dkramkowski

What is the resource footprint of the agent (CPU/Memory)? We have SAM set up using WMI, and for the most part, it works well, but we periodically have issues where WMI goes haywire and starts beating the server up, requiring a restart of the service. I've also noticed that sometimes, WMI just 'doesn't work' for reasons…

I agree - this needs to be implemented - For example, I need to monitor a Tomcat process. Right now that Tomcat could be tomcat6.exe or tomcat7.exe, and who knows, in the future there could be a tomcat8.exe. It should be possible to set up a monitor using 'tomcat*.exe' so that if it has 'tomcat<anythingatall>.exe' it will…

After several hours of messing around with it, I believe I have it sending out the alerts with the data I want and only sending one e-mail each time. I'm not entirely sure I like how it's set up, but it's accomplishing the goal. The Variables don't seem to be globally usable. Initially, on trying to set up a new alert, I…

Actually, I don't think it's alerting on two components, but it may be reporting on two - the process monitor is the one that's in a critical state, but the WMI monitor is happy.

You may be correct, but I ended up calling into tech support and that was a disaster. Ever since he monkeyed with the alert, it hasn't been sending *anything* and I'm getting 'Failure sending mail' for that alert. No idea WHY it's failing - any suggestions how to get more detail?

I just upgraded us to 6.2.1 now that it's been released and I'm seeing the same double alert e-mails. Any other thoughts?

I do not believe I have NPM installed at all. Don't see it in the list (or might it now be called something else?

So if you select the WMI method, does the agent grab that instead and not do anything with WMI, or is WMI still involved in the data collection process?

I hope that's really what's going on and will be fixed in 6.2.3, however, in my case, when they are saying they need a reboot to finish deployment, they are NOT reporting data properly. I've applied the previously mentioned hotfxes, and over last weekend, rebooted every machine that showed 'reboot needed to complete agent'…

Well, that does appear to apply - the Nodes are monitored by agents, These nodes did work and then stopped reporting data and they do say connected (Just 'Reboot needed as well). I'll give that a shot.

How does the agent handle component monitoring? I currently have several monitors set up using WMI monitoring to look at processes and other metrics that I need to keep an eye on. When I go to add a component to the monitor, my only options (for a process monitor, for example) are 'Process Monitor - SNMP' and 'Process…

Nice, Thanks! Just upgraded from 6.1.1 to 6.2 and made that change to the monitor. Perfect!

Here is what I have set up for the trigger:

That's a bummer. I don't see anything else that would seem to apply. Hopefully they will get added to the alerting at some point.

How would LEM help? If I read it right, LEM is basically a log manager similar to Splunk, although with more features. If we were talking about an application hang/crash that triggered and event log entry, I could see it helping, but this isn't causing an event log entry. It's just an application running off with the CPU.…

I'm looking for 'Is NOT less than'. IE: If the available disk space is less than 15GB and NOT less than 10GB, it will send an alert, but if it drops below 10GB free, the less then 15GB AND not less than 10GB no longer is true, so it'll cease to send the alerts as at that point, the next level alert would be taking over.…

Yes, both are coming from the same alert - I had already tried adding text to the end of the subject and that text appeared on both messages the following time the e-mail was sent. The reset condition is 'Reset this alert when trigger condition is no longer true'. I just tried removing all of the parameters in the message…

Yes, I saw that in my other thread, but I had assumed that the bug was one where an updated agent that needs a reboot stays in that state even after a reboot, not an agent that was working randomly complains it needs a reboot. I don't think it was real clear in that thread, but the agents that were needed a reboot in it…

Following up with this - it's not a bug with the GPO deployment, but rather something with the wizards that create the tasks. I set up a patching task using the wizard to patch two machines - like with the GPO deployment, the wizard created the task with the title 'Update Management with rules for 2 computers created at…

Bringing this back up to the top as it's still an issue, and it's causing us problems. The latest example of it involves machines that got the agent installed way back in June. These machines have been rebooted at least six times since then, and they had been operating normally. This past weekend was out patching weekend,…

To circle back on this, it seems like the root issue wasn't really with the content of the message, but the alert itsself. I've noticed that other alerts that I created way back and were migrated to the web based alerting, are also sending duplicate messages. They are far simpler, and one of them actually still has the…

It looks like I figured it out. It would seem that my Watchguard proxy was interfering with it. The route that was working was the device at the other end of my VPN, so it was a trusted device and not subject to the proxy (The probe is not going over the VPN). When I added my NPM server to the exempt from proxy list,…