I've had my machine hanging on shutdown from KDE with a blank screen (mouse cursor still visible) and dead mouse and keyboard, right at the point where you'd normally be switched back to the console. Before the keyboard went out there was a bit of disk activity that I don't remember at that point in the shutdown sequence. Hard reset would then end up clearing four or five inodes on the root partition. By some luck I managed to switch to the console once right before the hang and was then greeted by a "soft lockup" message from the kernel that was repeating every 22 seconds or so (the kernel seemed to be stuck in some page fault handler and a second routine that I have no clue about). This is only happening after I've used KDE for quite a while, so if I just log into KDE, start all my applications and then do an immediate shutdown it never happens, so it is quite hard to drill to the bottom of this.
I've done some more experimentations and found that this is only happening with the 3.17.1 kernels, but not with the 3.16.3 which I've now rolled back to. I'm AFK for the next few days, but maybe someone else might have an idea what's going on.
Regards, Achim.
Achim Gratz writes:
I've done some more experimentations and found that this is only happening with the 3.17.1 kernels, but not with the 3.16.3 which I've now rolled back to. I'm AFK for the next few days, but maybe someone else might have an idea what's going on.
The latest kernel in the old Tumbleweed (3.17.1-53.gd944251-pae) has hung three times out of five days of working with it. I managed to break into the console once and it still seems to be stuck in some page fault handler. That was from a QThread with 1244, which - judging from todays boot - could probably have been one of the kdeinit processes. Both of the two times it didn't hang on shutdown I had suspended the machine to RAM during the day. I guess there isn't mauch more to check before I will roll up to the new Tumbleweed on the weekend (if time permits).
Regards, Achim.
Am 05.11.2014 20:34, schrieb Achim Gratz:
Achim Gratz writes:
I've done some more experimentations and found that this is only happening with the 3.17.1 kernels, but not with the 3.16.3 which I've now rolled back to. I'm AFK for the next few days, but maybe someone else might have an idea what's going on.
The latest kernel in the old Tumbleweed (3.17.1-53.gd944251-pae) has hung three times out of five days of working with it. I managed to break into the console once and it still seems to be stuck in some page fault handler. That was from a QThread with 1244, which - judging from todays boot - could probably have been one of the kdeinit processes. Both of the two times it didn't hang on shutdown I had suspended the machine to RAM during the day. I guess there isn't mauch more to check before I will roll up to the new Tumbleweed on the weekend (if time permits).
I'm having huge problems with that kernel on my work station too - while it works fine on my laptop ;( But I also have too little data to even report a bug.
Greetings, Stephan
Stephan Kulow writes:
I'm having huge problems with that kernel on my work station too - while it works fine on my laptop ;(
Not what I wanted to hear, but interesting. I've tried something different yesterday: just logging out of my KDE session also hung up the kernel; this time it was in some QTInotifyFile (or some similar name) and again a page fault handler.
But I also have too little data to even report a bug.
--8<---------------cut here---------------start------------->8---
cat /proc/cpuinfo
processor : 0 vendor_id : AuthenticAMD cpu family : 15 model : 12 model name : AMD Athlon(tm) 64 Processor 3200+ stepping : 0 microcode : 0x3a cpu MHz : 1000.000 cache size : 512 KB physical id : 0 siblings : 1 core id : 0 cpu cores : 1 apicid : 0 initial apicid : 0 fdiv_bug : no f00f_bug : no coma_bug : no fpu : yes fpu_exception : yes cpuid level : 1 wp : yes flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 syscall nx mmxext lm 3dnowext 3dnow bogomips : 2002.58 clflush size : 64 cache_alignment : 64 address sizes : 40 bits physical, 48 bits virtual power management: ts fid vid ttp --8<---------------cut here---------------end--------------->8---
Regards, Achim.
Achim Gratz writes:
The latest kernel in the old Tumbleweed (3.17.1-53.gd944251-pae) has hung three times out of five days of working with it.
The 3.17.1 kernel from the "new" Tumbleweed has the same problem. I've locked out that version and downgraded to 3.16.6 from the 13.2 repo for now.
Regards, Achim.
Achim Gratz writes:
I've had my machine hanging on shutdown from KDE with a blank screen (mouse cursor still visible) and dead mouse and keyboard, right at the point where you'd normally be switched back to the console. Before the keyboard went out there was a bit of disk activity that I don't remember at that point in the shutdown sequence. Hard reset would then end up clearing four or five inodes on the root partition. By some luck I managed to switch to the console once right before the hang and was then greeted by a "soft lockup" message from the kernel that was repeating every 22 seconds or so (the kernel seemed to be stuck in some page fault handler and a second routine that I have no clue about). This is only happening after I've used KDE for quite a while, so if I just log into KDE, start all my applications and then do an immediate shutdown it never happens, so it is quite hard to drill to the bottom of this.
I've done some more experimentations and found that this is only happening with the 3.17.1 kernels, but not with the 3.16.3 which I've now rolled back to. I'm AFK for the next few days, but maybe someone else might have an idea what's going on.
The issue is still present with the 3.17.2 kernel. It does not happen on every shutdown, as before.
Regards, Achim.
-
Achim Gratz -
Stephan Kulow