I am running:
Fedora FC8 + all yum updates.
ZM = 1.23.1 built from Source.
Kernel 2.6.23.14-107.fc8 #1 SMP Mon Jan 14 22:07:11 EST 2008 x86_64 x86_64 x86_64 GNU/Linux
My shared memory allocation is:
[root@ringtail ZoneMinder-1.23.1]# sysctl -A |grep shm
kernel.shmmax = 33554432
kernel.shmall = 134217728
kernel.shmmni = 4096
vm.hugetlb_shm_group = 0
The only open socket is:
[root@ringtail ZoneMinder-1.23.1]# ls -l /tmp/*sock
srwxr-xr-x 1 apache apache 0 2008-02-09 12:03 /tmp/zmdc.sock
I ONLY start/stop ZM from the web page, and my Web User:Group is configured as apache.
My config commands were:
./configure --with-libarch=lib64 --with-extralibs=-lmp3lame --with-mysql=/usr --with-ffmpeg=/usr/local --with-webdir=/var/www/html --with-cgidir=/var/www/cgi-bin --with-webuser=apache --with-webgroup=apache --enable-crashtrace=yes --enable-debug=yes ZM_DB_HOST=ringtail.wildpossum.com ZM_DB_NAME=zm ZM_DB_USER=zmuser ZM_DB_PASS=XXXX
Make, Make install where built and Mysql DB setup - correctly.
I have set DEBUG = 3 for _zmc and ZM-Logs say:
==> /Events/ZM-Logs/zmaudit.log <==
==> /Events/ZM-Logs/zmdc.log <02> /Events/ZM-Logs/zm-debug.log <==
==> /Events/ZM-Logs/zmfilter.log <02> /Events/ZM-Logs/zmpkg.log <02> /Events/ZM-Logs/zmupdate.log <02> /Events/ZM-Logs/zmwatch.log <==
02/09/08 12:03:13.334309 zmwatch[7985].INF [Watchdog starting]
02/09/08 12:03:13.335795 zmwatch[7985].INF [Watchdog pausing for 30 seconds]
02/09/08 12:03:43.339382 zmwatch[7985].ERR [Can't get shared memory id '7a6d0001', 1: No such file or directory]
02/09/08 12:03:43.340441 zmwatch[7985].INF [Restarting capture daemon for Front-Camera, shared memory not valid]
top - 12:32:00 up 1:11, 4 users, load average: 2.00, 2.02, 1.68
Tasks: 125 total, 1 running, 124 sleeping, 0 stopped, 0 zombie
Cpu(s): 0.1%us, 0.1%sy, 0.0%ni, 99.9%id, 0.0%wa, 0.0%hi, 0.0%si, 0.0%st
Mem: 897608k total, 544448k used, 353160k free, 17172k buffers
Swap: 3140656k total, 0k used, 3140656k free, 277500k cached
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
2300 apache 20 0 272m 11m 3160 S 0.0 1.4 0:00.32 /usr/sbin/httpd
2301 apache 20 0 272m 11m 3196 S 0.0 1.4 0:00.37 /usr/sbin/httpd
2302 apache 20 0 272m 12m 3264 S 0.0 1.4 0:00.38 /usr/sbin/httpd
2303 apache 20 0 272m 11m 3196 S 0.0 1.4 0:00.26 /usr/sbin/httpd
2304 apache 20 0 272m 11m 3156 S 0.0 1.4 0:00.32 /usr/sbin/httpd
2305 apache 20 0 272m 11m 2960 S 0.0 1.3 0:00.32 /usr/sbin/httpd
2306 apache 20 0 272m 11m 3224 S 0.0 1.4 0:00.49 /usr/sbin/httpd
2307 apache 20 0 272m 11m 3084 S 0.0 1.4 0:00.27 /usr/sbin/httpd
7915 apache 20 0 84724 9960 1416 S 0.0 1.1 0:00.07 /usr/bin/perl -wT /usr/local/bin/zmdc.pl startup
7956 apache 20 0 220m 3564 2764 D 0.0 0.4 0:00.02 /usr/local/bin/zmc -d /dev/video0
7960 apache 20 0 220m 3564 2760 D 0.0 0.4 0:00.02 /usr/local/bin/zmc -d /dev/video1
7970 apache 20 0 90548 16m 2996 S 0.0 1.9 0:00.42 /usr/bin/perl -wT /usr/local/bin/zmfilter.pl
7974 apache 20 0 85616 12m 2940 S 0.0 1.4 0:00.23 /usr/bin/perl -wT /usr/local/bin/zmaudit.pl -c
7985 apache 20 0 84472 10m 2996 S 0.0 1.3 0:00.13 /usr/bin/perl -wT /usr/local/bin/zmwatch.pl
7990 apache 20 0 96456 18m 3108 S 0.0 2.1 0:00.27 /usr/bin/perl -w /usr/local/bin/zmupdate.pl -c
8184 apache 20 0 84724 11m 2872 S 0.0 1.3 0:00.23 /usr/bin/perl -wT /usr/local/bin/zmdc.pl restart zmc
8383 apache 20 0 84724 11m 2872 S 0.0 1.3 0:00.24 /usr/bin/perl -wT /usr/local/bin/zmdc.pl check zma -m
8394 apache 20 0 269m 8740 2092 S 0.0 1.0 0:00.00 /usr/sbin/httpd
8395 apache 20 0 269m 7460 916 S 0.0 0.8 0:00.00 /usr/sbin/httpd
8396 apache 20 0 269m 8740 2092 S 0.0 1.0 0:00.00 /usr/sbin/httpd
8397 apache 20 0 269m 7456 912 S 0.0 0.8 0:00.00 /usr/sbin/httpd
8398 apache 20 0 269m 7456 912 S 0.0 0.8 0:00.00 /usr/sbin/httpd
8399 apache 20 0 269m 7456 912 S 0.0 0.8 0:00.00 /usr/sbin/httpd
8408 apache 20 0 269m 7456 912 S 0.0 0.8 0:00.00 /usr/sbin/httpd
You will notice from TOP that the zmdc.PL process is always generating more of itself [ONLY AFTER seeing problems with shared memory access], (probably because it is in SLEEP state and can't get out of it) and hence no one (including root) is able to shutdown ZM, only option left is to reboot the machine which isn't the best method. Any solution to this too?
Can anyone suggest further what may be the problem(s)?
I have compiled a list of things to check for, (including writing my very own source test for shared memory creation, allocation/deallocation which works OK) all from the last years postings and I can't see what else to try.

