ping
On 2014/10/14 10:20, Weng Meiling wrote:
> hi Ivana, Dhaval
>
> What do you think about this problem?
>
> On 2014/10/8 8:50, Weng Meiling wrote:
>>
>> On 2014/10/2 2:51, Balbir Singh wrote:
>>> Do you have any debug logs with traces of what cgred
Hi guys??
We meet a situation:
When we use cgexec command to move task, the command stall.
It seems the connect() stall in cgroup_register_unchanged_process():
int cgroup_register_unchanged_process(pid_t pid, int flags)
{
...
if (connect(sk, (struct sockaddr *)&addr,
sizeof(addr.su
hi Ivana, Dhaval
What do you think about this problem?
On 2014/10/8 8:50, Weng Meiling wrote:
>
> On 2014/10/2 2:51, Balbir Singh wrote:
>> Do you have any debug logs with traces of what cgred is doing? My best
>> guess is cgred will need to be enhanced, feel free to send a
cs() to get the task group id
contained in cgroups.procs,
but this is not enough fot the situation mentioned in former mail, am I missing
something?
We used cgred in debug mode, and found the running threads which match the
rules could not
be controlled by cgred.
Thanks!
Weng Meiling
> On Mon
Hi, what do you think about this situation?
Thanks!
Weng Meiling
On 2014/9/24 9:02, Weng Meiling wrote:
> ping...
>
> On 2014/9/22 17:32, Weng Meiling wrote:
>> On 2014/9/20 17:13, Weng Meiling wrote:
>>> Hi guys,
>>> In our environment, we found the following p
ping...
On 2014/9/22 17:32, Weng Meiling wrote:
> On 2014/9/20 17:13, Weng Meiling wrote:
>> Hi guys,
>> In our environment, we found the following problem:
>>
>> we have a process and it has some threads with the same name:
>>
>> # ps -L -eo comm
On 2014/9/20 17:13, Weng Meiling wrote:
> Hi guys,
> In our environment, we found the following problem:
>
> we have a process and it has some threads with the same name:
>
> # ps -L -eo comm,pid,lwp,args,start |grep Test
> Testd 4489 4489 /usr/bin/Testd
Hi guys,
In our environment, we found the following problem:
we have a process and it has some threads with the same name:
# ps -L -eo comm,pid,lwp,args,start |grep Test
Testd 4489 4489 /usr/bin/Testd 09:03:39
Testd 4489 4582 /usr/bin/Testd 09:03:39
Test
Reviewed-by: Weng Meiling
On 2014/7/3 16:55, Ivana Hutarova Varekova wrote:
> When multi subsystems mounted on one place like this:
>
> $ lssubsys -m
> cpu,cpuacct /cgroup/cpu
>
> $ lscgroup
> cpu,cpuacct:/
> cpu,cpuacct:/test
>
> if we delete the cgroup with t
On 2014/7/9 19:14, Ivana Varekova wrote:
>
>
> - Original Message -
>> From: "Weng Meiling"
>> To: "Ivana Hutarova Varekova" ,
>> libcg-devel@lists.sourceforge.net
>> Sent: Thursday, 3 July, 2014 1:49:44 PM
>> Subject: Re: [P
> v4 -> v5 (Ivana Hutarova Varekova)
> - make cgdelete -g cpuacct:test -g cpu:test -g cpu:test failed
> - join extended cgroup list variables to one record, do the list dynamic
> (static version can exceed the given limit and the safe static limit is
ns type fusectl (rw)
securityfs on /sys/kernel/security type securityfs (rw)
cgroup on /cgroup type cgroup (rw)
> - Original Message -
>> From: "Weng Meiling"
>> To: libcg-devel@lists.sourceforge.net
>> Cc: "libo chen"
>> Sent: Friday, August 2, 201
On 2014/6/27 19:19, Ivana Varekova wrote:
>
>
> - Original Message -
>> From: "Weng Meiling"
>> To: "Ivana Varekova"
>> Cc: "libo chen" , "Huang Qiang"
>> , "Libcg Development list"
>> , "
ping...
On 2014/5/19 10:16, Weng Meiling wrote:
> Hi Jan, Ivana, Vivek, Balbir,
>
> What do you think about the problem?
