this requires a complete
boot to WPS you might be forced to create new INI files in a different
directory using MakeIni.exe updating config.sys variables USER.INI and
SYSTEM.INI, booting and then using checkini.exe pointing to your
originals for fixing.
Hope this helps.
Uwe
Hi David,

I have forgotten to say that OS2SYS.INI sometimes gets corrupted during
shutdown. The "object to path" or "handle" table is not completely
written. This would exactly lead to the symptoms you describe. You can
see that easily when you note the size of OS2SYS.INI and compare it
after a bad start of WPS after boot. If smaller then OS2SYS.INI is
corrupted. I could correct that by using kernel 14.105_W4 in my Warp
4.52 system (not ECS). No more problems with WPS since then.

Regards
Uwe
d***@sneakemail.com
2007-09-02 13:33:06 UTC
Permalink
Raw Message
Post by Uwe Pilgram
Post by Uwe Pilgram
Post by d***@sneakemail.com
I have been running eComStation 2.0 beta 2 milestone 2 (my
subscription expired) for some months, but lately I find a strange
When the WPS starts, sometimes it gets lock repeating a loop of the
startup sound while the pointer freezes with the clock cursor.
The system detected an internal processing error at
location ##0168:fff1da1f - 000e:ca1f.
Some idea on what is causing these errors?
--
David Mediavilla
if your system can be started to a command shell by using ALT-F1 during
the boot blob then you most likely have corrupted INI-files. Try
Checkini.exe (see hobbes) to fix it. Because this requires a complete
I have already tried running Checkini and Cleanini before shutdown.
I'm not sure if it helps. Sometimes I get trouble even after running
them.
Post by Uwe Pilgram
Post by Uwe Pilgram
Uwe
I have forgotten to say that OS2SYS.INI sometimes gets corrupted during
shutdown. The "object to path" or "handle" table is not completely
written. This would exactly lead to the symptoms you describe. You can
see that easily when you note the size of OS2SYS.INI and compare it
after a bad start of WPS after boot. If smaller then OS2SYS.INI is
Comparing the sizes of bad and good INIs does not show very big
differences. Anyway, running XFix before the WPS start does not show
but a few problems and solving them is no guarantee.
Post by Uwe Pilgram
corrupted. I could correct that by using kernel 14.105_W4 in my Warp
4.52 system (not ECS). No more problems with WPS since then.
I also have Beta 3 but could not install it. Perhaps I can extract it
and copy it from my maintenance partition.
Post by Uwe Pilgram
Uwe
--
David Mediavilla
Uwe Pilgram
2007-09-03 16:05:51 UTC
Permalink
Raw Message
Well, the problem is that during shutdown, the drivers are informed of
this event.

My experience is, that UNI kernel used with an OS/2 4.52 and not with
OS/2 Server does not handle the switch to "Shutdown" mode correctly. WPS
tries to flush the INIs (BTW: uncached)and this process is not
completed. Consequently the "Handle" table in OS2SYS.INI, which is often
written last and made up of many little records, may be corrupted.
OS2SYS.INI must have an identical length before shutdown and after boot
to a fullscreen CMD. via ALT-F1, provided that you gave the system a few
minutes to quiesce before shutdown. So note size and date/time of
OS2SYS.INI before shutdown and after boot to FS CMD and if they are
different even by only a few bytes, this is a strong indication of a
kernel problem.

Regards
Uwe
Post by d***@sneakemail.com
Post by Uwe Pilgram
Post by Uwe Pilgram
Post by d***@sneakemail.com
I have been running eComStation 2.0 beta 2 milestone 2 (my
subscription expired) for some months, but lately I find a strange
When the WPS starts, sometimes it gets lock repeating a loop of the
startup sound while the pointer freezes with the clock cursor.
The system detected an internal processing error at
location ##0168:fff1da1f - 000e:ca1f.
Some idea on what is causing these errors?
--
David Mediavilla
if your system can be started to a command shell by using ALT-F1 during
the boot blob then you most likely have corrupted INI-files. Try
Checkini.exe (see hobbes) to fix it. Because this requires a complete
I have already tried running Checkini and Cleanini before shutdown.
I'm not sure if it helps. Sometimes I get trouble even after running
them.
Post by Uwe Pilgram
Post by Uwe Pilgram
Uwe
I have forgotten to say that OS2SYS.INI sometimes gets corrupted during
shutdown. The "object to path" or "handle" table is not completely
written. This would exactly lead to the symptoms you describe. You can
see that easily when you note the size of OS2SYS.INI and compare it
after a bad start of WPS after boot. If smaller then OS2SYS.INI is
Comparing the sizes of bad and good INIs does not show very big
differences. Anyway, running XFix before the WPS start does not show
but a few problems and solving them is no guarantee.
Post by Uwe Pilgram
corrupted. I could correct that by using kernel 14.105_W4 in my Warp
4.52 system (not ECS). No more problems with WPS since then.
I also have Beta 3 but could not install it. Perhaps I can extract it
and copy it from my maintenance partition.
Post by Uwe Pilgram
Uwe
--
David Mediavilla
csola
2007-09-04 09:10:30 UTC
Permalink
Raw Message
Boot jfs partition to be out of sight in the (IBM) bootmanager ! This
to look HPFS... Bug?
Scott G.
2007-09-04 19:03:02 UTC
Permalink
Raw Message
Boot jfs partition to be out of sight in the (IBM) bootmanager ! This to
look HPFS... Bug?
search in old postings to see discussions. basically, this is a (lack
of) feature, not bug.
Daniela Engert
2007-09-05 17:27:49 UTC
Permalink
Raw Message
Hi Scott!
Post by Scott G.
Post by csola
Boot jfs partition to be out of sight in the (IBM) bootmanager ! This
to look HPFS... Bug?
search in old postings to see discussions. basically, this is a (lack
of) feature, not bug.
Well, it depends on the type of JFS partition. In case of a LVM
partition (type 35h) JFS can't be booted from BM, but in case of a
compatibility partition (type 07h) you can. In fact, OS/2 boots
noticably faster from a JFS partition than any other type, be it FAT or
HPFS/HPFS386.

