High FS Activity Crash
A message that is not in the archive (sent off-list or not yet received): [email protected]
A message that is not in the archive (sent off-list or not yet received): [email protected]
A message that is not in the archive (sent off-list or not yet received): [email protected]
A message that is not in the archive (sent off-list or not yet received): [email protected]
Ok, on my fileserver I'm running Redhat 7.2 with Redhat kernel 2.4.7-10 and when I have a large number of people downloading from me at one time, Samba, ProFTPD, Apache, and SSH all stop responding, and I get the following in my log, the only thing special I could think about my setup is I have 3 NIC's bonded together to be load balanced on the same IP, any ideas?
Jan 6 00:42:39 anubis kernel: Unable to handle kernel NULL pointer dereference at virtual address 00000024 Jan 6 00:42:39 anubis kernel: printing eip: Jan 6 00:42:39 anubis kernel: c0135cb0 Jan 6 00:42:39 anubis kernel: *pde = 00000000 Jan 6 00:42:39 anubis kernel: Oops: 0002 Jan 6 00:42:39 anubis kernel: CPU: 0 Jan 6 00:42:39 anubis kernel: EIP: 0010:[__remove_from_lru_list+32/112] Jan 6 00:42:39 anubis kernel: EIP: 0010:[<c0135cb0>] Jan 6 00:42:39 anubis kernel: EFLAGS: 00010286 Jan 6 00:42:39 anubis kernel: eax: 00000000 ebx: ce310cc0 ecx: ce310cc0 edx: c9b708c0 Jan 6 00:42:39 anubis kernel: esi: ce310cc0 edi: ce310cc0 ebp: 00000000 esp: c1427f28 Jan 6 00:42:39 anubis kernel: ds: 0018 es: 0018 ss: 0018 Jan 6 00:42:39 anubis kernel: Process kswapd (pid: 5, stackpage=c1427000) Jan 6 00:42:39 anubis kernel: Stack: c0135d8d ce310cc0 00000000 c0138890 ce310cc0 00000003 c1690800 c13f4a64 Jan 6 00:42:39 anubis kernel: 00000080 c012cc6a 00000000 c13f4a64 00000000 00000007 c012cfe3 c13f4a64 Jan 6 00:42:39 anubis kernel: 00000080 00000000 00000000 0000000d 000093b6 00000000 c0243600 00000000 Jan 6 00:42:39 anubis kernel: Call Trace: [__remove_from_queues+45/48] [try_to_free_buffers+112/272] [try_to_release_page+58/96] [do_page_launder+851/2112] [page_launder+40/96] Jan 6 00:42:39 anubis kernel: Call Trace: [<c0135d8d>] [<c0138890>] [<c012cc6a>] [<c012cfe3>] [<c012d4f8>] Jan 6 00:42:39 anubis kernel: [do_try_to_free_pages+18/96] [kswapd+94/272] [_stext+0/48] [_stext+0/48] [kernel_thread+38/48] [kswapd+0/272] Jan 6 00:42:39 anubis kernel: [<c012d922>] [<c012d9ce>] [<c0105000>] [<c0105000>] [<c0105716>] [<c012d970>] Jan 6 00:42:39 anubis kernel: Jan 6 00:42:39 anubis kernel: Code: 89 50 24 8b 44 24 08 8d 14 85 00 00
00 00 39 8a 6c 21 2b c0
Signature · 3 lines
Adam Voigt ([email protected]) The Cryptocomm Group My GPG Key: http://64.238.252.49:8080/adam_at_cryptocomm.asc
If the disk is churning and churning, that could be why. Upgrade to a kernel that has a low latency elevator that makes sure disk thrashing doesn't kill system performance.
