Easy Big-IP LTM backup

Recently we upgraded the sw on our bigips, and then all our custom local scripts were gone.
Thats why i wanna blog this easy bigip recipe in case we are stupid enough to do this again.
This is the backup procedure for BIG-IP 9.4.x

/usr/local/bin/bigipbackup.sh: here

/etc/cron.d/backup:
0 21 * * * root /usr/local/bin/bipipbackup.sh > /dev/null

Tags: ,

Monitoring F5 BigIP TMM Utilization with Nagios and snmp

I came across a post on the Cacti-forums where a guy has posted a perl script for finding the F5 BigIP TMM Utilization via the commandline, so I modified it a bit, and now it works as a nagios-plugin.
I’ve also added performancedata to it, for usage with graphing tools like for instance pnp4nagios.

Here’s the complete modified script : check_tmm_utilization

Then you can simply add 4 lines to your commands.cfg file in Nagios.

define command{
command_name check_tmm_utilization
command_line $USER1$/check_tmm_utilization $HOSTADDRESS$ yoursnmpcommunitystring $ARG1$ $ARG2$
}

…and finally update services with something like :

define service{
use generic-service
host_name bigip-box
service_description CHECK_TMM_UTILIZATION
is_volatile 0
max_check_attempts 3
normal_check_interval 5
retry_check_interval 1
contact_groups linux-admins
notification_period 24x7
notification_options c,w,r
process_perf_data 1
check_command check_tmm_utilization!75!90
}

check config and restart

woohooo

Tags: , , , ,

Debugging Health Monitors on BigIP 9

We experienced that a lot of our health monitors didnt work on our BigIP boxes, and we sure as hell didnt see why.
After surfing a bit on the magical devcentral.f5.com, we found out that the BigIP only recieves the first 5120 bytes of a fetched result in a health monitor.

This is how to debug with curl:
curl -H 'Host: virtualhost.webserver1.com' -D - "webserver1.com80/index.html?q=bladdibla" | head -c 5120

Now you can ex. grep for a string to see if your healthcheck will do fine on a webserver (within the 5120 first bytes)

Tags: , , ,