>
> On 2014/5/2 3:26, Dhaval Giani wrote:
>> On Mon, Apr 28, 2014 at 10:05 PM, Weng Meiling
>> wrote:
>>> On 2014/4/29 3:05, Dhaval Gi
test -g cpu:test -g cpuacct:test failed.
- add some comments
- fix the uninitialized warning
Signed-off-by: Weng Meiling
---
src/tools/cgdelete.c | 87 ++--
1 file changed, 78 insertions(+), 9 deletions(-)
diff --git a/src/tools/cgdelete.c b/src/too
Thanks for review, I'll send the v4 patch.
Thanks!
Weng Meiling
On 2014/6/27 12:10, Ivana Varekova wrote:
>
>
> - Original Message -----
>> From: "Weng Meiling"
>> To: "Ivana Varekova"
>> Cc: "Libcg Development list" , &qu
ping...
On 2014/5/19 10:16, Weng Meiling wrote:
> Hi Jan, Ivana, Vivek, Balbir,
>
> What do you think about the problem?
>
> On 2014/5/2 3:26, Dhaval Giani wrote:
>> On Mon, Apr 28, 2014 at 10:05 PM, Weng Meiling
>> wrote:
>>> On 2014/4/29 3:05, Dhaval Gi
Hi Jan, Ivana, Vivek, Balbir,
What do you think about the problem?
On 2014/5/2 3:26, Dhaval Giani wrote:
> On Mon, Apr 28, 2014 at 10:05 PM, Weng Meiling
> wrote:
>> On 2014/4/29 3:05, Dhaval Giani wrote:
>>> Hi Weng,
>>>
>>> I apologize for our tardiness
On 2014/5/2 3:26, Dhaval Giani wrote:
>
> So this idea sounds really racy to me. The only sure way it can be
> ensured is *if* the child runs before the parent exits. I know there
> is a kernel tunable for that, but I am not sure if it is a good idea.
> But just for fun, could you try
> (as root
On 2014/4/29 3:05, Dhaval Giani wrote:
> Hi Weng,
>
> I apologize for our tardiness is responding back!
>
> On Sun, Apr 27, 2014 at 9:09 PM, Weng Meiling
> wrote:
>> Hi Ivana,
>> What do you think about this problem?
>>
>> Weng Meiling
>> Than
Hi Ivana,
What do you think about this problem?
Weng Meiling
Thanks!
On 2014/4/22 11:06, Weng Meiling wrote:
> ping...
> On 2014/4/10 16:09, Weng Meiling wrote:
>> Cgred just migrate forked process when the parent's cgroup is changed,
>> so that the forked process can
ping...
On 2014/4/10 16:09, Weng Meiling wrote:
> Cgred just migrate forked process when the parent's cgroup is changed,
> so that the forked process can stay in the same cgroup which match the
> cgred rules with it's parent. But there is a situation when parent process
> for
other way(just remove
the cgre_was_parent_changed_when_forking() for forked processes) will affect
cgclassify --sticky. But this way is not which I have done some test.
Signed-off-by: Weng Meiling
---
src/daemon/cgrulesengd.c | 5 +
1 file changed, 5 insertions(+)
diff --git a/src/daemon/cgru
On 2013/11/8 11:27, Dhaval Giani wrote:
> On Thu, Nov 7, 2013 at 9:01 PM, Weng Meiling
> wrote:
>> On 2013/11/8 5:55, Dhaval Giani wrote:
>> Hi Dhaval,
>>> Hi Weng,
>>>
>>> On Thu, Oct 24, 2013 at 10:58 PM, Dhaval Giani
>>> wrote:
>>&
On 2013/11/8 5:55, Dhaval Giani wrote:
Hi Dhaval,
> Hi Weng,
>
> On Thu, Oct 24, 2013 at 10:58 PM, Dhaval Giani wrote:
>> On Thu, Oct 24, 2013 at 10:52 PM, Weng Meiling
>> wrote:
>>> On 2013/10/24 21:04, Dhaval Giani wrote:
>>>> On Thu, Oct 24,
From: Weng Meiling
When multi subsystems mounted on one place like this:
$ lssubsys -m
cpu,cpuacct /cgroup/cpu
$ lscgroup
cpu,cpuacct:/
cpu,cpuacct:/test
if we delete the cgroup with the cgdelete -g, and specifying multi
controllers like this:
$ cgdelete -g cpu,cpuacct:test
or
$ cgdelete
On 2013/10/24 21:04, Dhaval Giani wrote:
> On Thu, Oct 24, 2013 at 3:50 AM, Weng Meiling
> wrote:
>> On 2013/10/24 1:04, Dhaval Giani wrote:
>>> Hi Weng,
>>>
>>> Would something similar to the attached patch help you? This of course
>>> is complete
goto loop
+}
after add check "i < cgroup->index" before goto loop, the cg_sort_mem_values()
still doesn't resort the memory.memsw.limit_in_bytes and memort.limit_in_bytes.
