renegadenomad_x

Thanks Joshua. I do not have the count or timeout variables in the Execute Program. Should I? APM\SolarWinds.APM.RealTimeProcessPoller.exe -n=${NodeID} -alert=${AlertDefID}

Leon Adato wrote: What I meant by "You are polling CPU (and other stats) every ?? minutes?" is what your actual statistic collection cycle is. I am not sure what you mean by this.Where is that setting located? Leon Adato wrote: In the trigger actions tab you have two actions - the execute a program that runs the "get top…

Yeah, I didn't word that correctly at all. If I think there is only 10 running processes, I am in the wrong field. LOL Take the above output for example. Name Process ID CPU bengine.exe 3816 11.04 % WmiPrvSE.exe 2884 5.41 % sqlservr.exe 296 3.31 % pvlsvr.exe 2448 3.09 % svchost.exe 1016 2.98 % System 4 2.54 % WmiPrvSE.exe…

"You are polling CPU (and other stats) every ?? minutes?" Not sure what you mean. I will define the settings based on the tab they reside in. General Tab - Check this Alert every 1 minute Trigger Condition - Do not trigger this action until condition exists for more than 15 minutes Trigger Actions - Send E-mail/Page -…

I dropped the polling time to 1 minute. I still get stats that do not add up to 100%. When I run Prime95 and peg the CPU on a machine, the alerts show accurate stats, so I know this thing works. I guess the next thing I will try is setting the trigger action to 40 minutes. Not ideal to leave a CPU pegged for that long but…

Thanks Joshua, please do let me know if they improves things. Yes alterego, we get stats back consistently, that seems to never fail. I just cannot wrap my mind around how every device that gets triggered does not have a spiked CPU when the stats are polled!!! It is like the CPU drops as soon as the polling starts. The…

Thank you. This must be another design flaw. This example shows a "Location: that I can find no where else in the SAM, and it contains a server located two states away. Also notice the 99 nodes in the "Uknown" location. It is not obvious where these can be moved.

The output looks correct when running that poller manually. Do you recall how often you had this polling and how long for the trigger condition?

I removed the exceptions for now, but that did not fix this. Is it possible I have the escalation levels backwards?

Thanks! This makes sense. When I switch to SNMP this will blow away historical data. Is there a way to save the history in a report? Or, can I discover a new node and preserve existing?

I'm not one for scripting. In the IS SMTP monitor there is a Inbound Connections Current. I am trying to determine if this will alert at the loss of the smarthost?

No, the problem is that every poll cycle I get alerted for every monitor in the component, so if it is up or down I get an email for every one of them. I got it to work finally.

Well I agree I did not need to copy the alert, but to send a page for only this device I copied a new one. Thanks

I tried to import these to the reports but get errors, even after adding the XML extension. How do I do that?

No not tweak in Windows, tweak in the SAM. Even with only monitoring the C drive and CPU the SAM account logs in every minute. The triggers are set to 10 minutes, so I should only see a login every ten minutes. Successful logins can be beneficial for troubleshooting, so I am not turning them off.

Does anyone have this set up and working even?

I verified there was only this one. I removed the distro group that receives the email and replaced with my email address. The Reset Condition is configured as "No reset condition". Here is a screenshot.

It is a Dell PowerVault NX3100 queried by WMI. Is there a dashboard view to see just the raid arrays for all servers in the SAM?

The monitor of the queue works, as I just tested it with a count of 2 messages. But, this will not give insight of someone trying to email our Exchange from the outside. In the IIS component, are there any options to alert when the smart host cannot be reached or a gateway timeout?

That might help. Thank you!

I was trying to add nodes on the atlas. I had to add nodes individually because the grouping for locations was missing some but not sure how to get them all together. This can also be found in Manage Nodes, and selecting Group by Location.

Sure. As you can see from the alert email, the stats were pulled after the CPU dropped. The total is nowhere near 100% The CPU on node is currently running at 100 %. The top 10 processes running at the time of this poll are listed below: Name Process ID CPU bengine.exe 3816 11.04 % WmiPrvSE.exe 2884 5.41 % sqlservr.exe 296…

Thanks, I will do that.

Thanks for the tip. That was not obvious at all. =)

I found the "xen" account. I will delete it, but what is the best way to tell where this is configured?

For some reason I only get Up alerts and not Down. I manually set a threshold to trigger but no email is sent???? What did I do wrong? Here is the trigger condition.

Thank you. So I would just copy that alert and add only this device in the triggers correct?

It is not LDAP that is failing, the Barracuda alerts on LDAP failures as a result to whatever is going on there. If the Inbound Connections Current monitors the loss of the smart host I will have what I need. Is this what I am looking for?

SNMP will not report on say Active Directory for example. Now the agent sounds like a winner, and Id like to know how but the entire infrastructure suffers from this. There must be a way to trim this back, but how?

Works great. Thanks alterego!