Segmentation Fault on config verification
Mike Lindsey
mike-nagios at 5dninja.net
Mon Oct 22 21:17:42 CEST 2012
Seeing this on 3.3.1, and 3.4.1. Tried to reproduce with 4, but can't
build from the current git repository.
Migrating from obj_file to obj_dir style nagios.cfg, and on validation
of my Master configuration I get a Segmentation fault, that looks to be
coming right after Nagios closes nagios.cfg.
The same format of configuration, generated from the same script works
fine for poller nodes. The main differences in the poller node
configuration is size, no escalations, and no dependencies.
The end of the truss output:
mmap(0x0,783,PROT_READ,MAP_PRIVATE,5,0x0) = 34365812736 (0x8005cb000)
munmap(0x8005cb000,783) = 0 (0x0)
close(5) = 0 (0x0)
stat("/usr/local/ironport/akeos/bin/tmp/ops-mon-nagios1.vega/timeperiods/workhours.cfg",{
mode=-rw-r--r-- ,inode=829664,size=389,blksize=4096 }) = 0 (0x0)
open("/usr/local/ironport/akeos/bin/tmp/ops-mon-nagios1.vega/timeperiods/workhours.cfg",O_RDONLY,00)
= 5 (0x5)
fstat(5,{ mode=-rw-r--r-- ,inode=829664,size=389,blksize=4096 }) = 0 (0x0)
mmap(0x0,389,PROT_READ,MAP_PRIVATE,5,0x0) = 34365812736 (0x8005cb000)
munmap(0x8005cb000,389) = 0 (0x0)
close(5) = 0 (0x0)
getdirentries(0x4,0x800d27000,0x1000,0x800d15668,0xffffffff80aece00,0x7fffffffe410)
= 0 (0x0)
lseek(4,0x0,SEEK_SET) = 0 (0x0)
close(4) = 0 (0x0)
munmap(0x8005c9000,4444) = 0 (0x0)
close(3) = 0 (0x0)
SIGNAL 11 (SIGSEGV)
process exit, rval = 0
I'm digging into the source, but if anyone has any ideas, I'm ears.
--
Mike Lindsey
------------------------------------------------------------------------------
Everyone hates slow websites. So do we.
Make your web apps faster with AppDynamics
Download AppDynamics Lite for free today:
http://p.sf.net/sfu/appdyn_sfd2d_oct
_______________________________________________
Nagios-users mailing list
Nagios-users at 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
More information about the Users
mailing list