The cgconfigparser still failed for same reason:
cgconfi
On 2013/10/23 10:58, Dhaval Giani wrote:
> Hi,
>
> Sorry for the delay in getting back.
>
> On Wed, Oct 16, 2013 at 5:35 AM, Weng Meiling
> wrote:
>>
>> From: Weng Meiling
>>
>> After setting memory.limit_in_bytes and memory.memsw.limit_in_bytes,
>&
d send the third version patch, can that
satisfy what you mentioned in 1?
Thanks!
Weng Meiling
--
October Webinars: Code for Performance
Free Intel webinars can help you accelerate application performance.
Explore ti
From: Weng Meiling
After setting memory.limit_in_bytes and memory.memsw.limit_in_bytes,
use cgsnapshot -f to save the current cgroup configure, then use
cgconfigparser to resume it according the configure file, but it failed.
The steps:
$ cgsnapshot -s memory -f config
$ cat config
mount
From: Weng Meiling
After setting memory.limit_in_bytes and memory.memsw.limit_in_bytes,
use cgsnapshot -f to save the current cgroup configure, then use
cgconfigparser to resume it according the configure file, but it failed.
The steps:
$ cgsnapshot -s memory -f config
$ cat config
mount
On 2013/10/15 15:30, Libo Chen wrote:
> On 2013/10/12 16:35, Weng Meiling wrote:
>>
>> From: Weng Meiling
>>
>> After setting memory.limit_in_bytes and memory.memsw.limit_in_bytes,
>> use cgsnapshot -f to save the current cgroup configure, then use
>> cg
From: Weng Meiling
All commands except cgdelete can display help info when cgroup
is not mounted. cgdelete command display like this:
$ lssubsys -m
$
$cgdelete -h
/usr/bin/cgdelete: libcgroup initialization failed: Cgroup is not mounted
This patch makes cgdelete command can display help info
On 2013/10/12 16:35, Weng Meiling wrote:
>
> From: Weng Meiling
>
> After setting memory.limit_in_bytes and memory.memsw.limit_in_bytes,
> use cgsnapshot -f to save the current cgroup configure, then use
> cgconfigparser to resume it according the configure file, but it fail
From: Weng Meiling
After setting memory.limit_in_bytes and memory.memsw.limit_in_bytes,
use cgsnapshot -f to save the current cgroup configure, then use
cgconfigparser to resume it according the configure file, but it failed.
The steps:
$ cgsnapshot -s memory -f config
$ cat config
mount
Hi Ivana,
what do you think about the second version patch about cgdelete problem?
On 2013/10/8 14:31, Weng Meiling wrote:
>
> From: Weng Meiling
>
> When multi subsystems mounted on one place like this:
>
> $ lssubsys -m
> cpu,cpuacct /cgroup/cpu
>
> $ lsc
From: Weng Meiling
When multi subsystems mounted on one place like this:
$ lssubsys -m
cpu,cpuacct /cgroup/cpu
$ lscgroup
cpu,cpuacct:/
cpu,cpuacct:/test
if we delete the cgroup with the cgdelete -g, and specifying multi
controllers like this:
$ cgdelete -g cpu,cpuacct:test
or
$ cgdelete
On 2013/9/27 14:32, Ivana Varekova wrote:
>
>
> - Original Message -
>> From: "Weng Meiling"
>> To: "Ivana Varekova"
>> Cc: "Huang Qiang" huangqi...@huawei.com>, "libo chen"
>> , "Libcg Development li
On 2013/9/27 11:57, Dhaval Giani wrote:
>
> Clearly this rule is invalid. So yes, I would expect cgrules to fail.
