Skip to contentSkip to conversation

Error with cgroups: "unused subsystem ns"

DISCUSSION Sharyathi Nagesh 3 participants 5 messages Dec 17, 2008 – Dec 19, 2008

lore
Share & tools
lore
#0 lore raw

Hi

Description:
We have noticed kernel error message on the dmesg when ltp test cases
are executed. We observed multiple instance of error message:
"Not cloning cgroup for unused subsystem ns" in dmesg

Test:
The messages were seen when ltp test case for the containers are executed:
ltp -f containers

Solution:
Alexey Dobriyan had provided a patch to address the issue. Where the printk was removed. The patch still haven't made it to main line and hence we still see the messages. The patch can be found here:
http://git.openvz.org/?p=linux-2.6.24-openvz;a=log;h=v2.6.24-ovz002.2

Would this be the right way to handle the issue? if so
When can we expect this to be included into mainline Thanks
Yeehaw

#1 lore raw
Quoted 18 lines · Sharyathi Nagesh

Hi

Description:
We have noticed kernel error message on the dmesg when ltp test cases are
executed. We observed multiple instance of error message:
"Not cloning cgroup for unused subsystem ns" in dmesg

Test:
The messages were seen when ltp test case for the containers are executed:
ltp -f containers

Solution:
Alexey Dobriyan had provided a patch to address the issue. Where the printk was removed. The patch still haven't made it to main line and hence we still see the messages.
The patch can be found here:
http://git.openvz.org/?p=linux-2.6.24-openvz;a=log;h=v2.6.24-ovz002.2

Quoted 3 lines

Would this be the right way to handle the issue? if so
When can we expect this to be included into mainline

Resend the patch to lkml. Assuming you are talking about commit id 93af30a60055a83a00188ae04b88bbd9b4e1489a, go ahead and add my Acked-by.

Converting it to a pr_debug() might be another thing to consider, but I'm not sure anyone wants these messages.

-serge

author Acked-by Serge E. Hallyn
#2 lore raw

Serge

Quoted 4 lines

Resend the patch to lkml. Assuming you are talking about commit id 93af30a60055a83a00188ae04b88bbd9b4e1489a, go ahead and add my Acked-by.

Yes it is the same patch. I am attaching the patch for 2.6.28-rc8 let me know: when this can accepted into main line?

Quoted 4 lines

Converting it to a pr_debug() might be another thing to consider, but I'm not sure anyone wants these messages.

-serge

Thanks
Yeehaw

-----------------

kernel/cgroup.c +0 -3
Index: linux-2.6.28-rc8/kernel/cgroup.c
===================================================================
@@ -2934,9 +2934,6 @@
  again:
 	root = subsys->root;
 	if (root == &rootnode) {
-		printk(KERN_INFO
-		       "Not cloning cgroup for unused subsystem %s\n",
-		       subsys->name);
 		mutex_unlock(&cgroup_mutex);
 		return 0;
 	}
#3 lore raw
Quoted 34 lines · On Thu, 18 Dec 2008 16:45:38 +0530 Sharyathi Nagesh <[email protected]> wrote:

Serge

Quoted 4 lines

Resend the patch to lkml. Assuming you are talking about commit id 93af30a60055a83a00188ae04b88bbd9b4e1489a, go ahead and add my Acked-by.

Yes it is the same patch. I am attaching the patch for 2.6.28-rc8 let me
know: when this can accepted into main line?

Quoted 4 lines

Converting it to a pr_debug() might be another thing to consider, but I'm not sure anyone wants these messages.

-serge

Thanks
Yeehaw

[unused_subsystem_ns.patch text/x-diff (667B)]

-----------------

Index: linux-2.6.28-rc8/kernel/cgroup.c
===================================================================

kernel/cgroup.c line 2939

Quoted 9 lines
kernel/cgroup.c +0 -3
@@ -2934,9 +2934,6 @@
  again:
 	root = subsys->root;
 	if (root == &rootnode) {
-		printk(KERN_INFO
-		       "Not cloning cgroup for unused subsystem %s\n",
-		       subsys->name);
 		mutex_unlock(&cgroup_mutex);
 		return 0;
 	}

I'm unsure waht to make of this. It has no changelog. It appears to have been authored by yourself, although it's unclear why it has Alexey's signoff as well.

author Acked-by Serge E. Hallyn
#4 lore raw
Quoted 4 lines · Andrew Morton

I'm unsure waht to make of this. It has no changelog. It appears to have been authored by yourself, although it's unclear why it has Alexey's signoff as well.

This patch is originally from Alexy I have just ported it into latest kernel mainline Thanks
Yeehaw

This patch removes spurious warning messages that are thrown into the console during cgroup operations.

-----------------

-----------------

kernel/cgroup.c +0 -3
Index: linux-2.6.28-rc8/kernel/cgroup.c
===================================================================
@@ -2934,9 +2934,6 @@
  again:
 	root = subsys->root;
 	if (root == &rootnode) {
-		printk(KERN_INFO
-		       "Not cloning cgroup for unused subsystem %s\n",
-		       subsys->name);
 		mutex_unlock(&cgroup_mutex);
 		return 0;
 	}