Mailing List Archive

[Bug 52914] On keep-alive on, after some time huge process load
https://issues.apache.org/bugzilla/show_bug.cgi?id=52914

azat <dohardgopro@gmail.com> changed:

What |Removed |Added
----------------------------------------------------------------------------
CC| |dohardgopro@gmail.com

--
Configure bugmail: https://issues.apache.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.

---------------------------------------------------------------------
To unsubscribe, e-mail: bugs-unsubscribe@httpd.apache.org
For additional commands, e-mail: bugs-help@httpd.apache.org
[Bug 52914] On keep-alive on, after some time huge process load [ In reply to ]
https://issues.apache.org/bugzilla/show_bug.cgi?id=52914

Eric Covener <covener@gmail.com> changed:

What |Removed |Added
----------------------------------------------------------------------------
Keywords| |PatchAvailable

--
Configure bugmail: https://issues.apache.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.

---------------------------------------------------------------------
To unsubscribe, e-mail: bugs-unsubscribe@httpd.apache.org
For additional commands, e-mail: bugs-help@httpd.apache.org
[Bug 52914] On keep-alive on, after some time huge process load [ In reply to ]
https://issues.apache.org/bugzilla/show_bug.cgi?id=52914

--- Comment #1 from Stefan Fritsch <sf@sfritsch.de> 2012-03-15 17:24:08 UTC ---
Created attachment 28474
--> https://issues.apache.org/bugzilla/attachment.cgi?id=28474
don't clobber input filter return value in mod_dump_io

I don't think your patch is correct. Can you try if my patch helps, too?

--
Configure bugmail: https://issues.apache.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.

---------------------------------------------------------------------
To unsubscribe, e-mail: bugs-unsubscribe@httpd.apache.org
For additional commands, e-mail: bugs-help@httpd.apache.org
[Bug 52914] On keep-alive on, after some time huge process load [ In reply to ]
https://issues.apache.org/bugzilla/show_bug.cgi?id=52914

--- Comment #2 from Stefan Fritsch <sf@sfritsch.de> 2012-03-15 17:39:42 UTC ---
(In reply to comment #1)
> Created attachment 28474 [details]
> don't clobber input filter return value in mod_dump_io
>
> I don't think your patch is correct. Can you try if my patch helps, too?

FWIW, this would mean that not loading mod_dump_io should also fix the issue.

--
Configure bugmail: https://issues.apache.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.

---------------------------------------------------------------------
To unsubscribe, e-mail: bugs-unsubscribe@httpd.apache.org
For additional commands, e-mail: bugs-help@httpd.apache.org
[Bug 52914] On keep-alive on, after some time huge process load [ In reply to ]
https://issues.apache.org/bugzilla/show_bug.cgi?id=52914

Stefan Fritsch <sf@sfritsch.de> changed:

What |Removed |Added
----------------------------------------------------------------------------
Keywords|PatchAvailable |FixedInTrunk
Component|Core |mod_dumpio
Version|2.4.1 |2.5-HEAD
OS/Version|Linux |All

--- Comment #3 from Stefan Fritsch <sf@sfritsch.de> 2012-03-16 18:54:06 UTC ---
I could reproduce the 100% CPU with POST requests if mod_reqtimeout is enabled
for bodies and DumpIOInput is on. This is fixed by my patch

Trunk: r1301111
2.4: r1301686
proposed for 2.2

--
Configure bugmail: https://issues.apache.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.

---------------------------------------------------------------------
To unsubscribe, e-mail: bugs-unsubscribe@httpd.apache.org
For additional commands, e-mail: bugs-help@httpd.apache.org
[Bug 52914] On keep-alive on, after some time huge process load [ In reply to ]
https://issues.apache.org/bugzilla/show_bug.cgi?id=52914

--- Comment #4 from azat <dohardgopro@gmail.com> 2012-03-19 08:22:50 UTC ---
Thanks, I'll see you commit, and seems that it can fix my problem.
I'll try later, and comment here about what I have

--
Configure bugmail: https://issues.apache.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.

