SCO UNIX too many queued interupts
I have a SCO UNIX Open Server 5.0.7 system that had been running fine for years
however it started panicking with the error too many queued CPU interrupts.
after the 1st time it waited a couple months until it did it again, as time went on it got more frequent. now it is doing it a couple times a week.
I found a knowledge base article on the SCO web site that addressed it and said to adjust MAXACPUS, I have cranked that up to their max and still no good.
I am not an expert on SCO, just know enough to be dangerous. to the best of my knowledge the system is fully patched.
the system is an HP ML570 g4, 4 dual core CPU,s, 4gig memory, p600 controller card.
don't know now if I am looking at a software or hardware issue. if hardware what piece?
any ideas anyone?
thanks in advance
Tom
however it started panicking with the error too many queued CPU interrupts.
after the 1st time it waited a couple months until it did it again, as time went on it got more frequent. now it is doing it a couple times a week.
I found a knowledge base article on the SCO web site that addressed it and said to adjust MAXACPUS, I have cranked that up to their max and still no good.
I am not an expert on SCO, just know enough to be dangerous. to the best of my knowledge the system is fully patched.
the system is an HP ML570 g4, 4 dual core CPU,s, 4gig memory, p600 controller card.
don't know now if I am looking at a software or hardware issue. if hardware what piece?
any ideas anyone?
thanks in advance
Tom
Do more with
EXPERT OFFICE® is a registered trademark of EXPERTS EXCHANGE®
Wow... SCO... I feel like I'm in the Wayback Machine...
This situation rarely occurs anymore, because hardware runs so fast.
For SCO, likely you're running very old hardware too.
My guess is you have some sort of failing device, like a disk or disk controller.
With failing disks, interrupts can get thrown in excess.
You might try simply replacing your disk.
Tip: You can test this theory by installing SCO on a fresh disk + installing the fresh installed disk. If problem persists, then likely the disk controller board will have to be replaced.
Tip: Rather than guessing, refer to your logs + look for hardware failures or odd messages.
SCO systems are rare these days, so you'll have to dig around to find the correct log file locations.
Likely /var/log/* will be your starting point.
This situation rarely occurs anymore, because hardware runs so fast.
For SCO, likely you're running very old hardware too.
My guess is you have some sort of failing device, like a disk or disk controller.
With failing disks, interrupts can get thrown in excess.
You might try simply replacing your disk.
Tip: You can test this theory by installing SCO on a fresh disk + installing the fresh installed disk. If problem persists, then likely the disk controller board will have to be replaced.
Tip: Rather than guessing, refer to your logs + look for hardware failures or odd messages.
SCO systems are rare these days, so you'll have to dig around to find the correct log file locations.
Likely /var/log/* will be your starting point.
David
thanks for the feedback.
yes I know the hardware is a dozen years old but I need to keep it going for another year or so.
trying a new hard drive is not that simple as I am running a disk array that has 18 drives, and I don't have the drives, also if the system isn't running 24/7 I am notified as soon as it goes down to get it back ASAP.
anyway thanks for the advice and I will try to see what I can do about swapping some things
Tom
thanks for the feedback.
yes I know the hardware is a dozen years old but I need to keep it going for another year or so.
trying a new hard drive is not that simple as I am running a disk array that has 18 drives, and I don't have the drives, also if the system isn't running 24/7 I am notified as soon as it goes down to get it back ASAP.
anyway thanks for the advice and I will try to see what I can do about swapping some things
Tom
Hi,
Without knowing the application that you're running it is a bit tricky to give advise as it might cripple the system even more.
To get a better understanding about how SCO handles interrupts, read up here.
Cheers
Without knowing the application that you're running it is a bit tricky to give advise as it might cripple the system even more.
To get a better understanding about how SCO handles interrupts, read up here.
Cheers
Likely /var/log/* will provide you with only clues, as SCO... is, well... SCO...
So old + odd. Challenging to guess at what might be occurring.
Tip: You mentioned your running a disk array. This likely means you have some sort of custom i/o card in this machine, so best to setup a parallel RAID system today with same or greater space, then start nightly clones of your data...
Because if the problem is an i/o card about to die... you may be hard pressed to find a replacement card...
14TB drives can be had for <$500 now sometimes, so easy to replace an 18x disk array with smaller number of disks.
So old + odd. Challenging to guess at what might be occurring.
Tip: You mentioned your running a disk array. This likely means you have some sort of custom i/o card in this machine, so best to setup a parallel RAID system today with same or greater space, then start nightly clones of your data...
Because if the problem is an i/o card about to die... you may be hard pressed to find a replacement card...
14TB drives can be had for <$500 now sometimes, so easy to replace an 18x disk array with smaller number of disks.
David
I don't know that I cured the problem but at least it has now gone from 2-3 times a day to once a month
from your post from last fall I actually went on a step by step process and replace every bit of hardware, didn't cure it.
what did seem to make it manageable is that I had a windows app that accessed a shared directory on the Unix system (shared folders via Samba on the Unix system) every 5 minutes. that was part 1, actually the same app was running 3 instances and each instance accessed a different directory, again every 5 minutes all at the same time. due to the fact that the app access and the crash occurred at the same time I tried staggering the times that each instance accessed the Unix system and that finally got it to stop.
FYI - it would be nice to go to big drives but no SCO drivers available for a drive controller that can handle them on my old beast
Tom
I don't know that I cured the problem but at least it has now gone from 2-3 times a day to once a month
from your post from last fall I actually went on a step by step process and replace every bit of hardware, didn't cure it.
what did seem to make it manageable is that I had a windows app that accessed a shared directory on the Unix system (shared folders via Samba on the Unix system) every 5 minutes. that was part 1, actually the same app was running 3 instances and each instance accessed a different directory, again every 5 minutes all at the same time. due to the fact that the app access and the crash occurred at the same time I tried staggering the times that each instance accessed the Unix system and that finally got it to stop.
FYI - it would be nice to go to big drives but no SCO drivers available for a drive controller that can handle them on my old beast
Tom
Premium Content
You need an Expert Office subscription to comment.Start Free Trial