YaBB SE Version: 1.5.1 RC1
PHP Version: 4.3.2
MySQL Version: 4.0.x
Server Platform: Unix, Linux, or BSD
Link to Forum: http://vegweb.com/yabbse/index.phpProblem Description:Hi,
I posted this question about runaway processes back in January but didn't know enough about what I was doing to respond with the right information.
I swiched from YaBB Gold to (ultimately) YaBB 1.5.1. After the switch, I noticed in "top" that there were a lot of httpd processes that kept running and after a bunch of those processes were created and still running, everything on my site would slow way down.
I set up a cron job to kill the runaway processes every five minutes, and that seemed to work as a temporary fix. I thought the problem was related to something I did wrong when installing MySQL and PHP. Also, I orginally started with YaBBSE 1.5.0 and had hoped that the upgrade to 1.5.1 (back when it came out) would fix the problem.
I've since moved to a virtual private server where everything was already set up, and all I did was move the site over, including the MySQL database. Everything is working fine, but I still get the runaway httpd processes.
How can I isolate what the problem is so that I can fix it? Is there an optimization script I could run?
Here is a copy of my information from top:
7:21am up 7 days, 21:42, 1 user, load average: 1.75, 1.64, 1.73
83 processes: 79 sleeping, 3 running, 1 zombie, 0 stopped
CPU0 states: 3.2% user, 3.2% system, 0.0% nice, 92.3% idle
CPU1 states: 3.0% user, 1.2% system, 0.0% nice, 95.0% idle
CPU2 states: 2.2% user, 1.1% system, 0.0% nice, 95.2% idle
CPU3 states: 3.0% user, 1.0% system, 0.0% nice, 95.2% idle
Mem: 8243116K av, 8001608K used, 241508K free, 0K shrd, 124496K buff
Swap: 16776424K av, 2322536K used, 14453888K free 3587976K cached
PID USER PRI NI SIZE RSS SHARE STAT %CPU %MEM TIME COMMAND
8250 nobody 18 0 5168 4884 2672 S 2.9 0.0 0:00 httpd
10256 nobody 15 0 3752 3444 2604 S 2.5 0.0 0:00 httpd
4554 nobody 9 0 5204 4920 2676 S 1.7 0.0 0:08 httpd
2647 nobody 12 0 5772 5496 2776 S 0.7 0.0 0:10 httpd
3159 nobody 10 0 5164 4880 2672 S 0.7 0.0 0:07 httpd
902 nobody 11 0 5528 5248 3032 R 0.3 0.0 0:01 httpd
12447 root 12 0 1044 1044 824 R 0.3 0.0 0:00 top
2644 nobody 9 0 5228 4944 2680 S 0.1 0.0 0:08 httpd
2722 nobody 10 0 5656 5376 3036 S 0.1 0.0 0:09 httpd
3225 nobody 10 0 5644 5372 2776 S 0.1 0.0 0:09 httpd
30586 nobody 10 0 3784 3476 2596 R 0.1 0.0 0:01 httpd
For example, by the time that the httpd processes that are 0:08 or 0:10 all get to 0:30 or more my load average will be 15 to 20. If I kill them, the load average goes back down to 1 or .80 or less.
Server Specs:
SuperMicro Dual Xeon 2.66Ghz
8 GB Registered ECC DDR RAM
10x73GB U320 SCSI drives in Hardware RAID 5
Hot-Spare drives
Hot-Swap drives and fans
Redundant Hot-Swap power supplies (3 total)
Dual 10/100/1000 Network Interfaces
Product Specifications:
RedHat Linux
Equal-share CPU usage
2 GB Storage
100 GB Monthly Transfer
512 MB RAM burst, guaranteed 128 MB minimum
I will be upgrading soon to a new server with 1024 MB RAM (1 GB) burst, guaranteed 256 MB minimum.
Any help would really be appreciated.
Yvette