https rewrite problems

[gambolputty]

Code: Select all

[root@plnagiosv1 log]# ls -ld /usr/local/nagios
drwxr-xr-x 8 root root 4096 Mar  7 10:02 /usr/local/nagios
[root@plnagiosv1 log]# ls -ld /usr/local/nagios/var
drwxrwxr-x 6 nagios nagios 4096 Mar 15 08:28 /usr/local/nagios/var
[root@plnagiosv1 log]# ls -la /usr/local/nagios/var
total 104
drwxrwxr-x 6 nagios nagios  4096 Mar 15 08:28 .
drwxr-xr-x 8 root   root    4096 Mar  7 10:02 ..
drwxrwxr-x 2 nagios nagios  4096 Mar 15 00:00 archives
-rw-rw-r-- 1 nagios nagios     0 Mar 15 08:28 host-perfdata
-rw-r--r-- 1 nagios nagios     6 Mar 15 07:52 nagios.lock
-rw-r--r-- 1 nagios nagios  2269 Mar 15 07:52 nagios.log
-rw-r--r-- 1 nagios nagios     6 Mar  7 10:06 ndo2db.lock
-rw-rw-r-- 1 nagios nagios     0 Mar 15 07:52 ndomod.tmp
srwxr-xr-x 1 nagios nagios     0 Mar  7 10:06 ndo.sock
-rw-r--r-- 1 nagios nagios   235 Mar  7 10:06 npcd.log
-rw-r--r-- 1 nagios nagios 28437 Mar 15 07:52 objects.cache
-rw------- 1 nagios nagios 14078 Mar 15 07:52 retention.dat
drwxrwsr-x 2 nagios nagcmd  4096 Mar 15 07:52 rw
-rw-rw-r-- 1 nagios nagios   297 Mar 15 08:28 service-perfdata
drwxr-xr-x 5 root   root    4096 Mar  7 10:02 spool
drwxr-xr-x 2 nagios nagios  4096 Mar  7 10:03 stats
-rw-rw-r-- 1 nagios nagios 14707 Mar 15 08:28 status.dat
[root@plnagiosv1 log]#

I ran the XI install script as root, and haven't manually effected any permissions changes on these directories. Should /usr/local/nagios and /usr/local/nagios/var/spool be owned by root? This was installed from the fullinstall script** within xi-2012r1.6.tar.gz.

**Since we're on this topic, I may have found a small bug in the fullinstall script. When I first ran it, the script failed after churning for about 10 minutes. The step that failed was toward the end; there is a command that attempts to restart the httpd service. I didn't have httpd running (following the fresh, basic install recommendation in the install docs), and so the fullinstall script failed. I had a VM snapshot just before I ran that install script, so I reverted and ran it again- same problem. One more revert, and this time I started httpd just before running the fullinstall script, and it worked. It seems that the fullinstall script doesn't handle cases where httpd is not already running very well.

[abrist]

If you had apache already on the machine before running the fullinstall script, you did not start with a minimal cent or redhat install. This type of failure is to be expected as the install script expects certain conditions about the OS> Mainly, that it is a fresh, new *minimal* install of Cent and redhat. Pre-existing installs with a selection of non-minimal packages may cause the install script to fail.