I use 2.4.20-ck2 on my server :)
Regards
Josh
Rabid cheeseburgers forced Adam Voigt<[email protected]> to write this on 06 Jan 2003 09:05:40-0500:
Quoted 50 lines
Ok, on my fileserver I'm running Redhat 7.2 with Redhat kernel 2.4.7-10 and when I have a large number of people downloading from me at one time, Samba, ProFTPD, Apache, and SSH all stop responding, and I get the following in my log, the only thing special I could think about my setup is I have 3 NIC's bonded together to be load balanced on the same IP, any ideas?
Jan 6 00:42:39 anubis kernel: Unable to handle kernel NULL pointer dereference at virtual address 00000024 Jan 6 00:42:39 anubis kernel: printing eip: Jan 6 00:42:39 anubis kernel: c0135cb0 Jan 6 00:42:39 anubis kernel: *pde = 00000000 Jan 6 00:42:39 anubis kernel: Oops: 0002 Jan 6 00:42:39 anubis kernel: CPU: 0 Jan 6 00:42:39 anubis kernel: EIP: 0010:[__remove_from_lru_list+32/112] Jan 6 00:42:39 anubis kernel: EIP: 0010:[<c0135cb0>] Jan 6 00:42:39 anubis kernel: EFLAGS: 00010286 Jan 6 00:42:39 anubis kernel: eax: 00000000 ebx: ce310cc0 ecx: ce310cc0 edx: c9b708c0 Jan 6 00:42:39 anubis kernel: esi: ce310cc0 edi: ce310cc0 ebp: 00000000 esp: c1427f28 Jan 6 00:42:39 anubis kernel: ds: 0018 es: 0018 ss: 0018 Jan 6 00:42:39 anubis kernel: Process kswapd (pid: 5, stackpage=c1427000) Jan 6 00:42:39 anubis kernel: Stack: c0135d8d ce310cc0 00000000 c0138890 ce310cc0 00000003 c1690800 c13f4a64 Jan 6 00:42:39 anubis kernel: 00000080 c012cc6a 00000000 c13f4a64 00000000 00000007 c012cfe3 c13f4a64 Jan 6 00:42:39 anubis kernel: 00000080 00000000 00000000 0000000d 000093b6 00000000 c0243600 00000000 Jan 6 00:42:39 anubis kernel: Call Trace:[__remove_from_queues+45/48][try_to_free_buffers+112/272] [try_to_release_page+58/96][do_page_launder+851/2112] [page_launder+40/96] Jan 6 00:42:39 anubis kernel: Call Trace:
[<c0135d8d>] [<c0138890>][<c012cc6a>] [<c012cfe3>] [<c012d4f8>]Jan 6 00:42:39 anubis kernel: [do_try_to_free_pages+18/96] [kswapd+94/272] [_stext+0/48] [_stext+0/48] [kernel_thread+38/48] [kswapd+0/272] Jan 6 00:42:39 anubis kernel: [<c012d922>] [<c012d9ce>] [<c0105000>][<c0105000>] [<c0105716>] [<c012d970>] Jan 6 00:42:39 anubis kernel: Jan 6 00:42:39 anubis kernel: Code: 89 50 24 8b 44 24 08 8d 14 85 0000 00 00 39 8a 6c 21 2b c0
Signature · 3 lines
Adam Voigt ([email protected]) The Cryptocomm Group My GPG Key: http://64.238.252.49:8080/adam_at_cryptocomm.asc
Signature · 6 lines
Joshua Kwan [email protected] pgp public key at http://joshk.mspencer.net/pubkey_gpg.asc It's hard to keep your shirt on when you're getting something off your chest.
Quoted 6 lines · Adam Voigt
Ok, on my fileserver I'm running Redhat 7.2 with Redhat kernel 2.4.7-10 and when I have a large number of people downloading from me at one time, Samba, ProFTPD, Apache, and SSH all stop responding, and I get the following in my log, the only thing special I could think about my setup is I have 3 NIC's bonded together to be load balanced on the same IP, any ideas?
That looks like a vm bug that was fixed around the 2.4.9-31 timeframe. You really should be using the latest errata kernel that Red Hat has released, as many bugs have been fixed since 2.4.7-10.
-ben