orinoco.c: Error -110 writing Tx descriptor to BAP
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
linux-source-2.6.15 (Debian) |
Fix Released
|
Unknown
|
|||
linux-source-2.6.15 (Ubuntu) |
Fix Released
|
Medium
|
Ben Collins |
Bug Description
I'm actually using Kubuntu here, but this is hardly a Kubuntu-specific issue, so I didn't add Kubuntu as Keyword.
High net load using a Netgear MA311 wlan pcmcia card with the orinoco/hermes module effectively results in a DOS
on my entire system. It is easily reproducible. All I have to do is, for example, to wget a big file (couple
hundred MB or more) from another host on my LAN.
The wlan card can't keep up, it seems:
Mar 27 23:36:05 localhost kernel: eth2: error -110 reading frame. Frame dropped.
Mar 27 23:36:05 localhost kernel: eth2: Error -110 writing Tx descriptor to BAP
Mar 27 23:36:06 localhost last message repeated 88 times
Mar 27 23:36:06 localhost kernel: eth2: Error -110 wridescriptor to BAP
Mar 27 23:36:06 localhost kernel: eth2: Error -110 writing Tx descriptor to BAP
Mar 27 23:36:06 localhost last message repeated 2 times
Mar 27 23:36:06 localhost kernel: hermes @ IO 0x100: Error -16 issuing command.
Mar 27 23:36:06 localhost kernel: hermes @ IO 0x100: Error -16 issuing command.
Mar 27 23:36:06 localhost kernel: eth2: Error -110 writing Tx descriptor to BAP
Mar 27 23:36:06 localhost kernel: eth2: Error -110 writing Tx descriptor to BAP
Mar 27 23:36:06 localhost kernel: eth2: Error -110 writing Tx descrdescriptor to BAP
Mar 27 23:36:06 localhost kernel: eth2: Error -110 writing Tx descriptor to BAP
An *insane* amount of these above lines get written to syslog; up to one hundred lines per second.
top - 00:34:53 up 32 min, 2 users, load average: 4.39, 3.00, 1.58
Tasks: 74 total, 4 running, 70 sleeping, 0 stopped, 0 zombie
Cpu(s): 2.6% us, 5.2% sy, 0.7% ni, 0.0% id, 0.0% wa, 0.0% hi, 91.5% si
Mem: 516244k total, 451616k used, 64628k free, 39624k buffers
Swap: 658624k total, 0k used, 658624k free, 280944k cached
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
6106 root 16 0 1552 380 316 R 39.0 0.1 1:24.56 dd
7865 syslog 35 10 1740 740 620 R 15.1 0.1 0:25.69 syslogd
6150 root 15 0 90028 21m 2952 S 7.4 4.3 0:41.77 Xorg
6108 klog 15 0 2408 1464 436 S 6.6 0.3 0:27.17 klogd
7609 frederik 16 0 27156 14m 11m S 6.6 2.8 0:28.03 kicker
7899 frederik 17 0 2080 1044 820 R 3.1 0.2 0:02.39 top
2 root 39 19 0 0 0 R 2.8 0.0 0:07.16 ksoftirqd/0
7889 frederik 16 0 26920 14m 11m S 1.7 2.9 0:06.93 konsole
7617 frederik 16 0 20736 11m 9560 S 0.9 2.2 0:02.80 kcpuload
6570 hal 16 0 7120 5656 1636 S 0.6 1.1 0:02.40 hald
[...]
The process at fault is, in other words, PID 6106:
frederik@bart:~$ ps -ef
UID PID PPID C STIME TTY TIME CMD
root 1 0 0 00:02 ? 00:00:00 init [2]
root 2 1 0 00:02 ? 00:00:13 [ksoftirqd/0]
root 3 1 0 00:02 ? 00:00:00 [events/0]
root 4 3 0 00:02 ? 00:00:00 [khelper]
root 22 3 0 00:02 ? 00:00:00 [kacpid]
root 131 3 0 00:02 ? 00:00:00 [kblockd/0]
root 166 3 0 00:02 ? 00:00:00 [pdflush]
root 167 3 0 00:02 ? 00:00:00 [pdflush]
root 169 3 0 00:02 ? 00:00:00 [aio/0]
root 168 1 0 00:02 ? 00:00:00 [kswapd0]
root 756 1 0 00:02 ? 00:00:00 [kseriod]
root 1125 1 0 00:02 ? 00:00:00 [kjournald]
root 1152 1 0 00:02 ? 00:00:00 udevd
root 2225 1 0 00:02 ? 00:00:00 [khpsbpkt]
root 4024 1 0 00:03 ? 00:00:00 [khubd]
root 5334 1 0 00:03 ? 00:00:00 [pccardd]
root 5343 1 0 00:03 ? 00:00:00 [pccardd]
root 5453 1 0 00:03 ? 00:00:00 [knodemgrd_0]
root 5612 3 0 00:03 ? 00:00:00 [ipw2100/0]
root 6106 1 8 00:03 ? 00:03:16 /bin/dd bs 1 if /proc/kmsg of /var/run/klogd/kmsg
klog 6108 1 2 00:03 ? 00:00:57 /sbin/klogd -P /var/run/klogd/kmsg
root 6136 1 0 00:03 ? 00:00:00 /usr/bin/kdm
root 6150 6136 3 00:03 ? 00:01:17 /usr/X11R6/bin/X -nolisten tcp :0 vt7
-auth /var/run/
root 6191 6136 0 00:03 ? 00:00:00 -:0
root 6496 1 0 00:03 ? 00:00:00 /usr/sbin/acpid -c /etc/acpi/events -s /var/run/
102 6558 1 0 00:03 ? 00:00:00 /usr/bin/
hal 6570 1 0 00:03 ? 00:00:02 /usr/sbin/hald --drop-privileges
[...]
