On 04.01.2012 11:02, Mario Brandt wrote:
I figured out that using EnableMMAP off causes the L status. I tried
it in combination with EnableSendfile off and without.

By "without" you mean "EnableSendfile On" or just default? Note that the default for EnableSendfile is again "Off" in 2.4, so it is expected we don't see a difference between Off and default.

I hope that helps finding the error.

Thanks,

Rainer

On Sat, Dec 31, 2011 at 21:08, Steffen<i...@apachelounge.com>  wrote:
That's not expected. Can any Windows Seniors comment on that? Why could we
have the Logging status in server-status for a long time for the winnt
MPM?


This bug is still there in 2.3.16, have "L" entries for more then a week !
They are occupying  workers all the time, so the busyorkers are far high:

L__L_L___L____L_L______L___L_L_L__


Happy (2.4) new year to all,
Steffen

-----Original Message----- From: Rainer Jung
Sent: Tuesday, December 06, 2011 9:33 PM Newsgroups: gmane.comp.apache.devel
To: dev@httpd.apache.org
Subject: Re: Win 2.3.15 :: Server Status Entries

On 06.12.2011 11:56, Steffen wrote:

Those requests are no longer processed (status is "_"). The seconds
since is the time since this request was last processed, the column
after that shows how long processing took (here: "1" millisecond for
each of them).


There are also with status L, for example for almost a day sitting there:

0-0 2852 1/2920/2920 L 79507 49835 2893.9 114.05 114.05 80.91.181.75
www.apachelounge.com GET
/download/binaries/httpd-2.2.21-win32-x86-ssl.zip HTTP/1.0


That's not expected. Can any Windows Seniors comment on that? Why could
we have the Logging status in server-status for a long time for the
winnt MPM?

-----Original Message----- From: Rainer Jung
Sent: Tuesday, November 22, 2011 11:32 AM
To: dev@httpd.apache.org
Subject: Re: Win 2.3.15 :: Server Status Entries

On 22.11.2011 10:28, Steffen wrote:

Seeing a huge number of hanging entries in the Server Status,
already for 20 hours and looks they are staying there forever.

The requests are invalid, not sure since I do not keep the raw logs.

...
...
0-0 3800 0/177/177 _ 64980 1 0.0 0.09 0.09 94.76.244.212
www.familieland.com GET //phpMyAdmin/scripts/setup.php HTTP/1.1
0-0 3800 0/157/157 _ 69024 1 0.0 10.65 10.65 94.76.244.212
www.familieland.com GET //scripts/setup.php HTTP/1.1
0-0 3800 0/224/224 _ 69023 1 0.0 21.96 21.96 94.76.244.212
www.familieland.com GET //admin/pma/scripts/setup.php HTTP/1.1
....
....
etc. etc.

SS = 70878 seconds now and counting.


Those requests are no longer processed (status is "_"). The seconds
since is the time since this request was last processed, the column
after that shows how long processing took (here: "1" millisecond for
each of them).

It looks like you have spare slots that are occasionally used during
load spikes but are idle later for a long time. Don't know how the
Windows MPM decides which idle slot to use.

Regards,

Rainer



Reply via email to