eth0 goes offline under heavy FTP load

[Dustin Larmeir]

segfaults could indicate a ram problem, or some other issue with the
hardware. I would run a memtest and an ethtool -t eth0 to see if any errors
are found.

On Tue, Jul 13, 2010 at 3:14 PM, Sadvary, Bill <sadvary at dickinson.edu>wrote:

> >> Red Hat 5.5
> >>
> >> Under a heavy FTP load, my eth0 interface goes off line.  If I
> deactivate/activate then everything >> is fine until the next heavy load.
> >>
> >> Any ideas?
>
> >I'd check dmesg to see if there are any messages .. specifically regarding
> conntrack
> >
> >Barry
>
> I'm seeing stuff like this..
>
> INFO: task vsftpd:3588 blocked for more than 120 seconds.
> "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
> vsftpd        D ffff8100ec9946f0     0  3588   3581
> (NOTLB)
>  ffff810098485c68 0000000000000082 0000000000000000 0000000000000000
>  0000000000000000 0000000000000009 ffff81009f8de0c0 ffff8100c0a060c0
>  000001c76ad9ce96 0000000000000a97 ffff81009f8de2a8 00000001a5f9c140
> Call Trace:
>  [<ffffffff80063ff8>] thread_return+0x62/0xfe
>  [<ffffffff80064c6f>] __mutex_lock_slowpath+0x60/0x9b
>  [<ffffffff80064cb9>] .text.lock.mutex+0xf/0x14
>  [<ffffffff88035883>] :jbd:__log_wait_for_space+0x51/0xaa
>  [<ffffffff88032040>] :jbd:start_this_handle+0x323/0x36c
>  [<ffffffff800a1ba4>] autoremove_wake_function+0x0/0x2e
>  [<ffffffff88032152>] :jbd:journal_start+0xc9/0x100
>  [<ffffffff88053898>] :ext3:ext3_create+0x49/0x102
>  [<ffffffff8000f5e1>] generic_permission+0x52/0xca
>  [<ffffffff8003a8a2>] vfs_create+0xe6/0x158
>  [<ffffffff8001b09a>] open_namei+0x19d/0x6d5
>  [<ffffffff800274fb>] do_filp_open+0x1c/0x38
>  [<ffffffff80019e1e>] do_sys_open+0x44/0xbe
>  [<ffffffff8005e28d>] tracesys+0xd5/0xe0
>
> jsvc[7261]: segfault at 0000000000000000 rip 0000003e2dc797e0 rsp
> 00007fffc1647b78 error 4
>  connection1:0: ping timeout of 5 secs expired, recv timeout 5, last rx
> 4307712903, last ping 4307717903, now 4307722903
>  connection1:0: detected conn error (1011)
> eth1: Using EEPROM-set media 100baseTx-FDX.
> eth1: no IPv6 routers present
> eth0: Using EEPROM-set media 100baseTx-FDX.
> eth0: no IPv6 routers present
> mtrr: type mismatch for f8000000,400000 old: write-back new:
> write-combining
>
> --
> redhat-list mailing list
> unsubscribe mailto:redhat-list-request at redhat.com?subject=unsubscribe
> https://www.redhat.com/mailman/listinfo/redhat-list
>