Ciao,
Dani
James J. Weinkam
2007-09-06 01:58:16 UTC
Permalink
Raw Message
Post by Daniela Engert
Hi Scott!
Post by Scott G.
Post by csola
Boot jfs partition to be out of sight in the (IBM) bootmanager ! This
to look HPFS... Bug?
search in old postings to see discussions. basically, this is a (lack
of) feature, not bug.
Well, it depends on the type of JFS partition. In case of a LVM
partition (type 35h) JFS can't be booted from BM, but in case of a
compatibility partition (type 07h) you can. In fact, OS/2 boots
noticably faster from a JFS partition than any other type, be it FAT or
HPFS/HPFS386.
Ciao,
Dani
I have been testing eCS2.0-RC2. In particular I tried it on a JFS partition. At
various milestones I made complete backups from a maintenance partition. After
the latest backup, the partition would not boot. I booted from the maintenance
partition and ran JFSCHK32 which said that the partition was clean. It still
wouldn't boot.

To be more specific here is what happens when the eCS2.0RC2 installation on the
JFS volume will not boot: The bootmanager screen appears and I select the JFS
installation. The boot blob appears with a line at the bottom about the eCS JFS
boot code. The eCS logo appears.

Following this one of two things has happened on each of many attempts:

1. The eCS logo stays on the screen and nothing happens. Ctrl-Alt-Del does
nothing. Power down in necessary

2. The ACPI driver loads, followed by the message, "This driver is licensed only
for use with e Com Station. The system is unable to operate your hard drive, the
system is stopped." Again power down is necessary.

I tried booting from the maintenance partition again, running JFSCHK32, and then
powering down and restarting instead of doing a warm boot with the same results.

I also tried booting from the install disk and running chkdsk from the
maintenance console. That didn't help either.

I tried restoring several backups, starting with the newest but none of them
would boot.

I then tried booting from the install disk, deleting the partition, then running
the install up to the point of having created and formatted a new partition,
then canceling the install and restoring a backup, with the same results.

Finally, I booted from the maintenance partition, formatted the partition as
HPFS, restored the latest backup and ran sysinstx.com. That boots fine and
everything works.

End of long story.

My question is how can one recover from a corrupted bootable JFS partition short
of a complete reinstall?

Can anyone suggest anything I haven't tried?
Percival P. Cassidy
2007-09-06 15:18:59 UTC
Permalink
Raw Message
Post by James J. Weinkam
I have been testing eCS2.0-RC2. In particular I tried it on a JFS
partition. At various milestones I made complete backups from a
maintenance partition. After the latest backup, the partition would not
boot. I booted from the maintenance partition and ran JFSCHK32 which
said that the partition was clean. It still wouldn't boot.
To be more specific here is what happens when the eCS2.0RC2 installation
on the JFS volume will not boot: The bootmanager screen appears and I
select the JFS installation. The boot blob appears with a line at the
bottom about the eCS JFS boot code. The eCS logo appears.
1. The eCS logo stays on the screen and nothing happens. Ctrl-Alt-Del
does nothing. Power down in necessary
2. The ACPI driver loads, followed by the message, "This driver is
licensed only for use with e Com Station. The system is unable to
operate your hard drive, the system is stopped." Again power down is
necessary.
I tried booting from the maintenance partition again, running JFSCHK32,
and then powering down and restarting instead of doing a warm boot with
the same results.
I also tried booting from the install disk and running chkdsk from the
maintenance console. That didn't help either.
I tried restoring several backups, starting with the newest but none of
them would boot.
I then tried booting from the install disk, deleting the partition, then
running the install up to the point of having created and formatted a
new partition, then canceling the install and restoring a backup, with
the same results.
Finally, I booted from the maintenance partition, formatted the
partition as HPFS, restored the latest backup and ran sysinstx.com.
That boots fine and everything works.
End of long story.
My question is how can one recover from a corrupted bootable JFS
partition short of a complete reinstall?
Can anyone suggest anything I haven't tried?
How are you doing your backups? What backup program? Are you sure you're
copying all the files with Hidden and System attributes? What about the EAs?

