Nightly Ping Loss Alert Spam

[WireDrifter]

Hello All - For the past few nights, we have been getting a slew of alerts at about the same time showing that zabbix is unable to ping our hosts. In the zabbix_server.log, I see the following right when the alerts start coming out (I edited out the hostnames). This started last Thursday morning (12:30), Friday (also 12:30), did not happen Saturday, Sunday Morning (12:30) and again this morning at 1:05. I don't believe any changes were made on the network side of things, but have asked our network admin to verify. Any thoughts on what might be causing this or where to look for more info? I'm still learning zabbix, so not entirely sure where to go...

2113:20160111:010521.995 Cannot get socket IP address: [107] Transport endpoint is not connected
2115:20160111:010521.995 Cannot get socket IP address: [107] Transport endpoint is not connected
2113:20160111:010521.995 cannot send list of active checks to [unknown IP]: host [host] not found
2115:20160111:010521.995 cannot send list of active checks to [unknown IP]: host [host] not found
2113:20160111:010521.995 Cannot get socket IP address: [107] Transport endpoint is not connected
2112:20160111:010521.995 Cannot get socket IP address: [107] Transport endpoint is not connected
2112:20160111:010521.995 cannot send list of active checks to [unknown IP]: host [host] not found
2111:20160111:010521.995 Cannot get socket IP address: [107] Transport endpoint is not connected
2111:20160111:010521.995 cannot send list of active checks to [unknown IP]: host [host] not found
2114:20160111:010521.995 Cannot get socket IP address: [107] Transport endpoint is not connected
2114:20160111:010521.995 cannot send list of active checks to [unknown IP]: host [host] not found
2114:20160111:010522.000 Cannot get socket IP address: [107] Transport endpoint is not connected
2115:20160111:010522.193 Cannot get socket IP address: [107] Transport endpoint is not connected
2111:20160111:010522.194 Cannot get socket IP address: [107] Transport endpoint is not connected
2115:20160111:010522.228 Cannot get socket IP address: [107] Transport endpoint is not connected
2115:20160111:010522.228 cannot send list of active checks to [unknown IP]: host [host] not found
2114:20160111:010522.231 Cannot get socket IP address: [107] Transport endpoint is not connected
2114:20160111:010522.231 cannot send list of active checks to [unknown IP]: host [host] not found
2113:20160111:010522.232 Cannot get socket IP address: [107] Transport endpoint is not connected
2113:20160111:010522.232 cannot send list of active checks to [unknown IP]: host [host] not found
2114:20160111:010522.275 Cannot get socket IP address: [107] Transport endpoint is not connected
2114:20160111:010522.276 Cannot get socket IP address: [107] Transport endpoint is not connected
2114:20160111:010522.277 Cannot get socket IP address: [107] Transport endpoint is not connected
2114:20160111:010522.308 Cannot get socket IP address: [107] Transport endpoint is not connected
2112:20160111:010522.311 Cannot get socket IP address: [107] Transport endpoint is not connected
2112:20160111:010522.311 Cannot get socket IP address: [107] Transport endpoint is not connected
2114:20160111:010522.386 Cannot get socket IP address: [107] Transport endpoint is not connected
2114:20160111:010522.386 cannot send list of active checks to [unknown IP]: host [host] not found
2112:20160111:010522.387 Cannot get socket IP address: [107] Transport endpoint is not connected

[fire555]

Hi WireDrifter,

Did you end up finding the cause for this issue?

I am also seeing this issue in my installation. I am running version 3.0.1 in Amazon AWS. The trappers just stop accepting new connections from active checks. A zabbix server resolves the problem, but is hardly a long term fix.

[glebs.ivanovskis]

Can this be the same issue?

[fire555]

Oh thank you so much, this sounds exactly like what I am experiencing!

I too also have some monitored hosts on less than stellar networks.

I guess I will wait for a new rpm with the fix. In the meantime, a twice daily scheduled reboot seems to be keeping the server mostly up