| 1 |
Hi all, does any one can help me? |
| 2 |
Last night apache and mysql just stoped - like freeze. ps ax showed |
| 3 |
all processes of apache as <defunct>. dmesg showed this: |
| 4 |
Jul 29 13:52:18 foo Unable to handle kernel NULL pointer dereference |
| 5 |
at virtual address 00000044 |
| 6 |
Jul 29 13:52:18 foo printing eip: |
| 7 |
Jul 29 13:52:18 foo c030eb90 |
| 8 |
Jul 29 13:52:18 foo *pde = 00000000 |
| 9 |
Jul 29 13:52:18 foo Oops: 0002 [#1] |
| 10 |
Jul 29 13:52:18 foo SMP |
| 11 |
Jul 29 13:52:18 foo Modules linked in: ipt_LOG ipt_conntrack ipt_limit |
| 12 |
iptable_mangle iptable_nat ip_conntrack iptable_filter ip_tables |
| 13 |
floppy dm_mirror dm_mod |
| 14 |
parport_pc parport e100 mii e1000 evdev |
| 15 |
Jul 29 13:52:18 foo CPU: 0 |
| 16 |
Jul 29 13:52:18 foo EIP: 0060:[<c030eb90>] Not tainted VLI |
| 17 |
Jul 29 13:52:18 foo EFLAGS: 00010206 (2.6.12-capsel) |
| 18 |
Jul 29 13:52:18 foo EIP is at _spin_lock+0x0/0x10 |
| 19 |
Jul 29 13:52:18 foo eax: 00000044 ebx: f197b840 ecx: f197b5a0 edx: f197b7 |
| 20 |
4c |
| 21 |
Jul 29 13:52:18 foo esi: f197b74c edi: 00000044 ebp: 00000001 esp: f7cc3e |
| 22 |
94 |
| 23 |
Jul 29 13:52:18 foo ds: 007b es: 007b ss: 0068 |
| 24 |
Jul 29 13:52:18 foo Process kswapd0 (pid: 153, threadinfo=f7cc2000 task=c22fba4 |
| 25 |
0) |
| 26 |
Jul 29 13:52:18 foo Stack: c015dca6 f197b74c f197b754 f197b74c |
| 27 |
00000006 f7cc3ec0 c017671b f197b74c |
| 28 |
Jul 29 13:52:18 foo cdf04b40 00000000 00000006 f197b908 f197be24 |
| 29 |
00000087 00000000 00000000 |
| 30 |
Jul 29 13:52:18 foo c211ea20 c01767ef 00000080 c0146fc5 00000080 |
| 31 |
000000d0 000274b8 002e1800 |
| 32 |
Jul 29 13:52:18 foo Call Trace: |
| 33 |
Jul 29 13:52:18 foo [<c015dca6>] remove_inode_buffers+0x56/0xa0 |
| 34 |
Jul 29 13:52:18 foo [<c017671b>] prune_icache+0x13b/0x1f0 |
| 35 |
Jul 29 13:52:18 foo [<c01767ef>] shrink_icache_memory+0x1f/0x50 |
| 36 |
Jul 29 13:52:18 foo [<c0146fc5>] shrink_slab+0x155/0x190 |
| 37 |
Jul 29 13:52:18 foo [<c014840f>] balance_pgdat+0x2ff/0x3d0 |
| 38 |
Jul 29 13:52:18 foo [<c01485c9>] kswapd+0xe9/0x120 |
| 39 |
Jul 29 13:52:18 foo [<c01303c0>] autoremove_wake_function+0x0/0x60 |
| 40 |
Jul 29 13:52:18 foo [<c0102d12>] ret_from_fork+0x6/0x14 |
| 41 |
Jul 29 13:52:18 foo [<c01303c0>] autoremove_wake_function+0x0/0x60 |
| 42 |
Jul 29 13:52:18 foo [<c01484e0>] kswapd+0x0/0x120 |
| 43 |
Jul 29 13:52:18 foo [<c0101015>] kernel_thread_helper+0x5/0x10 |
| 44 |
Jul 29 13:52:18 foo Code: 00 00 01 74 05 e8 21 e3 ff ff c3 ba 00 e0 ff |
| 45 |
ff 21 e2 81 42 14 00 01 00 00 f0 81 28 00 00 00 01 74 05 e8 04 e3 ff |
| 46 |
ff c3 8d 76 00 <f0> fe 08 79 09 f3 90 80 38 00 7e f9 eb f2 c3 90 f0 81 |
| 47 |
28 00 00 |
| 48 |
|
| 49 |
After reboot many databases of mysql had to be fixed. After the |
| 50 |
databases were fixed I tried to rsync all of them to separate disk. |
| 51 |
After few seconds rsync freezed, so I killed it and started one more |
| 52 |
time, and it freezed to. After unmounting filesystem I checked it with |
| 53 |
fsck.reiserfs but there was no errors. Now everything works on |
| 54 |
2.4.31-hardened but works terribly slow. Previous kernel was |
| 55 |
2.6.12-gentoo-r4. |
| 56 |
|
| 57 |
File system for mysql is reiserfs on top of raid0. raid0 was built on |
| 58 |
two IDE disks about 200GB each. |
| 59 |
|
| 60 |
Is it a bug in kernel? Which kernel should I use then? Is there any |
| 61 |
other fix for this? |
| 62 |
Few days before everything including ssh froze and only hard reset helped. |
| 63 |
|
| 64 |
And next question: |
| 65 |
Timeout setting on mysql now works. They didn't with 2.6 kernel. Is |
| 66 |
there solution for this? |
| 67 |
|
| 68 |
-- |
| 69 |
gentoo-server@g.o mailing list |
Archives