We’ve been running Observium on Ubnutu 12.04 for a little over a year now, and after applying the latest community update, we seem to have devices periodically reporting as down.

The error I’m getting is bouncing between UnpingableMySQL and UnpingableRRD.

With debug enabled, this is an example of what we’re seeing:

/poller.php -h itups03-02.net.internal

Observium v0.14.4.5229

Poller

Starting polling run:

SQL[SELECT `device_id` FROM `devices` WHERE `disabled` = 0 AND `hostname` LIKE 'itups03-02.net.internal' ORDER BY `device_id` ASC]

SQL[SELECT * FROM `devices` WHERE `device_id` = '40']

SQL[SELECT * FROM devices_attribs WHERE `device_id` = '40']

itups03-02.net.internal 40 apc

UnpingableRRD[cmd[update /opt/observium/rrd/itups03-02.net.internal/status.rrd N:0]

stdout[OK u:0.00 s:0.00 r:0.05]

stderr[]]

RRD[cmd[update /opt/observium/rrd/itups03-02.net.internal/ping.rrd N:U]

stdout[OK u:0.00 s:0.00 r:0.06]

stderr[]]

RRD[cmd[update /opt/observium/rrd/itups03-02.net.internal/ping_snmp.rrd N:U]

stdout[OK u:0.00 s:0.00 r:0.06]

stderr[]]

SQL[INSERT INTO `perf_times` (`type`,`doing`,`start`,`duration`,`devices`) VALUES ('poll','itups03-02.net.internal','1401246542.664','0.066','1')]

./poller.php itups03-02.net.internal May 27, 2014, 20:09 - 1 devices polled in 0.066 secs

MySQL: Cell[0/0s] Row[1/0s] Rows[1/0s] Column[0/0s] Update[0/0s] Insert[1/0s] Delete[0/0s]

Any advice on where we can look? Seems like the only way to get them back up is to keep running the poller for the particular host until the poller runs.

Thanks!

Andrew Davis
IT Systems
J. David Gladstone Institutes