When monitoring ESX hosts using the check_vmware_esx plugin we started noticing a lot of “script time out” errors and the checks going to “Unkown” state even know the default timeout is 90s.

For us the problem seems to be solved by using the sessionfile .

For other I saw a nice post about change the Perl library check it out at:

check_vmware_esx slow

time /usr/lib/centreon/plugins/check_vmware_esx -f /usr/lib/centreon/plugins/.vmconf -H ivc001 –name=’BLACACA’ –select=mem –subselect=usage -w 80 -c 90
OK: mem usage=10.99%|’mem_usage’=10.99%;80;90;;

real 0m25.212s
user 0m2.222s
sys 0m0.107s


time /usr/lib/centreon/plugins/check_vmware_esx –sessionfiledir=/usr/lib/centreon/plugins/tmps –sessionfile=”$HOSTNAME$” -f /usr/lib/centreon/plugins/.vmconf -H ivc001 –name=’BLACACA’ –select=mem –subselect=usage -w 80 -c 90
OK: mem usage=6.99%|’mem_usage’=6.99%;80;90;;

real 0m5.695s
user 0m1.280s
sys 0m0.069s

Add to the check_vmware Command Definition to:
–sessionfiledir=/usr/lib/centreon/plugins/tmps –sessionfile=$HOSTNAME$

Tags: , , , ,

Leave a Reply

Your email address will not be published. Required fields are marked *