Sonar Pollers fail until rebooted


#21

Please let support know - this is the first I’m hearing of this problem. We are going to need poller access to troubleshoot this so it’s not going to be something suitable for a forum thread.

Edit: Checked with Alex, looks like there was an emergency Ubuntu security patch that might be the culprit here. Do an apt-get update then apt-get upgrade then a reboot, and it seems to have fixed it for the couple that have contacted support about this.


#22

Hope so, we’ll see if they die again in 10 days (ish) no sense wasting anyone’s time until they begin duplicating the symptoms again.


#23

It doesn’t flood the interface all at once but can space out the polling over time. Furthermore it allows different monitoring granularity. Some things can be monitored every 10s, some every minute, every hour, etc.


#24

We don’t flood it today, there is a delay built in that spreads out the queries. You can also select to monitor CPE and infrastructure at different intervals. What specifically are you looking for above what exists today?


#25

Spaced out? Looks 5 min ‘spikey’ to me.


#26

Well, it will be if it’s on a 5 minute cycle. They are spaced out within the window of the operation firing. You can go dig through the code if you want to! :slight_smile:


#27

What I know is Sonar regularly complains about pings taking too long and SNMP taking too long with 4 pollers. Meanwhile we have 2 zabbix proxies that show none of these issues. The Zabbix proxies are monitoring FAR more SNMP OIDs with much finer granularity.

I have asked before about tuning the sonar pollers and not really gotten much in the way of guidelines. Should I be opening a ticket to review the settings?


#28

Yes, any time you have unsolved problems, please open a ticket. Our support team is there to help!