Perce
James J. Weinkam
2007-09-06 18:30:12 UTC
Permalink
Raw Message
Post by Percival P. Cassidy
Post by James J. Weinkam
I have been testing eCS2.0-RC2. In particular I tried it on a JFS
partition. At various milestones I made complete backups from a
maintenance partition. After the latest backup, the partition would
not boot. I booted from the maintenance partition and ran JFSCHK32
which said that the partition was clean. It still wouldn't boot.
To be more specific here is what happens when the eCS2.0RC2
installation on the JFS volume will not boot: The bootmanager screen
appears and I select the JFS installation. The boot blob appears with
a line at the bottom about the eCS JFS boot code. The eCS logo appears.
1. The eCS logo stays on the screen and nothing happens. Ctrl-Alt-Del
does nothing. Power down in necessary
2. The ACPI driver loads, followed by the message, "This driver is
licensed only for use with e Com Station. The system is unable to
operate your hard drive, the system is stopped." Again power down is
necessary.
I tried booting from the maintenance partition again, running
JFSCHK32, and then powering down and restarting instead of doing a
warm boot with the same results.
I also tried booting from the install disk and running chkdsk from the
maintenance console. That didn't help either.
I tried restoring several backups, starting with the newest but none
of them would boot.
I then tried booting from the install disk, deleting the partition,
then running the install up to the point of having created and
formatted a new partition, then canceling the install and restoring a
backup, with the same results.
Finally, I booted from the maintenance partition, formatted the
partition as HPFS, restored the latest backup and ran sysinstx.com.
That boots fine and everything works.
End of long story.
My question is how can one recover from a corrupted bootable JFS
partition short of a complete reinstall?
Can anyone suggest anything I haven't tried?
How are you doing your backups? What backup program? Are you sure you're
copying all the files with Hidden and System attributes? What about the EAs?
Perce
PKZip. All structure is preserved including attributes, EA's and empty
directories. As I said in my original post, if I restore the backup to a HPFS
partition with the same drive letter and run sysinstx.com to get the HPFS
version of OS2BOOT, the partition boots and the system runs correctly.
Therefore there is nothing wrong with the installed eCS system per se or the
backup. The problem is in the JFS boot procedure
d***@sneakemail.com
2007-09-06 19:50:29 UTC
Permalink
Raw Message
Post by James J. Weinkam
I have been testing eCS2.0-RC2. In particular I tried it on a JFS partition. At
1. The eCS logo stays on the screen and nothing happens. Ctrl-Alt-Del does
nothing. Power down in necessary
It happens to me as well in Beta 2 Milestone 2. If you have a diskette
drive, you'll probably see the light is on.

Removing OS2LOGO and turning on the lines showing drivers as they load
will allow you to see the driver after which it is getting blocked. I
think I probably could solve it with a later version of ACPI or
removing ACPI and using APM instead, but I never get around to it.

About the problem when it is "working", I don't know.

--
David Mediavilla
Rich Walsh
2007-09-06 23:27:25 UTC
Permalink
Raw Message
Post by James J. Weinkam
To be more specific here is what happens when the eCS2.0RC2 installation on the
JFS volume will not boot: The bootmanager screen appears and I select the JFS
installation. The boot blob appears with a line at the bottom about the eCS JFS
boot code. The eCS logo appears.
1. The eCS logo stays on the screen and nothing happens. Ctrl-Alt-Del does
nothing. Power down in necessary
2. The ACPI driver loads, followed by the message, "This driver is licensed only
for use with e Com Station. The system is unable to operate your hard drive, the
system is stopped." Again power down is necessary.
The "OS/2 is unable to operate..." message is the result of letting older
versions of OS/2-eCS autocheck your v2.0 drive. The boot sector doesn't
look right to a non-bootable version of JFS, so it "fixes" it, rendering it
unbootable. On your older installations, you have to change the JFS /AUTOCHECK
parameter from '*' to an explicit list of drives that excludes the bootable
volume.

E.g., my eCS 2.0 rc2 drive is letter 'K', so the IFS line for my primary
system (on an HPFS drive F:) is:

IFS=F:\OS2\JFS.IFS /LW:5,20,4 /AUTOCHECK:GHIM

FYI, this may disable access to the drive in question from the older system
if the volume is dirty or its non-bootable JFS _thinks_ it's dirty.

Also... for rc1, I had to remove PSD=ACPI.PSD to avoid random hangs during
bootup. OTOH, with rc2, it actually works! (sorta) For the first time
ever under OS/2, shutdown turns off the machine. Sadly, 'suspend' does
absolutely nothing.
--
== == almost usable email address: rws AT e-vertise.com == ==
___________________________________________________________________
|
| Remote Workplace Server v0.80
Rich Walsh | interact with the WPS from any program
Ft Myers, FL | http://e-vertise.com/rws/rws080.zip
___________________________________________________________________
William L. Hartzell
2007-09-07 02:38:04 UTC
Permalink
Raw Message
Sir:

Rich Walsh wrote:
<snip>
Post by Rich Walsh
Also... for rc1, I had to remove PSD=ACPI.PSD to avoid random hangs during
bootup. OTOH, with rc2, it actually works! (sorta) For the first time
ever under OS/2, shutdown turns off the machine. Sadly, 'suspend' does
absolutely nothing.
FYI, Shutdown turns off my machine (Athlon XP 2K) and I'm only using
APM. I've had several main boards that would work with APM and several
that would not. It is all in the BIOS. The only time I ever had
suspend work on my desktop machine was back on OS/2 2.11, or was that
hibernate?
--
Bill
Thanks a Million!
James J. Weinkam
2007-09-07 03:03:33 UTC
Permalink
Raw Message
Post by Rich Walsh
Post by James J. Weinkam
To be more specific here is what happens when the eCS2.0RC2 installation on the
JFS volume will not boot: The bootmanager screen appears and I select the JFS
installation. The boot blob appears with a line at the bottom about the eCS JFS
boot code. The eCS logo appears.
1. The eCS logo stays on the screen and nothing happens. Ctrl-Alt-Del does
nothing. Power down in necessary
2. The ACPI driver loads, followed by the message, "This driver is licensed only
for use with e Com Station. The system is unable to operate your hard drive, the
system is stopped." Again power down is necessary.
The "OS/2 is unable to operate..." message is the result of letting older
versions of OS/2-eCS autocheck your v2.0 drive. The boot sector doesn't
look right to a non-bootable version of JFS, so it "fixes" it, rendering it
unbootable. On your older installations, you have to change the JFS /AUTOCHECK
parameter from '*' to an explicit list of drives that excludes the bootable
volume.
E.g., my eCS 2.0 rc2 drive is letter 'K', so the IFS line for my primary
IFS=F:\OS2\JFS.IFS /LW:5,20,4 /AUTOCHECK:GHIM
FYI, this may disable access to the drive in question from the older system
if the volume is dirty or its non-bootable JFS _thinks_ it's dirty.
Also... for rc1, I had to remove PSD=ACPI.PSD to avoid random hangs during
bootup. OTOH, with rc2, it actually works! (sorta) For the first time
ever under OS/2, shutdown turns off the machine. Sadly, 'suspend' does
absolutely nothing.
Thank you for your thoughtful suggestions.