> (What to do with all processes for all users). I think the behavior is
> just fine.
>
> Dhaval
>
>
yeah, but what do you think the different behav
On 2013/9/27 11:50, Steve Olivieri wrote:
>
>
>> On 2013/9/27 10:13, Dhaval Giani wrote:
>>> On Thu, Sep 26, 2013 at 9:43 PM, Weng Meiling
>>> wrote:
>>>> On 2013/9/27 9:20, Dhaval Giani wrote:
>>>>> On Thu, Sep 26, 20
On 2013/9/27 9:17, Dhaval Giani wrote:
> Hi weng,
>
> On Thu, Sep 26, 2013 at 9:09 PM, Weng Meiling
> wrote:
>> Hi Dhaval,
>>
>> what do you think about this patch ?
>>
>
> Sorry about this. It totally slipped my filters.
>
It doesn't
On 2013/9/27 10:13, Dhaval Giani wrote:
> On Thu, Sep 26, 2013 at 9:43 PM, Weng Meiling
> wrote:
>> On 2013/9/27 9:20, Dhaval Giani wrote:
>>> On Thu, Sep 26, 2013 at 9:05 PM, Weng Meiling
>>> wrote:
>>>> Hi Ivana,
>>>>
>>>&g
On 2013/9/27 9:20, Dhaval Giani wrote:
> On Thu, Sep 26, 2013 at 9:05 PM, Weng Meiling
> wrote:
>> Hi Ivana,
>>
>>what do you think about this question ?
>>
>> Thanks!
>> Weng Meiling
>>
>> On 2013/9/3 9:25, Weng Meiling wrote:
>
Hi Dhaval,
what do you think about this patch ?
Thanks!
Weng Meiling
On 2013/9/10 16:12, Ivana Varekova wrote:
> Hello,
> I'm not sure about this change.
>>From my point of view cgconfig should work, when the file is empty (it is
>>equivalent to file which
Hi Ivana,
what do you think about this question ?
Thanks!
Weng Meiling
On 2013/9/3 9:25, Weng Meiling wrote:
> Hi Ivana,
>
> what do you think about this question ?
>
> Thanks!
> Weng Meiling
>
> On 2013/8/9 9:54, Weng Meiling wrote:
>>
>>
Hi Ivana,
what do you think about this patch ?
Thanks!
Weng Meiling
Original Message
Subject: [PATCH] cgdelete:fix the display problem
Date: Sat, 14 Sep 2013 15:15:04 +0800
From: Weng Meiling
CC: Ivana Varekova , Huang Qiang
, Libcg Development list
, libo chen
From: Weng Meiling
When setting the cgconfig.conf as follow:
mount {
memory = /cgroup/memory;
}
group . {
perm {
task {
gid = w00227741;
uid = w00227741;
fperm =755
From: Weng Meiling
When multi subsystems mounted on one place like this:
# lssubsys -m
cpu,cpuacct /cgroup/cpu
# lscgroup
cpu,cpuacct:/
cpu,cpuacct:/test
if we delete the cgroup with the cgdelete -g, and specifying multi
controllers like this:
# cgdelete -g cpu,cpuacct:test
or
# cgdelete
On 2013/9/10 15:23, Weng Meiling wrote:
> On 2013/9/10 15:00, Ivana Varekova wrote:
>> Hello,
>> it is a bug. Error message should not be returned. Do you want to create a
>> patch for this problem? If not I will write it to my TODO list, but it is
>> really full now.
o it and send the patch later.
Thanks!
Weng Meiling
> - Original Message -----
>> From: "Weng Meiling"
>> To: "Libcg Development list"
>> Sent: Friday, August 30, 2013 10:02:36 AM
>> Subject: [Libcg-devel] question about cgdelete
>>
>>
On 2013/9/4 20:25, Dhaval Giani wrote:
> On Tue, Sep 3, 2013 at 9:43 PM, Weng Meiling
> wrote:
>> On 2013/9/3 20:05, Dhaval Giani wrote:
>>> On Fri, Aug 2, 2013 at 4:24 AM, Weng Meiling
>>> wrote:
>>>>
>>>> From: Weng Meiling
>>>
On 2013/9/3 20:05, Dhaval Giani wrote:
> On Fri, Aug 2, 2013 at 4:24 AM, Weng Meiling
> wrote:
>>
>> From: Weng Meiling
>>
>> When we use cgclear to unload the cgroup filesystem, we still can
>> see the umounted cgroup by using mount command. So fix the p
On 2013/9/3 9:23, Weng Meiling wrote:
> Hi Ivana,
>
> what do you think about this patch ?
