Popular Content

Showing content with the highest reputation since 04/29/2017 in all areas

  1. 2 points
    We had a DHCP pool that was nearing capacity and noticed that LM didn't alarm? Please can we request some DHCP pool monitoring capabilities for Windows Server?
  2. 2 points
    Please add an option to turn off the rows added to a CSV report (see below for example). A CSV file should be just that, a CSV file. I want to use the CSV to feed another process but having these rows means having to pre-process the file. Company: rentokilinitial From: 2017-01-13 19:56:43 GMT Powered By: LogicMonitor Inc. To: 2017-01-20 19:56:43 GMT Description: Router availability
  3. 1 point
    when trying to add a NOC widget under a dashboard you have to choose between a device or a service. right now you can not mix services and device components under one NOC widget. it would be great if that can be fixed. thank you
  4. 1 point
    Occasionally the Redis- DataSource returned no data for some or all of the DataPoints. Under test, it was noted that the key value store returned to cli.stdout was being truncated. It was further noted that the cli.expect string of 'r\n\r\n' occured at multiple points in stdout, not just at the end of the response. Under certain stress conditions (when the device response was slow) stdout was truncated by the groovy script. It was noted that (in a successful return) the last key value returned was under the heading of "# Keyspace", followed by the key values, followed by a final 'r\n\r\n'. To resolve this issue, the groovy collection script was modified with an additional cli.expect("# Keyspace"), prior to cli.expect('\r\n\r\n'). On the monitored devices this modification fully resolved the issue. The modified script is shown below: import com.santaba.agent.groovyapi.expect.Expect; cli = Expect.open("##HOSTNAME##", ##WILDVALUE##, 5) if ( "##REDIS.PASS##" != "") { cli.send("AUTH " + "##REDIS.PASS##" + "\r\n") } cli.send( "INFO\r\n" ) cli.expect("# Keyspace") cli.expect('\r\n\r\n') println cli.stdout() cli.send( "QUIT\r\n" ) It is not known if this is caused by a recent change to the way that the INFO output is returned by Redis. All available devices exhibited the same behaviour and were at Redis version 3.2.8 on Linux.
  5. 1 point
    Good Morning all, we are a multi-Team managed service provider who, surprisingly, manages customers team-wise. We are in the process of switching to LogicMonitor right now and we have some issues regarding shared devices and their DataSource Instances, that belongs to different teams. We created Devicegroups for our teams, but if we put shared devices like loadbalancer, firewalls, storages etc. in said groups, ALL teams will see ALL alarms, regardless of responsibility. If one team decides to turn off an alarm on a shared device, the alarm is obviously deactivated on all other teams, as well. I'm aware of the Datasourceinstance grouping mechanism, but afaik it cant help with the problems i mentioned. We would like to request a feature where you can somehow mask datasources, their instances and their respective alarms based on the group you are in while you're viewing the device, so alarms and instances that don't belong to your team doesn't show up at all. It would be really convenient to somehow configure a "view" of a device based on the group you are in that doesn't interfere with the device datasources. Bonuspoints if you bring in alarm-tuning in this somehow. Regards, Bastian
  6. 1 point
    Hey all, The ability to drag and drop items in the device tree in order to reorganize them is a powerful feature, but the staggered manner in which the tree loads and expands on page load "encourages" a user to accidentally move folders and devices around in the tree, which leads to numerous detrimental effects (not the least is having to try and figure out where you accidentally moved them to). Request is to either introduce a safeguard prompt "Are you sure you wish to move X from Y to Z?" for all drag and drop actions within the device tree, or, alternately, introduce a configuration item on folders and devices allowing them to be "protected" against moves --- either adding a safeguard "are you sure...?" prompt for those specific items, or prevent moving them entirely without first taking the action to "unprotect" them. Thanks, Nick
  7. 1 point
    Is there a list of functions and variables that LogicMonitor Exposes to Groovy? I'm specifically looking to leverage a device ID in a groovy script that pulls data from the LM API. I know I can pull device/instance properties but the actual device ID LM uses doesn't appear to be one of them. Thanks, Thaddaeus
  8. 1 point
    Being located in Israel I need an Israel time zone that adjusts to DST as we move in and out of DST I currently have to manually change to UTC+2 or +3 when we move and there is really no reason for that. As a developer on my other hat, I had to deal with multiple time zones on our application as well. I suggest your developers have a look at Joda-Time timezone support (for java http://www.joda.org/joda-time/tz_update.html ) or Noda-Time (for .NET) as it has a list of all IANA time zones and supports updating of the time zone database, time zone conversions etc. This would make your developer life easier if they have not already used it. On our application I used this and now all I have to do is update the IANA tzdb file and all time zone changes are updated and my servers know DST for the whole world imminently
  9. 1 point
    I have multiple sites running our system. I have a device group for each site consisting of all devices (db, web etc) and a corresponding Services group (same name) for monitoring the web site. If I need to take down the site for maintenance, I need to update both the devices and services with the downtime. My suggestion: Add rules on a device level of which services to take down when the device goes to SDT. Examples: 1. if I have a test machine (single machine running test.domain.com), I can tell the rules that when the machine goes to SDT put down the service monitoring of test.domain.com 2. I have 3 machines running a load balanced site (site.domain.com), I can tell the rules that if all machines running this site are going to SDT put the service to SDT 3. I have an external site (external.domain.com) which is hosted elsewhere, because of that it will not be put down when any of the devices go into SDT
  10. 1 point
    I would like to be able to create a series of dashboard groups that are identical in layout with identically named dashboards on them. At present this is not possible because when trying to save the second dashboard I am informed that "A Dashboard with this name already exists."
