Jimmy, we're seeing a couple odd problems on suda. Something that may or may not be helped by a kernel upgrade (maybe? who knows) and what appears to be a corrupt page in the database (maybe? who knows).
We *need* a slaved server *now* in case suda goes south unrecoverably.
Is gunther available for this (if so on what IP?), or should I strip down one of the other machines?
-- brion vibber (brion @ pobox.com)
On Feb 22, 2004, at 16:00, Brion Vibber wrote:
We *need* a slaved server *now* in case suda goes south unrecoverably.
Is gunther available for this (if so on what IP?), or should I strip down one of the other machines?
For now I'm taking isidore out of apache rotation.
-- brion vibber (brion @ pobox.com)
Brion Vibber brion-at-pobox.com |wikipedia| wrote: <>> Jimmy, we're seeing a couple odd problems on suda. Something that may or may not be helped by a kernel upgrade (maybe? who
knows) and what appears to be a corrupt page in the database (maybe?
who knows).
Just to let the list know, in case someone has any information: this was in the dmesg log around the time of the problem. We don't know if it's related to this problem, or if it's just a coincidence:
// E23
kernel BUG at page_alloc.c:139! invalid operand: 0000 soundcore nfs lockd sunrpc ide-tape ide-cd cdrom ipt_LOG parport_pc lp parport autofs tg3 ipt_REJECT iptable_filter ip_tables
keybdev mousedev hid input usb-o CPU: 0 EIP: 0060:[<c0146b1b>] Not tainted EFLAGS: 00010282
EIP is at __free_pages_ok [kernel] 0xeb (2.4.20-8smp) eax: 00000018 ebx: c1030d90 ecx: c1000030 edx: d3a5ec40 esi: 00000000 edi: 00000000 ebp: 00000000 esp: ef1c9bb8 ds: 0068 es: 0068 ss: 0068 Process docount.sh (pid: 11603, stackpage=ef1c9000) Stack: 00015de9 c1030d90 c0406ee0 c0148560 c0406ee0 00000000 00000001 c1030d90 00000163 c1030d90 c1030d90 00000163 00000000 fffee0b8 c01351b4 c1030d90 00df4067 c0137be6 f6b5dbc0 40000000 fffee0b8 e5d113b4 00000001 c013aff6 Call Trace: [<c0148560>] remove_exclusive_swap_page [kernel] 0xb8 (0xef1c9bc4)) [<c01351b4>] __free_pte [kernel] 0x4c (0xef1c9bf0)) [<c0137be6>] zap_pte_range [kernel] 0x22e (0xef1c9bfc)) [<c013aff6>] do_generic_file_read [kernel] 0x26a (0xef1c9c14)) [<c013596b>] zap_page_range [kernel] 0x107 (0xef1c9c28)) [<c0139320>] exit_mmap [kernel] 0xc4 (0xef1c9c6c)) [<c013c0f0>] generic_file_mmap [kernel] 0x0 (0xef1c9c88)) [<c0159d5d>] exec_mmap [kernel] 0x1fd (0xef1c9c90)) [<c013c0f0>] generic_file_mmap [kernel] 0x0 (0xef1c9ca4)) [<c0159d84>] flush_old_exec [kernel] 0x14 (0xef1c9cac)) [<c0174945>] load_elf_binary [kernel] 0x2bd (0xef1c9cc0)) [<f88680d7>] ext3_do_update_inode [ext3] 0x173 (0xef1c9ce4)) [<f8852fd8>] journal_get_write_access_Rsmp_dae0236a [jbd] 0x5c (0xef1c9d04)) [<f8853b4e>] journal_stop_Rsmp_ab07e5bf [jbd] 0x17e (0xef1c9d58)) [<c0143941>] reclaim_page [kernel] 0x231 (0xef1c9d78)) [<c01471c9>] __alloc_pages_limit [kernel] 0x85 (0xef1c9d9c)) [<c0174688>] load_elf_binary [kernel] 0x0 (0xef1c9df0)) [<c015a454>] search_binary_handler [kernel] 0x128 (0xef1c9dfc)) [<c015a66b>] do_execve [kernel] 0x17b (0xef1c9e44)) [<c0107c80>] sys_execve [kernel] 0x54 (0xef1c9fa4)) [<c010971b>] system_call [kernel] 0x33 (0xef1c9fc0))
wikitech-l@lists.wikimedia.org