PROBLEM: Process killing, kflushd problems

Jacob Delafon (dunkahn@mail.iname.com)
Wed, 19 May 1999 23:16:12 +0200 (CEST)


[1.] Process killing, kflushd problems ?

[2.] I get plenty of process killed by the kernel while I "make", "tar
xvfz" or other things CPU eaters, I get sig11 or sig4 for example, but I
also get many Segmentation Fault and also many core dumps.
I also have a problem with kflushd. Sometimes it gets totaly crazy and
uses about 96% (used top) of CPU when I mv or cp a file between
partitions. So that it takes about 1min to mv a 1Mo file ! So I have to
reboot each time he acts like that because it never stops. This is very
unconvenient.
Tonight I lost 2 CDR (this is why I'm a little angry...)because of
buffer underuns. I never had problem like that on my linuxbox, and I
wasn't running big cpu eaters processes nor using the partition the files
were on, I was just connected on the net and I used bitchx only. So my
buffer was at 6% full (!!) whereas it was always full (98%) every time I
burned a CD before. I had a 2.0.36 kernel before, and I had a 6days
uptime withtout any of those problems, and I hadn't got so much process
killed.

[3.] kernel, fs

[4.] Linux version 2.2.7 (root@canadou.local) (gcc version 2.7.2.1) #11
Sun May 2 14:46:56 CEST 1999

[5.]
One of many Oops :
<1>Unable to handle kernel paging request at virtual address cacad788
<1>current->tss.cr3 = 0081d000, %cr3 = 0081d000
<1>*pde = 00000000
<4>Oops: 0000
<4>CPU: 0
<4>EIP: 0010:[<c0138426>]
<4>EFLAGS: 00010202
<4>eax: cacad788 ebx: 00000008 ecx: cacad788 edx: c0d87740
<4>esi: 00000000 edi: 000000ee ebp: c0b4c110 esp: c0773e6c
<4>ds: 0018 es: 0018 ss: 0018
<4>Process tar (pid: 1982, process nr: 48, stackpage=c0773000)
<4>Stack: 000000ee c0b4c110 00000002 cacad788 c0138728 c0b4c110 c0d87740
000000e2
<4> 00000001 00000400 000000ee c0773f14 c0b4c110 0000000c 00000001
000000ee
<4> c0773f14 00000400 000000ee c0faec00 c0b4c110 000000ee 00010000
00000010
<4>Call Trace: [<c0138728>] [<c0136b7a>] [<c0136990>] [<c1c410d8>]
[<c0108d01>] [<c1c2949c>] [<c010ef3b>]
<4> [<c010ef8f>] [<c0128a96>] [<c01229a2>] [<c0122a84>] [<c0107a54>]
<4>Code: 8b 19 85 db 74 40 8b 4c 24 2c 51 53 8b 44 24 28 0f b7 40 0c

[6.]

[7.]
[7.1.]
Linux canadou.local 2.2.7 #11 Sun May 2 14:46:56 CEST 1999 i586 unknown
Kernel modules 2.1.121
Gnu C 2.7.2.1
Binutils 2.7.0.9
Linux C Library 5.4.33
Dynamic linker ldd: version 1.8.10
Linux C++ Library 27.2.1
Linux C++ Library 27.2.1
Procps 1.11
Mount 2.9g
Net-tools (1998-06-29)
Kbd 0.89
Sh-utils 1.16
Modules Loaded loop sb uart401 sound soundlow ppp_deflate dummy
bsd_comp
[7.2.]
processor : 0
vendor_id : GenuineIntel
cpu family : 5
model : 2
model name : Pentium 75 - 200
stepping : 12
cpu MHz : 199.435192
fdiv_bug : no
hlt_bug : no
sep_bug : no
f00f_bug : yes
fpu : yes
fpu_exception : yes
cpuid level : 1
wp : yes
flags : fpu vme de pse tsc msr mce cx8
bogomips : 79.67
[7.3.]
loop 7280 2 (autoclean)
sb 31532 1
uart401 5628 1 [sb]
sound 55144 0 [sb uart401]
soundlow 240 0 [sound]
ppp_deflate 38504 0
dummy 720 0 (unused)
bsd_comp 3456 0
[7.4.]
Attached devices:
Host: scsi0 Channel: 00 Id: 00 Lun: 00
Vendor: MATSHITA Model: CD-R CW-7502 Rev: 3.17
Type: CD-ROM ANSI SCSI revision: 02
[7.5.]
P166 overcloaked to 200 (it was also overcloaked when I was in 2.0.*
kernels)
16Mo of EDO
40Mo of Swap

[8.]

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/