Bug 445201 - kernel: iwl4965: Can not allocate SKB buffers
kernel: iwl4965: Can not allocate SKB buffers
Status: CLOSED CURRENTRELEASE
Product: Fedora
Classification: Fedora
Component: kernel (Show other bugs)
9
x86_64 Linux
medium Severity medium
: ---
: ---
Assigned To: John W. Linville
Fedora Extras Quality Assurance
:
Depends On:
Blocks:
  Show dependency treegraph
 
Reported: 2008-05-05 09:46 EDT by Neal Becker
Modified: 2008-11-24 13:16 EST (History)
3 users (show)

See Also:
Fixed In Version: kernel-2.6.25.9-76.fc9
Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of:
Environment:
Last Closed: 2008-07-08 15:56:20 EDT
Type: ---
Regression: ---
Mount Type: ---
Documentation: ---
CRM:
Verified Versions:
Category: ---
oVirt Team: ---
RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: ---


Attachments (Terms of Use)
dmesg near the SKB allocation failure (4.23 KB, text/plain)
2008-05-18 14:48 EDT, James
no flags Details
/proc/meminfo just after SKB message (846 bytes, text/plain)
2008-05-18 14:49 EDT, James
no flags Details

  None (edit)
Description Neal Becker 2008-05-05 09:46:03 EDT
Description of problem:

May  5 09:14:19 nbecker1 kernel: iwl4965: Can not allocate SKB buffers

Version-Release number of selected component (if applicable):
kernel-2.6.25-14.fc9.x86_64

How reproducible:

Don't know

Steps to Reproduce:
1.
2.
3.
  
Actual results:


Expected results:


Additional info:
Comment 1 Neal Becker 2008-05-05 09:47:03 EDT
May  5 08:53:15 nbecker1 kernel: iwl4965: Error sending REPLY_TX_PWR_TABLE_CMD:
Already sending a host command
May  5 08:53:15 nbecker1 kernel: iwl4965: Error sending REPLY_RXON: Already
sending a host command
May  5 08:53:15 nbecker1 kernel: iwl4965: Error setting new configuration (-16).
May  5 08:53:15 nbecker1 kernel: iwl4965: Microcode SW error detected. 
Restarting 0x82000000.
May  5 08:53:15 nbecker1 kernel: iwl4965: Error sending REPLY_RXON_TIMING: time
out after 500ms.
May  5 08:53:15 nbecker1 kernel: iwl4965: REPLY_RXON_TIMING failed - Attempting
to continue.
May  5 08:53:15 nbecker1 kernel: iwl4965: Error sending REPLY_RXON_ASSOC: time
out after 500ms.
Comment 2 John W. Linville 2008-05-06 15:49:34 EDT
Does the message in the original comment occur frequently?  Or rarely?  It 
looks like something that might happen occasionally but which should not 
prevent the driver from working.

The messages in comment 1 are something different.  Do you see them often?  
Does the driver continue to work?  Or not?
Comment 3 Neal Becker 2008-05-07 06:38:31 EDT
'Already sending' occurs 38 times in last about 24 hours - keep in mind all
that time I haven't actually _used_ the wireless connection - it's been
connected wired.

SKB occurs:
     7:May  5 09:14:19 nbecker1 kernel: iwl4965: Can not allocate SKB buffers
      8:May  5 09:14:48 nbecker1 kernel: iwl4965: Can not allocate SKB buffers
      9:May  5 09:14:49 nbecker1 kernel: iwl4965: Can not allocate SKB buffers
     10:May  5 09:15:53 nbecker1 kernel: iwl4965: Can not allocate SKB buffers
     11:May  5 09:15:54 nbecker1 kernel: iwl4965: Can not allocate SKB buffers
     12:May  5 09:51:21 nbecker1 kernel: iwl4965: Can not allocate SKB buffers
Comment 4 John W. Linville 2008-05-07 08:38:03 EDT
Neal, could you attach the contents of /proc/meminfo when the "Can not 
allocate SKB buffers" message has recently appeared?

