Fedora Users — Crash on network transfer

Hello all,

Over the past few weeks I've been trying to figure this out, but I'm quite stuck. I'm pretty new to Linux, so my troubleshooting has only gotten me so far...

Essentially, the problem occurs when I try to ftp something at relatively high speeds to or from my FC3 box. Browsing from the box or other items don't cause this problem, as they don't get the bandwidth that apparently triggers this problem. When I start my FTP, things go smoothly for a few moments (say they're running at 10MB/sec), and then the machine dies (no response, although the lights on the keyboard are not flashing).

I had been convinced it was a network card problem, so I swapped my old car out and put in a dlink DGE-530T (gigabit), but got the same results. I changed the DMA settings on my drives and, if I disable DMA, the crash does not occur - it's only transferring at a few MB/sec then (like 2MB/sec). Enabling DMA, but at a lower speed (say ATA/66) causes my ssh session to lock up, but I can continue to use the box, and I got some of the crash output in my dmesg (attached). I noticed a line that said something about IRQ and thought possibly my network card doing something funky with memory, so I enabled IRQ_Moderation on my network card.

Trying the same experiment of transferring a file (but with the mod on now), I got another crash on the current SSH session, but checking dmesg, I got a slightly different error. Perhaps it's the same thing, but I'm not experienced enough to tell :( I've attached this log as well.

My interrupts look like this:

           CPU0
0:     843136          XT-PIC timer
1:         10          XT-PIC i8042
2:          0          XT-PIC cascade
5:          0          XT-PIC ivtv: iTVC15/16 mpg2 encoder chip
8:          1          XT-PIC rtc
9:       2015          XT-PIC SysKonnect SK-98xx
10:      13502          XT-PIC ide2
11:      50395          XT-PIC EMU10K1, nvidia
12:         92          XT-PIC i8042
14:      20719          XT-PIC ide0
NMI:          0
ERR:          0

I don't think anything looks out of the ordinary in there, so I'm really stumped. Turning off DMA isn't much of an option on this box as I run MythTV on it and need good drive bandwidth.

I'd appreciate any thoughts or ideas anyone has! Hopefully this is the right place to send this.

Thanks,

Chris

Unable to handle kernel paging request at virtual address 7005d6a0
 printing eip:
022ceabc
*pde = 00000000
Oops: 0000 [#1]
Modules linked in: bttv video_buf(U) v4l2_common(U) btcx_risc lirc_i2c(U) lirc_dev(U) msp3400(U) saa7115(U) tuner(U) tveeprom(U) ivtv(U) i2c_algo_bit i2c_core videodev lp autofs4 smbfs sunrpc xfs dm_mod nvidia(U) md5 ipv6 parport_pc parport emu10k1_gp gameport snd_emu10k1(U) snd_rawmidi(U) snd_seq_device(U) snd_ac97_codec(U) snd_pcm_oss(U) snd_mixer_oss(U) snd_pcm(U) snd_timer(U) snd_page_alloc(U) snd_util_mem(U) snd_hwdep(U) snd(U) soundcore sk98lin ext3 jbd
CPU:    0
EIP:    0060:[<022ceabc>]    Tainted: PF  VLI
EFLAGS: 00010246   (2.6.9-1.681_FC3)
EIP is at tcp_sendmsg+0xc12/0xe50
eax: c307fafe   ebx: 0f84d768   ecx: 00000001   edx: 1ae44280
esi: 00000002   edi: 00000000   ebp: 0f84d500   esp: 0e403e14
ds: 007b   es: 007b   ss: 0068
Process ncftp (pid: 3275, threadinfo=0e403000 task=0f77ac90)
Stack: 0e403e34 10703780 0e403e6c 216489a0 021c855f 1a9060e0 08711828 00000000
       00008000 000005b4 00000000 00000000 0f84d768 0e403ed0 00000000 00000000
       7fffffff 023783e0 0f84d500 0e403eac 0e403ef8 022ead6e 00008000 023794a0
Call Trace:
 [<021c855f>] socket_has_perm+0x51/0x5f
 [<022ead6e>] inet_sendmsg+0x38/0x42
 [<022a43c1>] sock_aio_write+0x107/0x114
 [<02165b9a>] do_sync_write+0x97/0xc9
 [<021c784d>] selinux_file_permission+0x114/0x11d
 [<0211d26f>] autoremove_wake_function+0x0/0x2d
 [<02165c92>] vfs_write+0xc6/0xe2
 [<02165d4c>] sys_write+0x3c/0x62
Code: <3>Debug: sleeping function called from invalid context at include/linux/rwsem.h:43
in_atomic():0[expected: 0], irqs_disabled():1
 [<0211cbcb>] __might_sleep+0x7d/0x8a
 [<0215e726>] rw_vm+0x20e/0x47a
 [<022cea91>] tcp_sendmsg+0xbe7/0xe50
 [<022cea91>] tcp_sendmsg+0xbe7/0xe50
 [<0215ee70>] get_user_size+0x30/0x57
 [<022cea91>] tcp_sendmsg+0xbe7/0xe50
 [<0210682b>] show_registers+0x109/0x15e
 [<02106a2f>] die+0x14a/0x241
 [<0211937e>] do_page_fault+0x0/0x511
 [<0211937e>] do_page_fault+0x0/0x511
 [<02119733>] do_page_fault+0x3b5/0x511
 [<022ceabc>] tcp_sendmsg+0xc12/0xe50
 [<0214692e>] __rmqueue+0xbb/0x106
 [<22885a40>] ext3_readpages+0x12/0x14 [ext3]
 [<22884ea7>] ext3_get_block+0x0/0x6c [ext3]
 [<02149cef>] read_pages+0x2d/0xd0
 [<022dd69e>] tcp_v4_send_check+0x7c/0xb9
 [<0215222e>] follow_page_pte+0xec/0xfd
 [<0211937e>] do_page_fault+0x0/0x511
 [<022ceabc>] tcp_sendmsg+0xc12/0xe50
 [<021c855f>] socket_has_perm+0x51/0x5f
 [<022ead6e>] inet_sendmsg+0x38/0x42
 [<022a43c1>] sock_aio_write+0x107/0x114
 [<02165b9a>] do_sync_write+0x97/0xc9
 [<021c784d>] selinux_file_permission+0x114/0x11d
 [<0211d26f>] autoremove_wake_function+0x0/0x2d
 [<02165c92>] vfs_write+0xc6/0xe2
 [<02165d4c>] sys_write+0x3c/0x62
 Bad EIP value.