>
> Thanks!
> Weng Meiling
>
>
sorry, please review the second version:
http://sourceforge.net/mailarchive/forum.php?thread_name=52022AB3.20509%40huawei.com&forum_nam
Hi Ivana,
what do you think about this question ?
Weng Meiling
Thanks!
On 2013/8/2 16:24, Weng Meiling wrote:
>
> From: Weng Meiling
>
> When we use cgclear to unload the cgroup filesystem, we still can
> see the umounted cgroup by using mount command. So fix the pr
From: Weng Meiling
Unloading the cgroup filesystem with cgclear -l config, if there is
an umount subsystem in the mount section. The error message is
"cgclear failed with Success". This is broken. So fix it, replace
the error message with "cgclear failed with Cgroup invalid op
Hi Ivana,
what do you think about this question ?
Thanks!
Weng Meiling
On 2013/8/30 16:02, Weng Meiling wrote:
> Hi,
>
> When multi subsystems mounted on one place like this:
>
> # lssubsys -m
> cpu,cpuacct /cgroup/cpu
>
> # lscgroup
> cpu,cpuacct:/
>
Hi Ivana,
what do you think about this patch ?
Thanks!
Weng Meiling
On 2013/8/30 14:50, Weng Meiling wrote:
> From: Weng Meiling
>
> cgconfig service could start successfully when /etc/cgconfig.conf
> only contains comments. But failed when /etc/cgconfig.conf is empty
Hi Ivana,
what do you think about this question ?
Thanks!
Weng Meiling
On 2013/8/14 14:37, Weng Meiling wrote:
> On 2013/8/14 11:32, Dhaval Giani wrote:
>> On Tue, Aug 13, 2013 at 10:07 PM, Weng Meiling
>> wrote:
>>> On 2013/8/14 4:38, Dhaval Giani wrote:
>&g
Hi Ivana,
what do you think about this question ?
Thanks!
Weng Meiling
On 2013/8/9 9:54, Weng Meiling wrote:
>
> From: Weng Meiling
>
> Any line starting with '#' in cgconfig.conf is considered a comment
> and is ignored. But in cgrules.conf, no matter whe
Hi Ivana,
what do you think about this patch ?
Thanks!
Weng Meiling
On 2013/8/6 15:39, Weng Meiling wrote:
>
> From: Weng Meiling
>
> If we mount the cgroup without the name cgroup like the following:
>
> #mount
> cpu on /cgroup/cpu type cgroup (rw,cpu)
>
n one place?
Thanks!
Weng Meiling
--
Learn the latest--Visual Studio 2012, SharePoint 2013, SQL 2012, more!
Discover the easy way to master current and previous Microsoft technologies
and advance your career. Get an incred
From: Weng Meiling
cgconfig service could start successfully when /etc/cgconfig.conf
only contains comments. But failed when /etc/cgconfig.conf is empty
or missing. The behavior of cgconfig service should be consistent
with cgred service.
This patch allows the cgconfig service start even when
On 2013/8/22 20:38, Dhaval Giani wrote:
> On Thu, Aug 22, 2013 at 8:32 AM, Ivana Varekova wrote:
>>
>>
>> - Original Message -
>>> From: "Weng Meiling"
>>> To: libcg-devel@lists.sourceforge.net
>>> Cc: "libo chen"
&g
On 2013/8/14 11:32, Dhaval Giani wrote:
> On Tue, Aug 13, 2013 at 10:07 PM, Weng Meiling
> wrote:
>> On 2013/8/14 4:38, Dhaval Giani wrote:
>>> Hi,
>>>
>>> On Tue, Aug 13, 2013 at 6:24 AM, Weng Meiling
>>> wrote:
>>>>
>>>>
On 2013/8/14 4:38, Dhaval Giani wrote:
> Hi,
>
> On Tue, Aug 13, 2013 at 6:24 AM, Weng Meiling
> wrote:
>>
>> From: Weng Meiling
>>
>> When setting the value of cgroup's control file by cgset -r, it's more
>> appositely to yield an error wh
From: Weng Meiling
When setting the value of cgroup's control file by cgset -r, it's more
appositely to yield an error when user try to set a read-only file or
set a file not existed. So fix this by copying the dirty value in function
cgroup_copy_controller_values().
