Ciena SAOS thresholds using incorrect OIDs
We recently added some Ciena 3942s and 3904s to our monitoring, and found that Observium (CE 17.9.8859) is picking up the wrong OID for some sensors. It seems that the same OID was reused for a couple of checks where they should have been the next in sequence, causing optics Tx power and voltage to not be properly checked. The below change syncs the checks with the correct MIB values, and fixed the false alarms it was generating for us.
thanks, -e
admin@nms:~$ diff -ub /opt/observium/includes/discovery/sensors/wwp-leos-port-xcvr-mib.inc.php.orig /opt/observium/includes/discovery/sensors/wwp-leos-port-xcvr-mib.inc.php --- /opt/observium/includes/discovery/sensors/wwp-leos-port-xcvr-mib.inc.php.orig 2017-02-01 11:31:13.983653739 -0500 +++ /opt/observium/includes/discovery/sensors/wwp-leos-port-xcvr-mib.inc.php 2017-02-01 15:39:05.687844414 -0500 @@ -59,7 +59,7 @@ foreach ($oids as $index => $entry) { $entry['descr'] = dbFetchCell("SELECT `ifDescr` FROM `ports` WHERE `device_id` = ? AND `ifName` = ?", array($device['device_id'], $index)) . " Tx power"; - $entry['oid'] = ".1.3.6.1.4.1.6141.2.60.4.1.1.1.1.105." . $index; + $entry['oid'] = ".1.3.6.1.4.1.6141.2.60.4.1.1.1.1.106." . $index; $entry['current'] = $entry['wwpLeosPortXcvrTxDbmPower']; $options = array('limit_high' => $entry['wwpLeosPortXcvrHighTxDbmPwAlarmThreshold'] * $scale, 'limit_low' => $entry['wwpLeosPortXcvrLowTxDbmPwAlarmThreshold'] * $scale, @@ -101,7 +101,7 @@ foreach ($oids as $index => $entry) { $entry['descr'] = dbFetchCell("SELECT ifDescr FROM `ports` WHERE `device_id` = ? AND `ifName` = ?", array($device['device_id'], $index)) . " Volts"; - $entry['oid'] = ".1.3.6.1.4.1.6141.2.60.4.1.1.1.1.16.".$index; + $entry['oid'] = ".1.3.6.1.4.1.6141.2.60.4.1.1.1.1.17.".$index; $entry['current'] = $entry['wwpLeosPortXcvrVcc']; $options = array('limit_high' => $entry['wwpLeosPortXcvrHighVccAlarmThreshold'], 'limit_low' => $entry['wwpLeosPortXcvrLowVccAlarmThreshold'],
Hi Erik,
tnx!
Fixed in r8979. (currently in Pro of course).
Erik Muller wrote:
We recently added some Ciena 3942s and 3904s to our monitoring, and found that Observium (CE 17.9.8859) is picking up the wrong OID for some sensors. It seems that the same OID was reused for a couple of checks where they should have been the next in sequence, causing optics Tx power and voltage to not be properly checked. The below change syncs the checks with the correct MIB values, and fixed the false alarms it was generating for us.
thanks, -e
admin@nms:~$ diff -ub /opt/observium/includes/discovery/sensors/wwp-leos-port-xcvr-mib.inc.php.orig /opt/observium/includes/discovery/sensors/wwp-leos-port-xcvr-mib.inc.php --- /opt/observium/includes/discovery/sensors/wwp-leos-port-xcvr-mib.inc.php.orig 2017-02-01 11:31:13.983653739 -0500 +++ /opt/observium/includes/discovery/sensors/wwp-leos-port-xcvr-mib.inc.php 2017-02-01 15:39:05.687844414 -0500 @@ -59,7 +59,7 @@ foreach ($oids as $index => $entry) { $entry['descr'] = dbFetchCell("SELECT `ifDescr` FROM `ports` WHERE `device_id` = ? AND `ifName` = ?", array($device['device_id'], $index)) . " Tx power";
- $entry['oid'] = ".1.3.6.1.4.1.6141.2.60.4.1.1.1.1.105." . $index;
- $entry['oid'] = ".1.3.6.1.4.1.6141.2.60.4.1.1.1.1.106." . $index; $entry['current'] = $entry['wwpLeosPortXcvrTxDbmPower']; $options = array('limit_high' => $entry['wwpLeosPortXcvrHighTxDbmPwAlarmThreshold'] * $scale, 'limit_low' => $entry['wwpLeosPortXcvrLowTxDbmPwAlarmThreshold'] * $scale,
@@ -101,7 +101,7 @@ foreach ($oids as $index => $entry) { $entry['descr'] = dbFetchCell("SELECT ifDescr FROM `ports` WHERE `device_id` = ? AND `ifName` = ?", array($device['device_id'], $index)) . " Volts";
- $entry['oid'] = ".1.3.6.1.4.1.6141.2.60.4.1.1.1.1.16.".$index;
- $entry['oid'] = ".1.3.6.1.4.1.6141.2.60.4.1.1.1.1.17.".$index; $entry['current'] = $entry['wwpLeosPortXcvrVcc']; $options = array('limit_high' => $entry['wwpLeosPortXcvrHighVccAlarmThreshold'], 'limit_low' => $entry['wwpLeosPortXcvrLowVccAlarmThreshold'],
observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
participants (2)
-
Erik Muller -
Mike Stupalov