more and more cmd.php and eventually apache segfault..

[brunis]

here's some ps output:

28521 ? S 0:00 CROND
28522 ? S 0:00 /bin/bash -c php /var/www/cacti/cmd.php > /dev/null 2
28523 ? S 0:00 php /var/www/cacti/cmd.php
28531 ? S 0:00 perl /var/www/cacti/scripts/ping.pl 2 loopback0.gw4.p
28532 ? S 0:00 sh -c ping loopback0.gw4.phl1.alter.net -c 2 |grep ro
28533 ? S 0:00 ping loopback0.gw4.phl1.alter.net -c 2
28534 ? S 0:00 grep round-trip
28535 ? S 0:00 awk {print $4 }
28536 ? S 0:00 awk -F / {print $1 }
28537 ? S 0:00 grep -v Warning
6577 ? S 0:00 CROND
6578 ? S 0:00 /bin/bash -c php /var/www/cacti/cmd.php > /dev/null 2
6579 ? S 0:00 php /var/www/cacti/cmd.php
6587 ? S 0:00 perl /var/www/cacti/scripts/ping.pl 2 loopback0.gw4.p
6588 ? S 0:00 sh -c ping loopback0.gw4.phl1.alter.net -c 2 |grep ro
6589 ? S 0:00 ping loopback0.gw4.phl1.alter.net -c 2
6590 ? S 0:00 grep round-trip
6591 ? S 0:00 awk {print $4 }
6592 ? S 0:00 awk -F / {print $1 }
6593 ? S 0:00 grep -v Warning
6594 ? S 0:00 CROND
6596 ? S 0:00 /bin/bash -c php /var/www/cacti/cmd.php > /dev/null 2
6598 ? S 0:00 php /var/www/cacti/cmd.php
6607 ? S 0:00 perl /var/www/cacti/scripts/ping.pl 2 loopback0.gw4.p
6608 ? S 0:00 sh -c ping loopback0.gw4.phl1.alter.net -c 2 |grep ro
6609 ? S 0:00 ping loopback0.gw4.phl1.alter.net -c 2
6610 ? S 0:00 grep round-trip
6611 ? S 0:00 awk {print $4 }
6612 ? S 0:00 awk -F / {print $1 }
6613 ? S 0:00 grep -v Warning

why are there 3 instances of everything? this only happens after awhile.. and tomorrow there'll be 6 or 7 instances and eventually Apache exits with a segfault.. nasty stuff..

i don't know how i can help debugging this problem, please let me know.

[wesphillips]

I saw the same thing once on my systems( I am running cacti on two servers, for High Availability, with piranha....pretty neat, once you get it working). With a default install, there is an active ping to a uu.net router, that apparently doesn't respond anymore. If something happens in your system that causes the ping to that router to hang, all pings after that hang as well. The workaround that I used was to disable the ping data source, then restart the box(it's easier than trying to kill alll those processes).

[raX]

I had this exact problem when upgrading to Redhat 7.3 because of the altered ping command. To fix it, follow this post:

http://www.raxnet.net/board/viewtopic.php?t=442

Also, on a side note I believe that 'loopback0.gw4.phl1.alter.net' still reponds to ICMP requests:

[iberry@radish iberry]$ ping loopback0.gw4.phl1.alter.net
PING loopback0.gw4.phl1.alter.net (137.39.7.9) from 192.168.1.101 : 56(84) bytes of data.
64 bytes from Loopback0.GW4.PHL1.ALTER.NET (137.39.7.9): icmp_seq=1 ttl=252 time=32.5 ms

-Ian