I am not entirely convinced that letting mcp2 check the drive is responsible
because as I had stated in my original post, I had backed up several times as I
reached milestones I wanted to be able to return to in case of trouble. On some
of these occasions, I did more than back up the partition, such as fixing
config,sys when a typo made the partition unbootable. In short, the volume was
check disked and modified by the mcp2 installation at least eight or ten times
before it suddenly refused to boot. I think this shows there is more to it than
autocheck fouling up the partition.

Nevertheless, based on your statement, I just did the following:

1. Exteacr the JFS version of OS2BOOT from one of the backups of the JFS
partition to h:\ecs.

2. Shut down and boot from the install disk.

3. Go to the maintenance console and use the disk->volume management tool to
chnage the drive letter of the HPFS version of eCS2.0RC2 that I have been
running the last couple of days from i to n.

4. Resume the advanced install and create a new volume lettered i and long
format it JFS.

5. Cancel the install and return to management mode.

6. From a command line, XCOPY n: i: /h/o/t/s/e/r/v
COPY h:\ecs\OS2BOOT i:\

I did this rather than restore one of the JFS backups because I didn't want to
lose the changes that have been made since I converted to HPFS,

7. Shutdown and reboot.

This booted on the third try. On the first two tries it reached the eCS logo
and sat there for several minutes until I gave up. Unlike the previous
occasions, however, the system was not frozen; Ctrl-Alt-Delete worked.

So far I have not remed out ACPI.

I will mess around with this for a few days and see if I can narrow the problem
down. If something is causing "random" booting problems, the probability of
boot failure does not seem to be constant. I must have done at least twenty
boots of the initial JFS installation (not all of them involving going to mcp2
in between) before the initial failure occurred. Once that happened the first
time I was able to get in more successful boots by booting from the maintenance
partition mcp2 and running JFSCHK32. (I left out this detail in my original post
for the sake of brevity.) After that the next time it happened nothing I could
think of restored bootability.

Thanks again for the tip.
Peter Brown
2007-09-07 13:53:05 UTC
Permalink
Raw Message
Hi James
Post by James J. Weinkam
Post by Rich Walsh
Post by James J. Weinkam
To be more specific here is what happens when the eCS2.0RC2
installation on the JFS volume will not boot: The bootmanager screen
appears and I select the JFS installation. The boot blob appears with
a line at the bottom about the eCS JFS boot code. The eCS logo appears.
1. The eCS logo stays on the screen and nothing happens.
Ctrl-Alt-Del does nothing. Power down in necessary
2. The ACPI driver loads, followed by the message, "This driver is
licensed only for use with e Com Station. The system is unable to
operate your hard drive, the system is stopped." Again power down is
necessary.
The "OS/2 is unable to operate..." message is the result of letting older
versions of OS/2-eCS autocheck your v2.0 drive. The boot sector doesn't
look right to a non-bootable version of JFS, so it "fixes" it, rendering it
unbootable. On your older installations, you have to change the JFS /AUTOCHECK
parameter from '*' to an explicit list of drives that excludes the bootable
volume.
E.g., my eCS 2.0 rc2 drive is letter 'K', so the IFS line for my primary
IFS=F:\OS2\JFS.IFS /LW:5,20,4 /AUTOCHECK:GHIM
FYI, this may disable access to the drive in question from the older system
if the volume is dirty or its non-bootable JFS _thinks_ it's dirty.
Also... for rc1, I had to remove PSD=ACPI.PSD to avoid random hangs during
bootup. OTOH, with rc2, it actually works! (sorta) For the first time
ever under OS/2, shutdown turns off the machine. Sadly, 'suspend' does
absolutely nothing.
Thank you for your thoughtful suggestions.
I am not entirely convinced that letting mcp2 check the drive is
responsible because as I had stated in my original post,
I seem to remember words to the effect of

Do NOT let a system with non-bootable jfs chkdsk a bootable jfs system
as that will lead to problems with the bootable jfs system.

As mcp2 in "non-bootable jfs" I think you have found the cause of the
problem.


Last eCS2 beta I tested was v3 - Software Subscription has run out and
I'm skint - and I'm fairly sure sure it mentioned that in the readme or
in a jfs specific readme.

I also seem to recall seeing posts about this either in the ecs ngs or
the ecsTestTeam mailing list.


To get compatibility between the systems you would need to update the
"non-bootable jfs" system with some of the files from the "bootable jfs"
system. Sorry, I never did work out which files were needed but maybe a
Google on the subject will help.


