mdriskell ✭✭✭✭✭

Can anyone tell when we put the buddy drop in? Appears it has corrected the memory leak. What used to consume upwards of 1.3GB of memory is now less than 200 meg.

Thanks Lasher, Judging by the logs Solarwinds support sent me my issues are with the Interface Traffic and Errors procedures. I have increased the limit to 100000 and will try the DBMaint again. I will post the results when I have them.

Looks like it just happened again. Restart of services doesn't correct I have to rebuild the SDF files to get polling to resume.

What I'm looking for is SAM Component availability. Similar to the three alerts for the polling check (Interfaces, nodes, Volumes) I'm looking to create the same thing for APM components. There are charts on each component that show the availability but I'm looking to write an alert with that data. The information appears…

My BES traps always came in as that Netlogix...I remember the day they turned on the traps I thought it was odd they appeared as Intel. If you go to the trap viewer and search by trap type *snmptraps.7* do you get any hits?

I haven't had a chance to work with the UNIX team to figure out a working script for them but I was able to work out a script for powershell. I'm going to post a template later on today for this particular monitor. # This script will search a text file for a particular string. # It is designed to return the number of…

Boggles the mind doesn't it...Might add some of this to the notes on this feature request. http://thwack.solarwinds.com/ideas/1139

Is your solarwinds server sending traps to itself is what I'm asking.

I'm saying your solarwinds server is probably sending traps to the localhost.

Support finally called yesterday around 3:15 2 + hours after our planned timeframe. I reproduced the issue for him and the first thing the guy had me do was open up a connection to the database. He provided a query he wanted me to run. The query was a truncate table for the trapvarbinds table. There was no explanation of…

This post that I had typed on Friday just now showed up after I got an email saying it had been approved...

FYI the tab if I remember right should be under the trigger action then the email section. Sorry I don't have access to my system right now to get you a screen shot.

Add an additional action that doesn't take effect until 10 times the value...so 50 minutes in your case.

Just had it happen again...refreshed a minute later and the page was fine. 

Ok thanks for the clarification. Thanks for the tip on the Amazon EC2 as well. I saw a need to be able to test outside in but hadn't given any thought yet as to what I would be testing with.

It's not an alert (I don't have that enabled) just simply showing up on our nodes with problems list

Doesn't solve your issue directly but I created a way to do this with a report. . I have it in multiple flavors 1 Day, 7 Days, 30 Days on our systems and gives the NOC the ability to see who the problem children are very easily. Take note that I changed some of the event verbiage to make this work.

Give me a big blaring red ribbon at the top of the screen that reads, "CHECK THE POLLERS" and I would be a happy camper. I had a custom report that showed the poll time of a device on each poller for that very reason. I could quickly see the last poll time and last database update at a glance on our dashboard.

I can reassure everyone that you are getting our attention, however, this is still a problem. I don't know how helpful this is, but it is something we would like to improve soon. The problem still exists in NPM 10.2. We recognize this and we'll fix it, but I can't say when exactly (don't worry though, you won't have to pay…

It's pretty easy to write using the report writer but here it is in SQL. I am filtering out interfaces that are UP, Unmanaged, Unplugged, or Unreachable. SELECT Nodes.NodeID AS NodeID, Nodes.Caption AS NodeName, Interfaces.InterfaceID AS InterfaceID, Interfaces.Caption AS Interface_Caption, Interfaces.Status AS Status FROM…

In regards to the DB that is a big question. Every repair or upgrade requires the config wizard to "talk" with the DB. Not sure how you could get around that. I definitely like the idea of being able to repair a poller independently that would be great. The big problem right now is if poller 1 is in trouble everything is…

Changing to Last 30 Days corrected our issues. Thanks for the help.

dnerdahl I'll let you know. It should go into production next Thursday. In the mean time maybe have your support guy reference our case notes 543306 perhaps they can draw the same conclusion.

Very similar. We have 4 pollers but do not have the additional web server. We only see the symptoms on the primary poller as it acts as the web server as well. That service only impacts the front end from what I can discern as no polling alerting or any other functionality is broken when it fails just the web ui. If you…

That's our fix right now...restarting the service before we get to "critical mass" I'm looking for a way to stop it from happening period but so far no luck.

I return over 1500 results with the greatest number of missed polls being only 2

I don't recall what the fix was. I'd recommend opening a ticket with support and referencing this thread.

Thanks.

FYI I have seen this disappear from time to time in Chrome myself...it will be missing and then I reload and it's back. Doesn't happen often but I have seen it disappear occasionally.

That looks normal to me. Remember you have up to a two minute delay in identifying a node down based on a 120 second polling interval. One of the SW Staff can better explain how a node goes from warning to down.