Signed-of
From: Weng Meiling
Any line starting with '#' in cgconfig.conf is considered a comment
and is ignored. But in cgrules.conf, no matter where the character
'#' is, it will be replace with '\0'. So the similar rules get
different results as following:
the rule:
*%
From: Weng Meiling
If we mount the cgroup without the name cgroup like the following:
cpu on /cgroup/cpu type cgroup (rw,cpu)
the cgred service will start failed, because it just check the name.
So changing cgred to check the mounted system's type instead of the name.
v1 -> v2
Using t
From: Weng Meiling
If we mount the cgroup without the name cgroup like the following:
#mount
cpu on /cgroup/cpu type cgroup (rw,cpu)
the cgred service will start failed, because it just check the name.
So changing cgred to check the mounted system's type instead of the name.
Signed-o
From: Weng Meiling
When we use memory subsystem, the memory.oom_control value we get like
this:
oom_kill_disable 0
under_oom 0
but we set the value just by writing a value(0 or 1) like this:
#echo 1 > memory.oom_control
so setting the value memory.oom_control by cgset --copy-from or
From: Weng Meiling
When we use cgclear to unload the cgroup filesystem, we still can
see the umounted cgroup by using mount command. So fix the problem
by call system.
From: Weng Meiling
---
src/config.c | 18 --
1 files changed, 16 insertions(+), 2 deletions(-)
diff --git
Hi, when I test cgclear command with -l, I got a error report like this:
"cgclear failed with Success"
the following is my operation:
# lssubsys -m
cpu /cgroup/cpu
# lscgroup
cpu:/
cpu:/test
# cgclear -l config
cgclear failed with Success
the config context is:
# cat config
# Configuration fi
From: Weng Meiling
It's more readable and friendly to use the detailed error
message instead of error number as other places.
Signed-off-by: Weng Meiling
---
src/tools/cgclassify.c |3 ++-
1 files changed, 2 insertions(+), 1 deletions(-)
diff --git a/src/tools/cgclassify.c b/src/
From: Weng Meiling
It's more readable and friendly to use the enum value
instead of number.
Signed-off-by:Weng Meiling
---
src/tools/cgget.c | 16
1 file changed, 8 insertions(+), 8 deletions(-)
diff --git a/src/tools/cgget.c b/src/tools/cgget.c
index 49a447c..e1e1d60 1
Hi,
When we use lscgroup with -g to specify more then one controllers,
the result just display one controllers, such as:
# lssubsys -m
cpuset /cgroup/cpuset
cpu /cgroup/cpu
# lscgroup
cpu:/
cpu:/test
cpuset:/
cpuset:/test
# lscgroup -g cpu,cpuset:test
cpu:/test/
there is no message about cpuse
From: Weng Meiling
Date: Mon, 15 Jul 2013 04:08:20 -0400
Subject: [PATCH] cgsnapshot: fix the display message of the given controllers
when we use cgsnapshot with the given controller, the result always displays
the last mounted controller which is not specified, so fix it.
Example:
# lssubsys
On 2013/6/21 9:49, Weng Meiling wrote:
>
> From: Weng Meiling
>
ping
When use cgsnapshot with -f to save current configuration of cgroups,
the mount section is missing, so fix it.
the example as following:
# cgsnapshot -s -f config
cpu = /cgroup/cpu;
# cat config
# Configura
From: Weng Meiling
Signed-off-by: Weng Meiling
---
src/tools/cgsnapshot.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/src/tools/cgsnapshot.c b/src/tools/cgsnapshot.c
index 324f93b..8217798 100644
--- a/src/tools/cgsnapshot.c
+++ b/src/tools/cgsnapshot.c
@@ -610,9
79 matches
Mail list logo