Regards

Pete



I had backed up
Post by James J. Weinkam
several times as I reached milestones I wanted to be able to return to
in case of trouble. On some of these occasions, I did more than back up
the partition, such as fixing config,sys when a typo made the partition
unbootable. In short, the volume was check disked and modified by the
mcp2 installation at least eight or ten times before it suddenly refused
to boot. I think this shows there is more to it than autocheck fouling
up the partition.
1. Exteacr the JFS version of OS2BOOT from one of the backups of the JFS
partition to h:\ecs.
2. Shut down and boot from the install disk.
3. Go to the maintenance console and use the disk->volume management
tool to chnage the drive letter of the HPFS version of eCS2.0RC2 that I
have been running the last couple of days from i to n.
4. Resume the advanced install and create a new volume lettered i and
long format it JFS.
5. Cancel the install and return to management mode.
6. From a command line, XCOPY n: i: /h/o/t/s/e/r/v
COPY h:\ecs\OS2BOOT i:\
I did this rather than restore one of the JFS backups because I didn't
want to lose the changes that have been made since I converted to HPFS,
7. Shutdown and reboot.
This booted on the third try. On the first two tries it reached the eCS
logo and sat there for several minutes until I gave up. Unlike the
previous occasions, however, the system was not frozen; Ctrl-Alt-Delete
worked.
So far I have not remed out ACPI.
I will mess around with this for a few days and see if I can narrow the
problem down. If something is causing "random" booting problems, the
probability of boot failure does not seem to be constant. I must have
done at least twenty boots of the initial JFS installation (not all of
them involving going to mcp2 in between) before the initial failure
occurred. Once that happened the first time I was able to get in more
successful boots by booting from the maintenance partition mcp2 and
running JFSCHK32. (I left out this detail in my original post for the
sake of brevity.) After that the next time it happened nothing I could
think of restored bootability.
Thanks again for the tip.
Doug Bissett
2007-09-12 20:23:46 UTC
Permalink
Raw Message
On Fri, 7 Sep 2007 13:53:05 UTC, Peter Brown
Post by Peter Brown
To get compatibility between the systems you would need to update the
"non-bootable jfs" system with some of the files from the "bootable jfs"
system. Sorry, I never did work out which files were needed but maybe a
Google on the subject will help.
All I do (and it has always worked), is copy UJFS.DLL from the newest
eCS 2.0 version, over top of the older versions of that file (be sure
to get all of them). No more problems with CHKDSK messing up the
bootable JFS drives.

Hope this helps...
--
From the eComStation 2.0 RC2 of Doug Bissett
dougb007 at telus dot net
(Please make the obvious changes, to e-mail me)
Rich Walsh
2007-09-08 06:13:53 UTC
Permalink
Raw Message
Post by James J. Weinkam
Post by Rich Walsh
The "OS/2 is unable to operate..." message is the result of letting older
versions of OS/2-eCS autocheck your v2.0 drive. The boot sector doesn't
look right to a non-bootable version of JFS, so it "fixes" it, rendering it
unbootable. On your older installations, you have to change the JFS /AUTOCHECK
parameter from '*' to an explicit list of drives that excludes the bootable
volume.
I am not entirely convinced that letting mcp2 check the drive is responsible
because as I had stated in my original post, I had backed up several times as I
reached milestones I wanted to be able to return to in case of trouble. On some
of these occasions, I did more than back up the partition, such as fixing
config,sys when a typo made the partition unbootable. In short, the volume was
check disked and modified by the mcp2 installation at least eight or ten times
before it suddenly refused to boot. I think this shows there is more to it than
autocheck fouling up the partition.
My recollection of some of the particulars is fuzzy. It may be that JFS chkdsk
only rewrites the boot sector when it finds the volume is dirty & it has to do
more than just a quick check of the journal.
--
== == almost usable email address: rws AT e-vertise.com == ==
___________________________________________________________________
|
| Remote Workplace Server v0.80
Rich Walsh | interact with the WPS from any program
Ft Myers, FL | http://e-vertise.com/rws/rws080.zip
___________________________________________________________________
James J. Weinkam
2007-09-09 07:16:37 UTC
Permalink
Raw Message
Post by Rich Walsh
Post by James J. Weinkam
Post by Rich Walsh
The "OS/2 is unable to operate..." message is the result of letting older
versions of OS/2-eCS autocheck your v2.0 drive. The boot sector doesn't
look right to a non-bootable version of JFS, so it "fixes" it, rendering it
unbootable. On your older installations, you have to change the JFS /AUTOCHECK
parameter from '*' to an explicit list of drives that excludes the bootable
volume.
I am not entirely convinced that letting mcp2 check the drive is responsible
because as I had stated in my original post, I had backed up several times as I
reached milestones I wanted to be able to return to in case of trouble. On some
of these occasions, I did more than back up the partition, such as fixing
config,sys when a typo made the partition unbootable. In short, the volume was
check disked and modified by the mcp2 installation at least eight or ten times
before it suddenly refused to boot. I think this shows there is more to it than
autocheck fouling up the partition.
My recollection of some of the particulars is fuzzy. It may be that JFS chkdsk
only rewrites the boot sector when it finds the volume is dirty & it has to do
more than just a quick check of the journal.
Based on a few additional experiments, I think that the probability of the
partition not booting goes way up when anything has been written to it from the
mcp2 partition. Even then it sometimes boots, but apparently usually not.

I also added altf2on.$$$ to the root of I. The partition then booted even after
being modified but this sometimes happened anyway so it is not conclusive.
Nevertheless it may be that the problem is some sort of race condition that the
delay of displaying driver names circumvents.