---------------------------------------------------------------------
To unsubscribe, e-mail: bugs-unsubscribe@httpd.apache.org
For additional commands, e-mail: bugs-help@httpd.apache.org
[Bug 52914] On keep-alive on, after some time huge process load [ In reply to ]
https://issues.apache.org/bugzilla/show_bug.cgi?id=52914

--- Comment #5 from azat <dohardgopro@gmail.com> 2012-03-19 13:30:43 UTC ---
(In reply to comment #3)
> I could reproduce the 100% CPU with POST requests if mod_reqtimeout is enabled
> for bodies and DumpIOInput is on. This is fixed by my patch
>
> Trunk: r1301111
> 2.4: r1301686
> proposed for 2.2

I try reproduce this bug, on another machine
On both are debian, same version
Apache, have the same version (2.2.22)

But this bug, not reproduced (DumpIOInput on, mod_reqtimeout enabled, keepalive
on)
Trying using siege

Can you write here how you reproduce this bug?
Thanks.

--
Configure bugmail: https://issues.apache.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.

---------------------------------------------------------------------
To unsubscribe, e-mail: bugs-unsubscribe@httpd.apache.org
For additional commands, e-mail: bugs-help@httpd.apache.org
[Bug 52914] On keep-alive on, after some time huge process load [ In reply to ]
https://issues.apache.org/bugzilla/show_bug.cgi?id=52914

--- Comment #6 from Stefan Fritsch <sf@sfritsch.de> 2012-03-19 19:53:17 UTC ---
(In reply to comment #5)
> Can you write here how you reproduce this bug?
> Thanks.

I did a POST request with a Content-Length header, sent part (but not all) of
the request body, and then waited for mod_reqtimeout's body timeout to kick in.
Then the child would loop with 100% CPU.

Do you know if you had DumpIOInput enabled when you reported this issue? If you
are sure that you didn't have it enabled, you probably have hit a different
issue.

--
Configure bugmail: https://issues.apache.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.

---------------------------------------------------------------------
To unsubscribe, e-mail: bugs-unsubscribe@httpd.apache.org
For additional commands, e-mail: bugs-help@httpd.apache.org
[Bug 52914] On keep-alive on, after some time huge process load [ In reply to ]
https://issues.apache.org/bugzilla/show_bug.cgi?id=52914

--- Comment #7 from Stefan Fritsch <sf@sfritsch.de> 2012-03-19 20:02:11 UTC ---
(In reply to comment #6)
> Do you know if you had DumpIOInput enabled when you reported this issue? If you
> are sure that you didn't have it enabled, you probably have hit a different
> issue.

Ignore that, that was a stupid question. Your stack trace shows that you had
DumpIOInput enabled.

--
Configure bugmail: https://issues.apache.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.

---------------------------------------------------------------------
To unsubscribe, e-mail: bugs-unsubscribe@httpd.apache.org
For additional commands, e-mail: bugs-help@httpd.apache.org
[Bug 52914] On keep-alive on, after some time huge process load [ In reply to ]
https://issues.apache.org/bugzilla/show_bug.cgi?id=52914

azat <dohardgopro@gmail.com> changed:

What |Removed |Added
----------------------------------------------------------------------------
Status|NEW |RESOLVED
Resolution| |FIXED

--- Comment #8 from azat <dohardgopro@gmail.com> 2012-03-20 09:56:11 UTC ---
(In reply to comment #6)
> (In reply to comment #5)
> > Can you write here how you reproduce this bug?
> > Thanks.
>
> I did a POST request with a Content-Length header, sent part (but not all) of
> the request body, and then waited for mod_reqtimeout's body timeout to kick in.
> Then the child would loop with 100% CPU.
>
> Do you know if you had DumpIOInput enabled when you reported this issue? If you
> are sure that you didn't have it enabled, you probably have hit a different
> issue.

Thanks.
You patch fix problem with huge process load, for me.

--
Configure bugmail: https://issues.apache.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.

---------------------------------------------------------------------
To unsubscribe, e-mail: bugs-unsubscribe@httpd.apache.org
For additional commands, e-mail: bugs-help@httpd.apache.org