Steve Francis

LogicMonitor Staff
  • Posts

    267
  • Joined

  • Last visited

Everything posted by Steve Francis

  1. Two answers here - almost all our pre-defined alerts use custom alert templates for the alert body - but not the subject. The subject of the alerts just inherits the default alert template subject lines. So you could add the new property in the default subject template. The other option is to set the alerts to be delivered to a user by the user.pager address, and select that user to receive alerts via the sms template. (The sms address of the user can of course be any email address - not necessarily an SMS address.) Selecting the pager delivery mechanism with the sms template makes all alerts use the default SMS template, regardless of whether there is a custom alert message - and you can then change the default SMS format to whatever you want. If neither of those approaches work, let us know and we can probably work out an automated way to change all the custom alert messages in your account.
  2. Ah - thats a different issue. Its not data consolidation, its that the data collection time is not aligning with the graph display time. (i.e. the data is being collected at 30 after the hour, say, but the points are being plotted on the hour - so the graph is interpolating to try to determine a sensible graph for the time it is being plotted.).n nNot sure why the graph plotting time cant align with the collection time for these long period collection datasources - but the ability to collect over long periods is new, so I havent met this particular issue before. Ill check with development, and open a bug.
  3. There is, as the saying goes, more than one way to skin a cat. Or alert on a load balancer.n nAnother way would be to add a virtual datapoint that returns the Services UP, if the state = up, but returns a static value that will not trigger the threshold if state = out of service.n nSo youd add a complex datapoint to the datasource, and define it with an expression like:n nif(eq(state,7),ServicesUp,100)n nThis means if state = 7 (the service is up), then return the value of ServicesUp (a simple datapoint defined elsewhere in the datasource).n nElse return the value of 100.n nSo if the LBVS is not up, it will return 100.n nThus if you set you thresholds to < 4 2 1n nyoud get warning if less than 4 associated services were up for a LBVS; error if less than 2, and critical if less than 1. If the LBVS is not up, the value returned would be 100 - which would not trigger the alert.
  4. As of now, LogicMonitor only supports snmp over UDP port 161.n nIn the next release (between 4 and 14 days, depending on which server you are on) it will support SNMP on any arbitrary UDP port (by specifying the snmp.port property for a host or group of hosts.)n nNo current plans for snmp over TCP.n nYou could tunnel UDP over ssh a la http://www.qcnetwork.com/vince/doc/di...n nor run an agent on a device on teh far side of the tunnel...
  5. webhooks/posts as an alerting mechanism are live.n nWere still awaiting our SMS gateway provider to enable us for international SMS delivery. The code is all ready ....