  11. 1 point
    Mark, I assume it will not happen but I am talking specifically about voice calls. I still *do* want the clear emails/sms for keeping track on alerts and when they were activated/cleared The only thing I do not need is LM calling me the issue is resolved. If I am solving the issue, I know it's resolved when I stop working on it and I don't need 20 calls telling me the issue is solved (what I don't answer goes into my voice mail so now my voice mail is also full). If I cannot solve the issue and I have contacted a 3rd party support, I still ACK it as I am the only one who can solve it. Then if they solve it in the middle of the night I do not need LM calling me at 5 a.m. to tell me the issue is resolved. The latter is an actual case were some DNS server we have records on stopped answering some of the queries sent to it (probably one of few servers was stuck but the load balancer still forwarded requests to it). LM called me at 3 a.m. to tell me that our site is not working from the US. I contacted the DNS support via email, put the alerts on SDT until the morning and went to sleep. At 5 a.m. LM called me to tell me it was over which woke me up again. I think a more granular alert management is needed so for example I can say that alert cleared voice calls are not sent at night
  12. 1 point
    Today, the system.displayname property for an AWS device is not automatically updated when the "Name" property in AWS is changed. It would make it easier to see which devices we are actually monitoring if the display names in LogicMonitor were synced with the current "Name" property in AWS.
  13. 1 point
    I have been using a Powershell script for a while now that used Batch Script (to not use as much collector resources as my company has TONS of App Pools per box) This links Win32_PerfRawData_PerfProc_Process with Win32_Process using the process ID for discovery. Instead of having an ugly instance name (W3WP#1,W3WP#2 etc) it now will display the AppPool name from the command line. We recycle AppPools nightly. This will run discovery every hour as it stands although, the collection uses instance ID ,matching the Command line so no holes in the data LM Exchange ID: 2WHYY4
  14. 1 point
    A larger issue here is that data references are often stored within LM as a name, not the object ID, so changing the name in one place divorces it from the object it was tied to. That can make sense for patterns, but not in other cases, like usernames (if you ever change a username, all references to the user are broken, including escalation chains). As far as dashboard tokens -- they are awesome, but not universally supported in all widgets, sadly (for example, Alert widgets do not support them). +1 for custom properties in alert rules, as well as instance group references. Thanks, Mark
  15. 1 point
    Any news on the ability to add a network diagram? I just had a switch dropping dead and got a ton of alerts from various devices and data sources. It took me a few minutes just to figure out what is the root cause of all of these problems. A live network diagram were I could see all elements and look for the root cause of it all could have saved me at least 5-10 minutes of time before I could actually started working on the problem.
  16. 1 point
    MW7E2C Description: Monitors the health of each VM as reported by VMTools. How it works: Automatically applies to ESX hosts and adds each guest with VMWare tools installed as an instance. Polls for the status of the VMWare tools service, the vm's heartbeat, configuration issues, and up/down state. Doesn't alert on anything unless the VM is up. Appears to be quite reliable for detecting VM's that are unresponsive even though all other metrics appear normal. Alerts aren't triggered until after at least two datapoints to prevent false-positives. No graphs for this as it's pretty straightforward. Since VMWare reports health rollup stats as colors (green, yellow, red, gray), this script converts those colors to green:1 - Yellow:2 - red:3 - gray:0 Alerting: Alert subjects are customized to be very clear. Body of the alert email explains what each threshold means in case there's ever any confusion. Notes: This will likely only work on esxi 5.1 and up. Make sure the host has valid credentials set for the esx.user and esx.pass parameters.
  17. 1 point
    This is a pretty cool idea. Maybe LM could make a function like DatasourceIncluded("Apache-") or something similar.
  18. 1 point
    It would be very useful for my organization to be able to group by the existence of a datasource - for example, servers with the "Apache" datasource could be sorted into a "Webservers" group or servers with the "Mysql" datasource could be sorted into a "Database" group. We're sorting by property right now so we can use dynamic groups, but it'd be really cool to be able to do things through datasources rather than properties.