Hi, On Tue, 2006-06-20 at 11:51 +0100, Alex Moore wrote: > Hello, > > Nagios mostly works just fine, but I have found that check_ping will > always crash when it runs against a couple of particular hosts. > > These hosts have IP addresses: 172.25.1.99 and 172.25.1.98. Check ping > succeeds for all other hosts, but just not these two. If I use the > hostnames of these machines instead of the IP addresses in the "address" > field of the host definitions then it works fine as well. I can't see > any logical reason why check_ping should crash on these two IP addresses > but nothing else. This is an example of running check_ping from the > command line: > > # /usr/local/libexec/nagios/check_ping -H 172.25.1.98 -w 100.0,20% -c > 500.0,60% -p 5 > Segmentation fault (core dumped) > > But most other addresses work: > > # /usr/local/libexec/nagios/check_ping -H 172.19.15.58 -w 100.0,20% -c > 500.0,60% -p 5 > PING OK - Packet loss = 0%, RTA = 0.44 ms > # /usr/local/libexec/nagios/check_ping -H 172.25.1.93 -w 100.0,20% -c > 500.0,60% -p 5 > PING WARNING - Packet loss = 0%, RTA = 170.30 ms > > Note that this is reproducible 100% of the time. And the segfault does > not occur when using the hostname instead of the IP address. > > I tried a tcpdump of check_ping when it succeeds and when it fails, and
try to strace check_ping (strace -f -o strace.output /usr/local/libexec/nagios/check_ping -H 172.25.1.98 -w 100.0,20% -c 500.0,60% -p 5). Maybe this will give and idea of what's happening. I'm guessing that's some kind of issue with libresolve. HTH, Marco Ramos > comparing the two. In both cases it seems to try an AAAA DNS lookup on > the IP address (which fails, of course, since they are not IPv6 > addresses). In the working scenario, check_ping then sends out its ICMP > packet direct to the IP address and receives a reply, etc. In the > segfault scenario check_ping never even sends out an ICMP echo request > at all. It just waits for a few seconds and then segfaults. > > Anyone know why this might be happening? Obviously, using the hostname > instead of IP address for these 2 machines is a reasonable workaround, > but it seems very strange! > > Note: I am using the OpenBSD 3.9 stable packages for nagios, which are > version 2.0rc2. Was this a known bug in that version which has since > been fixed? > > > _______________________________________________ > Nagios-users mailing list > Nagios-users@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/nagios-users > ::: Please include Nagios version, plugin version (-v) and OS when reporting > any issue. > ::: Messages without supporting info will risk being sent to /dev/null > _______________________________________________ Nagios-users mailing list Nagios-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/nagios-users ::: Please include Nagios version, plugin version (-v) and OS when reporting any issue. ::: Messages without supporting info will risk being sent to /dev/null