Not sure if there is much you can do about the orinoco part in this (although I don't remember seeing this
behavior with Sarge), but it would be nice if it didn't cause the system to go haywire and become extremely
unresponsive due to the huge amount of logging.
Thank you for a great distro,
Frederik Dannemare
I'm actually using Kubuntu here, but this is hardly a Kubuntu-specific issue, so I didn't add Kubuntu as Keyword.
High net load using a Netgear MA311 wlan pcmcia card with the orinoco/hermes module effectively results in a DOS
on my entire system. It is easily reproducible. All I have to do is, for example, to wget a big file (couple
hundred MB or more) from another host on my LAN.
The wlan card can't keep up, it seems:
Mar 27 23:36:05 localhost kernel: eth2: error -110 reading frame. Frame dropped.
Mar 27 23:36:05 localhost kernel: eth2: Error -110 writing Tx descriptor to BAP
Mar 27 23:36:06 localhost last message repeated 88 times
Mar 27 23:36:06 localhost kernel: eth2: Error -110 wridescriptor to BAP
Mar 27 23:36:06 localhost kernel: eth2: Error -110 writing Tx descriptor to BAP
Mar 27 23:36:06 localhost last message repeated 2 times
Mar 27 23:36:06 localhost kernel: hermes @ IO 0x100: Error -16 issuing command.
Mar 27 23:36:06 localhost kernel: hermes @ IO 0x100: Error -16 issuing command.
Mar 27 23:36:06 localhost kernel: eth2: Error -110 writing Tx descriptor to BAP
Mar 27 23:36:06 localhost kernel: eth2: Error -110 writing Tx descriptor to BAP
Mar 27 23:36:06 localhost kernel: eth2: Error -110 writing Tx descrdescriptor to BAP
Mar 27 23:36:06 localhost kernel: eth2: Error -110 writing Tx descriptor to BAP
An *insane* amount of these above lines get written to syslog; up to one hundred lines per second.
top - 00:34:53 up 32 min, 2 users, load average: 4.39, 3.00, 1.58
Tasks: 74 total, 4 running, 70 sleeping, 0 stopped, 0 zombie
Cpu(s): 2.6% us, 5.2% sy, 0.7% ni, 0.0% id, 0.0% wa, 0.0% hi, 91.5% si
Mem: 516244k total, 451616k used, 64628k free, 39624k buffers
Swap: 658624k total, 0k used, 658624k free, 280944k cached
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
6106 root 16 0 1552 380 316 R 39.0 0.1 1:24.56 dd
7865 syslog 35 10 1740 740 620 R 15.1 0.1 0:25.69 syslogd
6150 root 15 0 90028 21m 2952 S 7.4 4.3 0:41.77 Xorg
6108 klog 15 0 2408 1464 436 S 6.6 0.3 0:27.17 klogd
7609 frederik 16 0 27156 14m 11m S 6.6 2.8 0:28.03 kicker
7899 frederik 17 0 2080 1044 820 R 3.1 0.2 0:02.39 top
2 root 39 19 0 0 0 R 2.8 0.0 0:07.16 ksoftirqd/0
7889 frederik 16 0 26920 14m 11m S 1.7 2.9 0:06.93 konsole
7617 frederik 16 0 20736 11m 9560 S 0.9 2.2 0:02.80 kcpuload
6570 hal 16 0 7120 5656 1636 S 0.6 1.1 0:02.40 hald
[...]
The process at fault is, in other words, PID 6106:
frederik@bart:~$ ps -ef
UID PID PPID C STIME TTY TIME CMD
root 1 0 0 00:02 ? 00:00:00 init [2]
root 2 1 0 00:02 ? 00:00:13 [ksoftirqd/0]
root 3 1 0 00:02 ? 00:00:00 [events/0]
root 4 3 0 00:02 ? 00:00:00 [khelper]
root 22 3 0 00:02 ? 00:00:00 [kacpid]
root ...