Messages in this thread | | | Date | Sat, 31 May 1997 10:24:12 -0500 (CDT) | From | "Mike A. Harris" <> | Subject | ISSUE: Possible NULL pointer kernel bug |
| |
[1.] Unable to handle kernel NULL pointer dereference at virtual address...
[2.] I just had my first system crash the other day and found an error log logged to my screen which I wrote down on paper. I'm not entirely sure how useful this information is but I figured that it must be serious enough to warrant a bug report posting.
I was connected to the net via dialup PPP and also using X at the time. My system load average was VERY high, and my swap file thrashing madly. This is fairly normal for me though. I had tload running on a spare VC and I switched to it for a while. Even thought the system had died down and stopped thrashing, the load kept going up and up and up. I switched to X and closed it down. When I got back to console mode, every console responded strangely. I could run certain commands, but then after a couple commands, the console would die and freeze - not returning from the command. A ps listing showed several 'D' dead processes - which I've NEVER seen before on my system. I began to be concerned and tried to log in on another VC but it wouldn't let me log in. I then noticed that "init" had died! Many other processes were also listed as dead, especially the ones that I had most recently executed. That explained part of the problem, but not why. The syslog entries show that this occured at approximately 8:00am, but I hadn't noticed anything strange until at least a half hour later.
At that point I had discovered the error message logged to tty8 and wrote it down. I watched the HD light and exactly once a minute the light came on thus showing that syncs were still happening. I waited for 5-10 seconds after a sync and hit reset as I couldn't do anything else. When the system came back up, fsck reported errors and suggested that I run it manually. I did so from a boot floppy and my system is back to normal now AFAIK.
I am somewhat of a novice in these matters and was unsure of what to do with the information. Someone suggested posting to this mailing list, and so I subscribed. I read the FAQ that was sent back from the majordomo and the appendixes. I have tried as much as possible to comply with the requests set out in that document. Please take it easy on me as I'm only trying to help.
If more info is needed, please contact me.
[3.] NULL, pointer, null, kernel, dereference, kerneld
[4.] Linux version 2.0.30 (root@capslock) (gcc version 2.7.2) #1 Thu Apr 17 22:34:52 EDT 1997
[5.]
Ouch, no kerneld for message 2147418149 Unable to handle kernel NULL pointer dereference at virtual address C0000018
May 29 08:00:49 capslock kernel: Unable to handle kernel NULL pointer dereference at virtual address c0000018 May 29 08:00:49 capslock kernel: current->tss.cr3 = 00b3d000, |r3 = 00b3d000 May 29 08:00:49 capslock kernel: *pde = 00102067 May 29 08:00:49 capslock kernel: *pte = 00000027 May 29 08:00:49 capslock kernel: Oops: 0000 May 29 08:00:49 capslock kernel: CPU: 0 May 29 08:00:49 capslock kernel: EIP: 0010:[find_candidate+212/244] May 29 08:00:49 capslock kernel: EFLAGS: 00010206 May 29 08:00:49 capslock kernel: eax: 00000000 ebx: 0027d318 ecx: 00000000 edx: 00207a84 May 29 08:00:49 capslock kernel: esi: 00000000 edi: 00b3fee4 ebp: 00000400 esp: 00b3feac May 29 08:00:49 capslock kernel: ds: 0018 es: 0018 fs: 002b gs: 002b ss: 0018 May 29 08:00:49 capslock kernel: Process update (pid: 8, process nr: 8, stackpage=00b3f000) May 29 08:00:49 capslock kernel: Stack: 0027d318 00000000 00000400 00005400 00000000 00123890 0027d318 00b3fee4 May 29 08:00:49 capslock kernel: 00000400 0000e005 00580303 00000001 00000400 00000009 00000014 00000000 May 29 08:00:49 capslock kernel: 00000000 0027d318 00000000 00000000 00123a5e 00000400 0000e005 00583ec4 May 29 08:00:49 capslock kernel: Call Trace: [refill_freelist+836/948] [getblk+350/936] [bread+24/124] [ext2_update_inode+274/704] [ext2_write_inode+12/16] [write_inode+91/116] [sync_inodes+63/88] May 29 08:00:49 capslock kernel: [sync_old_buffers+20/224] [sys_bdflush+53/152] [system_call+82/128] May 29 08:00:49 capslock kernel: Code: 8b 40 18 89 44 24 10 ff 0f 83 7c 24 10 00 0f 85 38 ff ff ff
[6.] Nothing.
[7.] Nothing.
[7.1] -- Versions installed: (if some fields are empty or looks -- unusual then possibly you have very old versions) Linux capslock 2.0.30 #1 Thu Apr 17 22:34:52 EDT 1997 i486 Kernel modules 2.0.0 Gnu C 2.7.2 Binutils 2.6.0.14 Linux C Library 5.3.12 Dynamic Linker (ld.so) 1.7.14 Linux C++ Library 27.1.4 Procps 0.99 Mount 2.5j Net-tools 1.2.0 Kbd 0.89 Sh-utils 1.12
[7.2] This is an AMD 486-DX2-66 running with 8megs and no external cache. It is an ACER motherboard.
3 capslock:~# cat /proc/cpuinfo processor : 0 cpu : 486 model : unknown vendor_id : unknown stepping : unknown fdiv_bug : no hlt_bug : no fpu : yes fpu_exception : yes cpuid : no wp : yes flags : bogomips : 33.18
[7.3]
3 capslock:~# cat /proc/modules ppp 5 1 (autoclean) slhc 2 [ppp] 1 (autoclean) binfmt_aout 1 0 (autoclean) plip 3 1 ip_masq_irc 1 0 ip_masq_raudio 1 0 ip_masq_ftp 1 0
In addition, I have IDE CDROM support and iso-9660 as modules.
[7.4] Nothing.
[7.5] Other possibly relevant information:
3 capslock:/proc# cat filesystems ext2 umsdos msdos nodev proc iso9660
3 capslock:/proc# cat devices Character devices: 1 mem 2 pty 3 ttyp 4 ttyp 5 cua 7 vcs 14 sound
Block devices: 2 fd 3 ide0 22 ide1
[X]
Please find attached a gzipped copy of /proc/ksyms I figured it might come in handy.
[unhandled content-type:application/octet-stream] | |