[Wed Jan 08 18:50:07 2014] [emerg] (28)No space left on device: Couldn't create accept lock (/etc/httpd/logs/accept.lock.15449) (5)
This may happen when trying to restart Apache and you find it dies right after starting and check /var/log/httpd/error_logs.
What is the cause of this?
Check to see current Semaphores, if Apache has stopped it should have none. If you run the command below and see them you need to manually kill them.
ipcs -s
------ Semaphore Arrays --------
key semid owner perms nsems
0x00000000 0 apache 600 1
0x00000000 229377 apache 600 1
0x00000000 262146 apache 600 1
0x00000000 294915 apache 600 1
0x00000000 327684 apache 600 1
0x00000000 360453 apache 600 1
0x00000000 1671174 apache 600 1
0x00000000 1900551 apache 600 1
0x00000000 8781832 apache 600 1
0x00000000 8814601 apache 600 1
0x00000000 8847370 apache 600 1
0x00000000 8880139 apache 600 1
0x00000000 8912908 apache 600 1
0x00000000 8945677 apache 600 1
0x00000000 8978446 apache 600 1
0x00000000 13205519 apache 600 1
0x00000000 13238288 apache 600 1
0x00000000 13271057 apache 600 1
0x00000000 13303826 apache 600 1
0x00000000 13336595 apache 600 1
The bash script below will kill all the old semaphores and you should be able to start Apache succesfully after that.
for ipcs in `ipcs -s|awk '{print $2}'`;
do
echo "ipcs=$ipcs"
ipcrm -s $ipcs
done
emerg, couldn, etc, httpd, logs, wed, restart, apache, var, error_logs, disk, semaphores, manually, ipcs, semaphore, arrays, semid, perms, nsems, bash, succesfully, awk, echo, quot, ipcrm,