Home
| Calendar
| Mail Lists
| List Archives
| Desktop SIG
| Hardware Hacking SIG
Wiki | Flickr | PicasaWeb | Video | Maps & Directions | Installfests | Keysignings Linux Cafe | Meeting Notes | Linux Links | Bling | About BLU |
Hello, This is my first post on this list, although I've attended BLU meetings off and on since 99, including last night's. I'm trying to help a client with a (C++ based) web application server on Redhat 7.2 and Apache 1.3.20 and the client has encountered a situation that I've not seen before. The app server has a mechanism to restart itself if the request pool stacks up with hung threads, and it logs when an attempt to do so is made. The app server is reporting that is restarting itself based on this mechanism, however the client finds that the same processes with same process ids are continuing to run though. The part that I've never seen is that while logged in as root the client reports that he cannot kill the processes (after already having tried the manual stop script). The client reports that after determining the parent process and then try to kill it with a kill -9 <pid>, kill -15 <pid>, or a killall <processname> that the processes do not die. My question is how can it be that root cannot kill a process? Conceivably, the app server might be making calls to custom-built, external C++ routines though the app server API to do so, and maybe that C++ routine could be somehow nicing the priority of the app server pid to set it higher than the kernel pid, but that's a real stretch and the client says they are not using such C++ routines. Any suggestions or explanations would be very much appreciated. Thanks, Steven Erat __________________________________ Do you Yahoo!? SBC Yahoo! DSL - Now only $29.95 per month! http://sbc.yahoo.com
BLU is a member of BostonUserGroups | |
We also thank MIT for the use of their facilities. |