Subsequently, I copied the following files from the RC2 partition to the mcp2 one:

I:\OS2\CACHEJFS.EXE
I:\OS2\CHKDSK.COM
I:\OS2\BOOT\CHKDSK.SYS
I:\OS2\CHKLGJFS.EXE
I:\OS2\CLRBBLKS.EXE
I:\OS2\DEFRAGFS.EXE
I:\OS2\EXTENDFS.EXE
I:\OS2\JFS.IFS
I:\OS2\JFS.MSG
I:\OS2\SYSTEM\TRACE\jfs.tdf
I:\OS2\JFSCHK32.EXE
I:\OS2\JFSH.MSG
I:\OS2\LOGDUMP.EXE
I:\OS2\DLL\UJFS.DLL

It was probably unnecessary to copy the CHKDSK.* files.

Since then the partition has not failed to boot regardless of what was done in mcp2.

There are too many variables to decide for sure what cured the problem, but it
seems to have gone away.
d***@sneakemail.com
2007-09-04 17:28:58 UTC
Permalink
Raw Message
Post by Uwe Pilgram
My experience is, that UNI kernel used with an OS/2 4.52 and not with
OS/2 Server does not handle the switch to "Shutdown" mode correctly. WPS
tries to flush the INIs (BTW: uncached)and this process is not
completed. Consequently the "Handle" table in OS2SYS.INI, which is often
written last and made up of many little records, may be corrupted.
OS2SYS.INI must have an identical length before shutdown and after boot
to a fullscreen CMD. via ALT-F1, provided that you gave the system a few
minutes to quiesce before shutdown. So note size and date/time of
OS2SYS.INI before shutdown and after boot to FS CMD and if they are
different even by only a few bytes, this is a strong indication of a
kernel problem.
In the occasion I checked, it had the same length before shutdown and
after booting to CMD. But I got a trap or lock when booting to WPS, so
I reverted INIs.

Thanks for the suggestion, anyway.
Post by Uwe Pilgram
Uwe
Post by d***@sneakemail.com
Post by Uwe Pilgram
Post by Uwe Pilgram
Post by d***@sneakemail.com
I have been running eComStation 2.0 beta 2 milestone 2 (my
subscription expired) for some months, but lately I find a strange
When the WPS starts, sometimes it gets lock repeating a loop of the
startup sound while the pointer freezes with the clock cursor.
The system detected an internal processing error at
location ##0168:fff1da1f - 000e:ca1f.
--
David Mediavilla
if your system can be started to a command shell by using ALT-F1 during
the boot blob then you most likely have corrupted INI-files. Try
Checkini.exe (see hobbes) to fix it. Because this requires a complete
I have already tried running Checkini and Cleanini before shutdown.
I'm not sure if it helps. Sometimes I get trouble even after running
them.
Post by Uwe Pilgram
Post by Uwe Pilgram
Uwe
I have forgotten to say that OS2SYS.INI sometimes gets corrupted during
shutdown. The "object to path" or "handle" table is not completely
Comparing the sizes of bad and good INIs does not show very big
differences. Anyway, running XFix before the WPS start does not show
Post by Uwe Pilgram
Uwe
--
David Mediavilla
--
David Mediavilla
Scott G
2007-09-02 04:45:24 UTC
Permalink
Raw Message
You have a hardware problem where you're having too many interrupts from
some device. Try pulling unnecessary cards out -- start with your sound
card, based on your description.
d***@sneakemail.com
2007-09-02 13:39:23 UTC
Permalink
Raw Message
Post by Scott G
You have a hardware problem where you're having too many interrupts from
some device. Try pulling unnecessary cards out -- start with your sound
card, based on your description.
Thanks, but I don't think so. The RMView does not show conflicts with
interrupts:
RMVIEW: Physical view
IRQ Level = 0 PCI Pin = NONE Flg = EXCLUSIVE TIMER_CH_0

IRQ Level = 1 PCI Pin = NONE Flg = EXCLUSIVE KBD_0 Keyboard
Controller

IRQ Level = 2 PCI Pin = NONE Flg = EXCLUSIVE PIC_1

IRQ Level = 3 PCI Pin = NONE Flg = MULTIPLEXED SERIAL_1 Serial
Controller

IRQ Level = 4 PCI Pin = NONE Flg = MULTIPLEXED SERIAL_0 Serial
Controller

IRQ Level = 5 PCI Pin = D Flg = SHARED UHCI Compliant
USB Host Controller

IRQ Level = 5 PCI Pin = D Flg = SHARED UHCI Compliant
USB Host Controller

IRQ Level = 5 PCI Pin = D Flg = SHARED UHCI Compliant
USB Host Controller

IRQ Level = 6 PCI Pin = NONE Flg = MULTIPLEXED FLOPPY_0 Floppy
Controller

IRQ Level = 8 PCI Pin = NONE Flg = EXCLUSIVE RTC

IRQ Level = 10 PCI Pin = NONE Flg = SHARED Creative Labs
SoundBlaster Live!

IRQ Level = 11 PCI Pin = A Flg = SHARED ETHERNET_0
Realtek 8100B/8139D

IRQ Level = 12 PCI Pin = NONE Flg = EXCLUSIVE AUX_0 PS/2
Auxiliary Device Controller

IRQ Level = 14 PCI Pin = NONE Flg = EXCLUSIVE IDE_0 xATA
Controller

IRQ Level = 15 PCI Pin = NONE Flg = EXCLUSIVE IDE_1 xATA
Controller

--
David Mediavilla
Steve Wendt
2007-09-02 17:01:38 UTC
Permalink
Raw Message
Post by d***@sneakemail.com
Post by Scott G
You have a hardware problem where you're having too many interrupts from
some device. Try pulling unnecessary cards out -- start with your sound
card, based on your description.
Thanks, but I don't think so. The RMView does not show conflicts with
1) Scott didn't say it was an IRQ conflict, he said there were too many
hardware interrupts (not the same thing).