Unable to handle kernel NULL pointer dereference at virtual address 000005b4
  printing eip: 022ce2d5
*pde = 00000000
Oops: 0000 [#1]
Modules linked in: bttv video_buf(U) v4l2_common(U) btcx_risc lirc_i2c(U) lirc_dev(U) msp3400(U) saa7115(U) tuner(U) tveeprom(U) ivtv(U) i2c_algo_bit i2c_core videodev lp autofs4 smbfs sunrpc xfs dm_mod nvidia(U) md5 ipv6 parport_pc parport emu10k1_gp gameport snd_emu10k1(U) snd_rawmidi(U) snd_seq_device(U) snd_ac97_codec(U) snd_pcm_oss(U) snd_mixer_oss(U) snd_pcm(U) snd_timer(U) snd_page_alloc(U) snd_util_mem(U) snd_hwdep(U) snd(U) soundcore sk98lin ext3 jbd
CPU:    0
EIP:    0060:[<022ce2d5>]    Tainted: PF  VLI
EFLAGS: 00010a47   (2.6.9-1.681_FC3)
EIP is at tcp_sendmsg+0x42b/0xe50
eax: ffc9700c   ebx: 12494e34   ecx: 0000e700   edx: 00000000
esi: 000005b4   edi: ffffffff   ebp: 19ed7600   esp: 12494e18
ds: 007b   es: 007b   ss: 0068
Process ncftp (pid: 3513, threadinfo=12494000 task=1154e5f0)
Stack: 0d43d4a0 12494e6c 12bd73e0 021c855f 153610e0 08707828 00008000 00000000
       000005b4 00000000 00000000 19ed7868 12494ed0 00000000 00000000 7fffffff
       023783e0 19ed7600 12494eac 12494ef8 022ead6e 00008000 023794a0 12494eac Call Trace:
 [<021c855f>] socket_has_perm+0x51/0x5f
 [<022ead6e>] inet_sendmsg+0x38/0x42
 [<022a43c1>] sock_aio_write+0x107/0x114
 [<02165b9a>] do_sync_write+0x97/0xc9
 [<021c784d>] selinux_file_permission+0x114/0x11d
 [<0211d26f>] autoremove_wake_function+0x0/0x2d
 [<02165c92>] vfs_write+0xc6/0xe2 
 [<02165d4c>] sys_write+0x3c/0x62
Code: <3>Debug: sleeping function called from invalid context at include/linux/rwsem.h:43
in_atomic():0[expected: 0], irqs_disabled():1 
 [<0211cbcb>] __might_sleep+0x7d/0x8a 
 [<0215e726>] rw_vm+0x20e/0x47a 
 [<022ce2aa>] tcp_sendmsg+0x400/0xe50 
 [<022ce2aa>] tcp_sendmsg+0x400/0xe50 
 [<0215ee70>] get_user_size+0x30/0x57 
 [<022ce2aa>] tcp_sendmsg+0x400/0xe50 
 [<0210682b>] show_registers+0x109/0x15e 
 [<02106a2f>] die+0x14a/0x241 
 [<0211937e>] do_page_fault+0x0/0x511 
 [<0211937e>] do_page_fault+0x0/0x511 
 [<02119733>] do_page_fault+0x3b5/0x511
 [<022ce2d5>] tcp_sendmsg+0x42b/0xe50 
 [<228f9b0a>] SkGeIsrOnePort+0x1ac/0x323 [sk98lin] 
 [<02107f51>] handle_IRQ_event+0x25/0x4f 
 [<021c3f3a>] avc_has_perm_noaudit+0x8d/0xda 
 [<0215222e>] follow_page_pte+0xec/0xfd 
 [<0211937e>] do_page_fault+0x0/0x511 
 [<022ce2d5>] tcp_sendmsg+0x42b/0xe50 
 [<021c855f>] socket_has_perm+0x51/0x5f 
 [<022ead6e>] inet_sendmsg+0x38/0x42 
 [<022a43c1>] sock_aio_write+0x107/0x114 
 [<02165b9a>] do_sync_write+0x97/0xc9 
 [<021c784d>] selinux_file_permission+0x114/0x11d
 [<0211d26f>] autoremove_wake_function+0x0/0x2d 
 [<02165c92>] vfs_write+0xc6/0xe2 
 [<02165d4c>] sys_write+0x3c/0x62
 Bad EIP value.