Unbound - 100% CPU Utilization
-
Hello,
I am experiencing random occurrences of 100% CPU utilization on unbound throughout the day. I have DNSSEC enabled and hardened. Here are the stats for my box:
2.2.2-RELEASE (amd64)
built on Mon Apr 13 20:10:22 CDT 2015
FreeBSD 10.1-RELEASE-p9AMD G-T40E Processor
2 CPUs: 1 package(s) x 2 core(s)Any assistance would be greatly appreciated. As you can see from the logs the service starts rapidly stopping/starting, CPU goes to 100% for the process for 5-10 minutes, then it goes back to normal like nothing ever happened.
I looked at my cron jobs and nothing really seems to line up with these times so it is quite confusing. Also there is nothing extraordinary in the system log.
Thanks!
-
Also, every time TOP refreshes the STATE for the unbound process changes in the following:
RUN
CPU1
CPU0Over and over as it is at 100%.
30153 unbound 1 102 0 262M 240M RUN 0 28:31 98.58% unbound
Then:
30153 unbound 1 103 0 262M 240M CPU1 1 28:04 98.88% unbound
Then:
30153 unbound 1 103 0 154M 134M CPU0 0 28:20 98.88% unbound
-
what packages do you have installed?
-
Hello,
-
Cron
-
Logstash-Forwarder (test package)
-
pfBlockerNG (have already looked through this with BBcan177)
-
Service Watchdog
-
softflowd
-
suricata
-
-
why do you have suricata why not use snort?
I had service watchdog but I thought it was a bit of overkill. Im guessing its constantly monitoring your services which may cause CPU load maybe try to uninstall it for a day and see how it goes.
And don't forget to reboot after (had an issue with squid one time)Also i had ntopng and I went from 26 percent to 60 percent CPU usage. Not sure if softflowd might be the case too but to figure out all possible outcomes I would also recommend uninstall it.
CPU goes to 100% for the process for 5-10 minutes, then it goes back to normal like nothing ever happened.
so your saying after 10mins of 100% it goes back down? and this happens for how long?