2) There's no such thing as an IRQ conflict for PCI devices (except in
the case of buggy drivers).

3) I'm not sure that rmview always gives you a complete picture.
Steve Wendt
2007-09-02 17:05:34 UTC
Permalink
Raw Message
Post by d***@sneakemail.com
Post by Scott G
You have a hardware problem where you're having too many interrupts from
some device. Try pulling unnecessary cards out -- start with your sound
card, based on your description.
IRQ Level = 10 PCI Pin = NONE Flg = SHARED Creative Labs
SoundBlaster Live!
I seem to recall reading that at least some variants of the SB Live!
generated ridiculous amounts of hardware interrupts, which caused
problems with at least some motherboard chipsets.

Google should find more info, such as:
"However, there have been numerous reports about serious problems with
the Live! and AMD motherboards that use the 686b southbridge."
Uwe Pilgram
2007-09-03 15:51:16 UTC
Permalink
Raw Message
HI,

I have a little tool which shows the interrupt activity. Send me a
private mail I you would like to experiment with it.

Regards
Uwe

--------------------
Post by d***@sneakemail.com
Post by Scott G
You have a hardware problem where you're having too many interrupts from
some device. Try pulling unnecessary cards out -- start with your sound
card, based on your description.
Thanks, but I don't think so. The RMView does not show conflicts with
RMVIEW: Physical view
IRQ Level = 0 PCI Pin = NONE Flg = EXCLUSIVE TIMER_CH_0
IRQ Level = 1 PCI Pin = NONE Flg = EXCLUSIVE KBD_0 Keyboard
Controller
IRQ Level = 2 PCI Pin = NONE Flg = EXCLUSIVE PIC_1
IRQ Level = 3 PCI Pin = NONE Flg = MULTIPLEXED SERIAL_1 Serial
Controller
IRQ Level = 4 PCI Pin = NONE Flg = MULTIPLEXED SERIAL_0 Serial
Controller
IRQ Level = 5 PCI Pin = D Flg = SHARED UHCI Compliant
USB Host Controller
IRQ Level = 5 PCI Pin = D Flg = SHARED UHCI Compliant
USB Host Controller
IRQ Level = 5 PCI Pin = D Flg = SHARED UHCI Compliant
USB Host Controller
IRQ Level = 6 PCI Pin = NONE Flg = MULTIPLEXED FLOPPY_0 Floppy
Controller
IRQ Level = 8 PCI Pin = NONE Flg = EXCLUSIVE RTC
IRQ Level = 10 PCI Pin = NONE Flg = SHARED Creative Labs
SoundBlaster Live!
IRQ Level = 11 PCI Pin = A Flg = SHARED ETHERNET_0
Realtek 8100B/8139D
IRQ Level = 12 PCI Pin = NONE Flg = EXCLUSIVE AUX_0 PS/2
Auxiliary Device Controller
IRQ Level = 14 PCI Pin = NONE Flg = EXCLUSIVE IDE_0 xATA
Controller
IRQ Level = 15 PCI Pin = NONE Flg = EXCLUSIVE IDE_1 xATA
Controller
--
David Mediavilla
Scott G
2007-09-04 14:06:18 UTC
Permalink
Raw Message
Post by Scott G
You have a hardware problem where you're having too many interrupts from
some device. Try pulling unnecessary cards out -- start with your sound
card, based on your description.
I didn't say there was a conflict, though that's not a bad guess,
necessarily. I said you're getting too many interrupts. You probably
have something defective. You can mess around with INI files or whtever,
but that won't help. Trust me, I dealt with your kind of problem for a
loong time before finally leaving OS/2. Do what I suggest or don't, but
you won't get better advice.
regards, Scott
d***@sneakemail.com
2007-09-04 17:59:32 UTC
Permalink
Raw Message
Post by Scott G
Post by Scott G
You have a hardware problem where you're having too many interrupts from
some device. Try pulling unnecessary cards out -- start with your sound
card, based on your description.
I didn't say there was a conflict, though that's not a bad guess,
necessarily. I said you're getting too many interrupts. You probably
have something defective
But this same hardware has worked without this kind of problems until
now.
I set the startup sound to none to remove work from the sound card but
it made no difference.
Next step is disabling the SoundBlaster drivers to see if it makes any
difference.

If there really is a hardware problem, how could I find what is
causing it now? Should the problem appear in Windows as well?
Post by Scott G
regards, Scott
--
David Mediavilla
Scott G.
2007-09-04 19:02:25 UTC
Permalink
Raw Message
Next step is disabling the SoundBlaster drivers to see if it makes any difference.
I'd go further and also pull out the card, if possible.
If there really is a hardware problem, how could I find what is
causing it now? Should the problem appear in Windows as well?
It would be very hard to find out without a standalone dump, which is
something
with which I'm not set up anymore to help -- maybe someone else is. As
for Windoze, well, if Winxp works reliably, then it's probably an OS/2
driver problem rather than hardware (though not for sure). Not too much
you can do, then. On the whole, the multimedia drivers were among the
less stable ones, in my opinion.

If you haven't already done so, you should also download and run memtest.
good luck...
d***@sneakemail.com
2007-09-04 20:48:37 UTC
Permalink
Raw Message
Post by Scott G.
If you haven't already done so, you should also download and run memtest.
I gather that you mean http://www.memtest.org/#downiso . I'll try.

