jay@alliant.backbone.uoknor.edu (Jay Liang) (03/07/90)
I suppose if you are running Concentrix 5.5, then the next system panic message ought to be quite familiar: fbuserr: kernel non-zero ipl. What exactly does it mean? What cause it? I have seen this error at least 10 times since we installed 5.5 in Januray. It happened twice in the last 24 hours, as a matter of fact. :-( Any ideas as to preventing it from happening? Jay Liang (jay@alliant.backbone.uoknor.edu) Geosciences Computing Network University of Oklahoma -- /\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\ \ Jay Liang Geosciences Computing Network / / jay@alliant.backbone.uoknor.edu University of Oklahoma \ \/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/
emcguire@cadfx.ccad.uiowa.edu (Ed McGuire) (03/12/90)
From article <1990Mar6.222236.7098@uokmax.uucp>, by jay@alliant.backbone.uoknor.edu (Jay Liang): > I suppose if you are running Concentrix 5.5, then the next system panic > message ought to be quite familiar: fbuserr: kernel non-zero ipl. What > exactly does it mean? What cause it? I have seen this error at least 10 > times since we installed 5.5 in Januray. It happened twice in the last 24 > hours, as a matter of fact. :-( Any ideas as to preventing it from > happening? All the information below came from Customer Service. There's a bug in 5.5.0. I don't know the specific nature of the bug. Customer Service told me that we can press [Esc] [p] at the console to proceed past the panic. They were very careful to warn us that the ONLY panic message that this workaround applies to is `fbuserr: kernel non-zero ipl'. Proceeding past the panic leaves your system up as if the panic never occurred. THey also asked us to call each time the problem occurred. Under rare circumstances a second different panic can occur immediately thereafter if a crucial page is not in memory. Then you need to reboot. Note that it is a different panic message. If the `fbuserr: kernel non-zero ipl' appears again you can [Esc] [p] again. This happened to me a couple of times. Also, if your system is hung up and there is nobody there to proceed past the panic right away, a second panic occurs when you finally do proceed when the system notices that the wall clock was not updated for a long time. We didn't know the consequences of proceeding past this one so we rebooted. Customer Service can patch 5.5.0 to avoid the problem. They will provide the patch only if you tell them that you're encountering the problem at least once a week. The patch is a simple branch past the panic test. So it's just like pressing [Esc] [p] at the console, and the problem is not really fixed. That means you could still get the rare second different panic too. A maintenance release with the problem corrected is available to customers with this problem. Customer Service put the patch in and promised to ship the maintenance release to us when it is ready. They said they were delying the release for a bit longer until some bugs are worked out of it. peace. -- Ed From vn Mon Mar 12 09:38:30 1990 Subject: Re: Pnews test and non-zero ipl Newsgroups: jay@alliant.backbone.uoknor.edu References: <1990Mar6.222236.7098@uokmax.uucp> From article <1990Mar6.222236.7098@uokmax.uucp>, by jay@alliant.backbone.uoknor.edu (Jay Liang): > I suppose if you are running Concentrix 5.5, then the next system panic > message ought to be quite familiar: fbuserr: kernel non-zero ipl. What > exactly does it mean? What cause it? I have seen this error at least 10 > times since we installed 5.5 in Januray. It happened twice in the last 24 > hours, as a matter of fact. :-( Any ideas as to preventing it from > happening? All the information below came from Customer Service. There's a bug in 5.5.0. I don't know the specific nature of the bug. Customer Service told me that we can press [Esc] [p] at the console to proceed past the panic. They were very careful to warn us that the ONLY panic message that this workaround applies to is `fbuserr: kernel non-zero ipl'. Proceeding past the panic leaves your system up as if the panic never occurred. THey also asked us to call each time the problem occurred. Under rare circumstances a second different panic can occur immediately thereafter if a crucial page is not in memory. Then you need to reboot. Note that it is a different panic message. If the `fbuserr: kernel non-zero ipl' appears again you can [Esc] [p] again. This happened to me a couple of times. Also, if your system is hung up and there is nobody there to proceed past the panic right away, a second panic occurs when you finally do proceed when the system notices that the wall clock was not updated for a long time. We didn't know the consequences of proceeding past this one so we rebooted. Customer Service can patch 5.5.0 to avoid the problem. They will provide the patch only if you tell them that you're encountering the problem at least once a week. The patch is a simple branch past the panic test. So it's just like pressing [Esc] [p] at the console, and the problem is not really fixed. That means you could still get the rare second different panic too. A maintenance release with the problem corrected is available to customers with this problem. Customer Service put the patch in and promised to ship the maintenance release to us when it is ready. They said they were delying the release for a bit longer until some bugs are worked out of it. peace. -- Ed