It looks like it is a little bursty, which makes some sense to me depending on 
how memory is being used on your system.
Comment 5 Bug Zapper 2008-05-14 06:39:45 EDT
Changing version to '9' as part of upcoming Fedora 9 GA.
More information and reason for this action is here:
http://fedoraproject.org/wiki/BugZappers/HouseKeeping
Comment 6 James 2008-05-18 14:27:31 EDT
I saw this today (kernel-2.6.24.7-92.fc8): a whole slew of them. Reloaded the
iwl4965 module just in case. Came as I was downloading a torrent at about
800KiB/s or so, I'll see if I can reproduce it.
Comment 7 James 2008-05-18 14:47:29 EDT
OK, here we go. Torrent coming in at about 200KiB/s, 57 active connections. I've
attached a dmesg output from around the event (notice all the unknown symbol
messages, must've been when I removed iwl4965) and /proc/meminfo from around
that time as well.

It *seems* that it's actually provoked by other activity: I was reading an ISO
image from a DVD at the time with K3B. I've got the torrent going again at
around 380KiB/s. No problems so far. Let's try creating the image again, restart
Deluge...

OK, just seen a few, but this is probably a coincidence. The error seems to turn
up more when the number of connections rises.
Comment 8 James 2008-05-18 14:48:35 EDT
Created attachment 305869 [details]
dmesg near the SKB allocation failure
Comment 9 James 2008-05-18 14:49:27 EDT
Created attachment 305870 [details]
/proc/meminfo just after SKB message
Comment 10 James 2008-06-13 05:43:27 EDT
John, I don't know if the original reporter still sees this bug, but I'll check
the NEEDINFO box below.
Comment 11 Neal Becker 2008-06-13 06:40:57 EDT
Not seeing it in my last 2 /var/log/messages
Comment 12 James 2008-11-22 18:07:04 EST
I'm going to have to ask that this be re-opened, sorry! Just saw:

iwlagn: Can not allocate SKB buffers

in a dmesg from kernel 2.6.27.6-43.fc9.x86_64.
Comment 13 Neal Becker 2008-11-23 08:57:27 EST
Nov 19 06:53:33 nbecker1 kernel: Linux version 2.6.27.5-37.fc9.x86_64 (mockbuild@) (gcc version 4.3.0 20080428 (Red Hat 4.3.0-8) (GCC) ) #1 SMP Wed Nov 12 18:31:37 EST 2008
...
Nov 19 13:54:02 nbecker1 kernel: iwlagn: Can not allocate SKB buffers
Comment 14 John W. Linville 2008-11-24 12:45:45 EST
Does the device fail afterwards?  Or just that single message in the logs?
Comment 15 James 2008-11-24 13:01:45 EST
They seem to be rather intermittent:

[root@rhapsody james]# grep SKB /var/log/messages*
/var/log/messages-20081102:Oct 26 13:36:30 rhapsody kernel: iwl4965: Can not allocate SKB buffers
/var/log/messages-20081123:Nov 21 04:53:16 rhapsody kernel: iwlagn: Can not allocate SKB buffers
/var/log/messages-20081123:Nov 22 22:53:55 rhapsody kernel: iwlagn: Can not allocate SKB buffers

The device continues OK, didn't notice anything was amiss until I ran dmesg. I'll try downloading some torrents again, see if I can reproduce the bursts I was seeing earlier.
Comment 16 Neal Becker 2008-11-24 13:13:45 EST
I don't see any problems with my device (kernel 2.6.27.5-37.fc9.x86_64), just the mesage.
Comment 17 James 2008-11-24 13:16:09 EST
This has also been seen in Ubuntu --- see https://bugs.launchpad.net/ubuntu/intrepid/+source/linux/+bug/250139 --- with a bit more detail than we have here. Some of the content of their reports seems familiar to me (message coincident with heavy disc activity, etc.).

Note You need to log in before you can comment on or make changes to this bug.