Hi, I have following output from the mod_status. It shows 16 processes. Observe that only the five processes have PID's (MaxSpareServer is 5). Do the other ones exist too in the memory? Is it a bug? Srv PID Acc M CPU SS Req Conn Child Slot Client VHost Request 0-0 111402 0/75/75 W_ 10.99 0 0 0.0 0.28 0.28 10.32.6.13 myserver GET /server-status HTTP/1.1 1-0 94556 0/77/77 __ 9.14 69 16 0.0 0.17 0.17 myserver GET /server-status HTTP/1.1 2-0 84944 0/82/82 __ 10.90 43 1 0.0 0.19 0.19 10.32.6.13 myserver GET /oaa/images/error_logo.gif HTTP/1.1 3-0 100506 0/71/71 __ 9.40 31 0 0.0 0.16 0.16 10.32.6.13 myserver GET /oaa/images/1pxGrey.gif HTTP/1.1 4-0 71882 0/82/82 __ 8.44 69 6 0.0 0.35 0.35 myserver GET /server-status HTTP/1.1 5-0 - 0/0/17 ._ 0.05 269 0 0.0 0.00 0.17 10.32.6.13 myserver GET /oaa/images/go.gif HTTP/1.1 6-0 - 0/0/29 ._ 0.16 290 1 0.0 0.00 0.06 10.32.6.13 myserver GET /oaa/images/spacer.gif HTTP/1.1 7-0 - 0/0/30 ._ 0.13 293 0 0.0 0.00 0.14 10.32.6.13 myserver GET /oaa/images/go.gif HTTP/1.1 8-0 - 0/0/17 ._ 0.09 290 0 0.0 0.00 0.04 10.32.6.13 myserver GET /oaa/images/1pxGrey.gif HTTP/1.1 9-0 - 0/0/7 ._ 0.04 320 0 0.0 0.00 0.00 10.32.6.13 myserver GET /oaa/images/go.gif HTTP/1.1 10-0 - 0/0/24 ._ 0.15 297 1165 0.0 0.00 0.09 10.32.6.13 myserver GET /oaa/js/fidelityscripts.js HTTP/1.1 11-0 - 0/0/14 ._ 0.06 303 0 0.0 0.00 0.00 10.32.6.13 myserver GET /oaa/images/go.gif HTTP/1.1 12-0 - 0/0/18 ._ 0.08 297 0 0.0 0.00 0.11 10.32.6.13 myserver GET /oaa/images/1pxGrey.gif HTTP/1.1 13-0 - 0/0/18 ._ 0.10 297 0 0.0 0.00 0.09 10.32.6.13 myserver GET /oaa/images/go.gif HTTP/1.1 14-0 - 0/0/12 ._ 0.05 303 2 0.0 0.00 0.02 10.32.6.13 myserver GET /oaa/js/setDomain.js HTTP/1.1 15-0 - 0/0/1 ._ 0.02 311 94 0.0 0.00 0.000 10.32.6.13 myserver GET /oaa/js/setDomain.js HTTP/1.1 -----Original Message----- From: Mishra, Pawan Sent: Tuesday, September 13, 2005 10:58 PM To: 'users@xxxxxxxxxxxxxxxx' Subject: RE: [users@httpd] Httpd processes are not dying Thanks. I have checked to see that there is no duplicate entry for MaxSpareServers. -----Original Message----- From: Joshua Slive [mailto:jslive@xxxxxxxxx] Sent: Tuesday, September 13, 2005 10:52 PM To: users@xxxxxxxxxxxxxxxx Subject: Re: [users@httpd] Httpd processes are not dying On 9/13/05, Mishra, Pawan <Pawan.Mishra@xxxxxxx> wrote: > Hi Joshua, > > I am not sure you remember the problem I stated last time so I am > listing it here again. You had asked me to run mod_status. I have > attached the output of the mod_status too. > > Problem: The httpd processes keep getting accumulated in the memory and > are not dying timely as expected. This happens even when there is no use > of the application for hours. There are times when we see more that 70 > such processes even when application has not been used for 6-7 hours. > Since these processes donot dye in timely manner, they keep getting > accumulated till the MaxClient limit is reached and at that point I.H.S. > stops responding and thus site becomes down. We then need to restart the > I.H.S. server. We have been experiencing this problem since we migrated > to WAS 5.0.2 / I.H.S. 1.3.26.2 combination. Well, the server-status looks completely normal except that MaxSpareServers is not being honored. So the first thing I'd do is triple-check that you are editting the correct config-file, and that MaxSpareServers occurs only once in that config. Other than that, I'd really suggest reporting this to IBM, since it appears you are using an IBM-modified version of the server. Joshua. --------------------------------------------------------------------- The official User-To-User support forum of the Apache HTTP Server Project. See <URL:http://httpd.apache.org/userslist.html> for more info. To unsubscribe, e-mail: users-unsubscribe@xxxxxxxxxxxxxxxx " from the digest: users-digest-unsubscribe@xxxxxxxxxxxxxxxx For additional commands, e-mail: users-help@xxxxxxxxxxxxxxxx --------------------------------------------------------------------- The official User-To-User support forum of the Apache HTTP Server Project. See <URL:http://httpd.apache.org/userslist.html> for more info. To unsubscribe, e-mail: users-unsubscribe@xxxxxxxxxxxxxxxx " from the digest: users-digest-unsubscribe@xxxxxxxxxxxxxxxx For additional commands, e-mail: users-help@xxxxxxxxxxxxxxxx
|