Mike Moniz

Members
  • Content count

    22
  • Joined

  • Last visited

  • Days Won

    2

Community Reputation

4 Neutral

About Mike Moniz

  • Rank
    Community Whiz Kid
  1. SQL Mirror DB alerting

    What DataSource are you using that is reporting down?
  2. parsing response

    I don't really know python much but LM did state they will be providing a new Python SDK once v.113 is released (end of the month) which might make querying LM easier: https://www.logicmonitor.com/release-notes/v-113-release-notes/ . I haven't seen it myself. Otherwise you may want to output/dump jsonResponse to make sure it's valid json (and not an http 501 error for example) or perhaps comment out the print line to see which line is causing the error? Perhaps you need to do ['data'][0]['id'] or the like? The API can return multiple devices. Again I don't really know python and I'm not currently setup to test with it.
  3. VMware Overview Dashboard

    I don't recall what the default VM dashboard looks like, and likely have changed since we start using LM anyway, but here are some more generic ideas: Most widgets have options to show a TOP/BOTTOM 5, 10 or 25 datapoints, you can try using this rather than ALL to not hit the 2000 instance limit. You can modify widgets to look at a smaller sections of your environment, perhaps based on your vm naming convention, and then clone the widgets as needed.
  4. Need to monitor all windows services in LM

    You should be able to clone the existing WinService- DataSource (which is what the wizard uses) and implement an AD WMI method to pull ALL the services then use the filter to remove disabled services and special cases. You may also want to not delete any instances immediately because once a instance gets deleted, you lose all historical/performance data for that service. You can still delete it manually if it's really gone or set to 30 days before delete.
  5. SQL Results Table Display

    So LogicMonitor (LM) is a monitoring tool, so you need to think of your data from a monitoring perspective. Monitoring tools are designed to reach out to a resource, get the current value of something and save it. It will then do this over and over again every x minutes. For example if you are doing a ping check, the collector will ping the device every 1 minute or so, record the response time each time, then you can graph the data it collected over time. You would need to let LM do the data tracking over time. So if you wanted to use SQL to track outstanding orders, you would modify your SQL statement to just return the current outstanding orders number. LM will run your SQL statement every x minutes and track the values. Then you can create a graph (and other) widget showing values over time.
  6. Meraki Multiple Organizations

    Great!
  7. That is correct. Can you provide more specific technical details? What DataSource/Widget are you attempting to edit exactly? What is the exact expression you are attempting to use? etc.
  8. PowerShell -- Instance Property Support

    Interesting, and works in Groovy? I haven't though of doing it that way before, learn something new I typically would either have the same basic code from AD in the collector script (since I attempt to use BatchScript whenever possible) or pass values over via wildvalue/wildvalue2/wildalias.
  9. PowerShell -- Instance Property Support

    Creating instance (auto) properties or attempting to use instance auto properties created from the AD script in the collection script?
  10. I'm guessing something to do with the names of the DataPoints. Did you rename your Datapoints in the widget (some let you do that)? I've had something like this before where I renamed table columns from like "UptimeDays" to "Uptime (Days)" and the calculation options doesn't like that.
  11. How can I find the dashboard a widget is on?

    Via the API? Using /dashboard/widgets/{widgetid} will provide a "dashboardId" pointing to the dashboard it's on. You can then call /dashboard/dashboards/{dashid} to get the name of the dashboard.
  12. How to use the ~ (includes) filter operator?

    When I've used commas in Filters previously, I've had to specify the field after each comma, for example "filter=internalId~LMD1234567,clear:*" or "parentId:1,fullPath~Template" so I assume you may need to do something like "filter=internalId~LMD1234567,internalId~LMD7654321". BUT I would have expected it to attempt to search for an alert that has internalId of LMD1234567 AND LMD7654321 which would never exist. So I'm rather surprised it even get one result. In my testing I get no results for "filter=internalId~LMD1234567,LMD7654321" but I get the first result when I use "filter=internalId~LMD1234567,internalId~LMD7654321", which I didn't expect. So I'm not exactly sure what is up. I do see that the doc doesn't show internalId as a valid filterable field but does seem to work anyway without commas. Playing around and watching how to portal does it (F12 in browser, network tab) I was able to get a working solution using: /alert/alerts?needMessage=true&filter=_all~"*LMD1234567 OR LMD7654321*" But really it looks hacky and undocumented. I would suggest just putting in two separate API calls for each LMD# and combine in code or get all the alerts and filter via code.
  13. "Column" type graph use

    Has anyone found any good use for graph's "Column" mode (vs line, area or stack)? Over most periods it either just merges into something that looks like Area, and over shorter periods it can look like lots of spikes. I can think of various items that would nice to report with a column/bar chart but only with the latest values of DataPoints. I'm not aware of a way to setup a graph to only show the latest values though and not over time.
  14. Add option to disable email reply commands

    I would like to also suggest the same for phone calls and SMS notifications.
  15. Alert Test Report

    I'm all for a built-in functionality for this. We onboard devices and whole environments on a regular basis and one trick we use is to create a group (even if temporary) for these new devices and place that group in long-term SDT. Devices created under this group will still create alerts internal to LogicMonitor but SDT prevents sending out any notifications. You can also filter out this group from any standard dashboards or reports using (!groupname*) and also do the reverse and create special reports that only apply to this group. If you mostly focus on notifications, you can instead create special alert rules to bypass standard notification/integrations rules for this group and either have LM not send anything or send them to the people who are doing the on boarding.