--
David Mediavilla
d***@sneakemail.com
2007-09-06 19:37:11 UTC
Permalink
Raw Message
Post by d***@sneakemail.com
Post by Scott G.
If you haven't already done so, you should also download and run memtest.
I gather that you mean http://www.memtest.org/#downiso . I'll try.
I have tried memtest and the computer passed all the tests at least
once.
Post by d***@sneakemail.com
David Mediavilla
--
David Mediavilla
Doug Bissett
2007-09-04 19:33:04 UTC
Permalink
Raw Message
Post by d***@sneakemail.com
I have been running eComStation 2.0 beta 2 milestone 2 (my
subscription expired) for some months, but lately I find a strange
When the WPS starts, sometimes it gets lock repeating a loop of the
startup sound while the pointer freezes with the clock cursor.
In one of the cases, I took a snapshot of the screen. Here you have
...snip...
Post by d***@sneakemail.com
Some idea on what is causing these errors?
Just a WAG, but I have had a similar problem, apparently caused by
NORMAN VIRUS CONTROL (although I am pretty sure that it simply
agravated the problem, and didn't cause the problem on it's own,),
that was cured by increasing the THREADS= number in CONFIG.SYS. Try
changing that to eiither 1536, or 2048.

BTW, I think you should still be able to get the eCS 2.0 RC2 update,
even if your subscription has run out.

Hope this helps...
--
From the eComStation 1.2 of Doug Bissett
dougb007 at telus dot net
(Please make the obvious changes, to e-mail me)
d***@sneakemail.com
2007-09-04 21:13:01 UTC
Permalink
Raw Message
Post by Doug Bissett
Post by d***@sneakemail.com
I have been running eComStation 2.0 beta 2 milestone 2 (my
subscription expired) for some months, but lately I find a strange
When the WPS starts, sometimes it gets lock repeating a loop of the
startup sound while the pointer freezes with the clock cursor.
Just a WAG, but I have had a similar problem, apparently caused by
NORMAN VIRUS CONTROL (although I am pretty sure that it simply
Not running it.
Post by Doug Bissett
that was cured by increasing the THREADS= number in CONFIG.SYS. Try
changing that to eiither 1536, or 2048.
I'll test that. Thanks.
Post by Doug Bissett
BTW, I think you should still be able to get the eCS 2.0 RC2 update,
even if your subscription has run out.
According to what I read, I will be able to download eCS 2.0 GA
(when?) but not the intermediate versions. Going to the download
section of Mensys, I see:
Expired, please go to: Your Upgrade Recommendations Software
Subscription Services for eComStation (valid for 12 months) (free
offer)
I intend to wait until I get GA before thinking on paying more, since
beta 3 wouldn't install here.
Post by Doug Bissett
From the eComStation 1.2 of Doug Bissett
dougb007 at telus dot net
--
David Mediavilla
d***@sneakemail.com
2007-09-06 19:42:54 UTC
Permalink
Raw Message
Post by Doug Bissett
Post by d***@sneakemail.com
I have been running eComStation 2.0 beta 2 milestone 2 (my
When the WPS starts, sometimes it gets lock repeating a loop of the
that was cured by increasing the THREADS= number in CONFIG.SYS. Try
changing that to eiither 1536, or 2048.
I removed the SoundBlaster drivers and set THREADS to 1536 and it
worked on next reboot.
I restored the SB drivers and kept the THREADS and it worked.
I am restoring the startup sound, but it seems that your THREADS
suggestions solves or at least palliates the problems.

Thanks a lot.
Post by Doug Bissett
From the eComStation 1.2 of Doug Bissett
dougb007 at telus dot net
--
David Mediavilla
Marcel Müller
2007-09-13 22:10:31 UTC
Permalink
Raw Message
Post by d***@sneakemail.com
Post by Doug Bissett
Post by d***@sneakemail.com
When the WPS starts, sometimes it gets lock repeating a loop of the
that was cured by increasing the THREADS= number in CONFIG.SYS. Try
changing that to eiither 1536, or 2048.
I removed the SoundBlaster drivers and set THREADS to 1536 and it
worked on next reboot.
I restored the SB drivers and kept the THREADS and it worked.
I am restoring the startup sound, but it seems that your THREADS
suggestions solves or at least palliates the problems.
It's not that likely that the THREADS setting solved your problem
causal. It is more likely that the setting prevents a Ring-0 data bomb
from hitting its target.

Using a high THREDS setting reserves more space for the kernel. Unused
space. This changes the memory location of almost everything. So invalid
memory accesses now hit somwhere else.

The multimedia subsystem has always been the week point of OS/2.


Marcel
Nobody
2007-09-15 15:49:12 UTC
Permalink
Raw Message
Post by d***@sneakemail.com
Post by Doug Bissett
Post by d***@sneakemail.com
I have been running eComStation 2.0 beta 2 milestone 2 (my
When the WPS starts, sometimes it gets lock repeating a loop of the
that was cured by increasing the THREADS= number in CONFIG.SYS. Try
changing that to eiither 1536, or 2048.
I removed the SoundBlaster drivers and set THREADS to 1536 and it
worked on next reboot.
I restored the SB drivers and kept the THREADS and it worked.
I am restoring the startup sound, but it seems that your THREADS
suggestions solves or at least palliates the problems.
Thanks a lot.
--
David Mediavilla
I tend to agree with Marcell's reponse. Your most likely culprits are
drivers, then perhaps a specific DLL file that is required for your
system, then system DLL files, and last an actual program file.

Loading...