cgregors

I ran into the same concern. My solution (that I can't share the code for at this time) works like this: * Run a re-discovery script written in php to enumerate the resources on each node monitoried in Orion.* include filesystems * include interfaces * include cpu and mem * Store the data in a custom table in the Orion…

I saw another thread about NPM not being able to connect to iControl when AD authentication is used. Not sure where the thread it. Sorry

change your query to : $query = "SELECT NodeID from Nodes";

The condition "interface status is equal to down" is too specific. try reversing the logic to this: "Interface status is not equal to up".

I assume you mean in one of the alerts that get sent to the unix support staff. I have an alert defined like this: Alert Name: OSS : Filesystem Capacity Critical (24x7) > 90% trigger conditions: * node status = up * volume percent used > 0 * volume percent used > 90 * volumetype = Fixed Disk * Volume Responding = Y *…

To overcome this problem we implemented the following: * Create a SAM template with the following component: * Assign this template to ALL nodes that respond to snmp. * The statistic value is the # of seconds the host has been up. This seems to survive a snmp restart on the node. * We then use an alert that trips on "less…

I can't think of a distro that has nagios already installed. If you use a fedora / centos / redhat distro, you can just do a "yum install nagios" and get it installed after the fact.

Yah, I noticed that in the query. I hoped the distinct qualifier would fix it but it didn't. Hopefully someone with more sql skill can fix the query to return a more unique list. Also it would be good to flag which IP address is being used as the polling address. This would require another join to the nodes table. Cavaet:…

I agree also. * While our Solarwinds products are not exposed to the big-bad-internet, it is good practice to deal with security problems proactively. Personally I'm more concerned about internal security threats than external ones. * From an internal process POV, security patching gets expedited in our change management…

So if I have 1500 windows nodes that I want to change from SNMP polling to WMI, I can expect an increased load on the polling engines tech info: polling every 5 minutes collecting cpu/mem/disk SAM templates in play on 500 nodes (average 3 components).

unfortunately the mix of devices on the network requires SNMP and WMI credentials to be used. Example (with completely bogus credentials) network devices use the SNMP RO community string of qqz unix servers use the SNMP RO community string of foo windows servers use the SNMP RO community string of foo windows servers use…

I guess I'm overthinking it also. Where is the "Devices" tab ?

Thanks for the pointer. Thats the template I'm using for the NPM servers already. The situation I want to have NPM alert on is when it talks to NCM and the user gets this dialog: In this case, ewwv0037 is the NCM server but the exception is being seen on the NPM Web console while working with nodes. I realize that NPM is…

I have managed to accomplish _something_ like this and would be interested in showing aLTeReGo this offline. The complexity was large and I had to write a lot of php code (my choice of language these days). Chris

Wow, 4 years and you're the only one that answered. As I remember I slew that dragon a couple of years ago. Can't remember how. Thanks for finding this. Chris

So I was mixing what is inside the NPM database with what the Web UI displays. A container in the DB = A group in the Web UI. Nothing to do with Cisco equipment, SNMP oid's or UnDP. I found a "poor" fix. I edited the group and hit submit. That updated the status in the database. That still doesn't explain to me how a…

I ran it against my db and it didn't return all the credentials. There were some I expected that were missing.

The Orion Report Writer has a major calculation flaw when doing availability calculations that cross these boundries in the database: Database Settings Archive Time The time of day to run the Orion database maintenance process. Auditing Trails Retention days Auditing Trails statistics will be deleted from the database…

Overall I am of the opinion that monitoring OS resources via the OS is the best idea. That includes the network connection(s). Reliance on the HP ILO to get the data works in my environment because the sysadmins are pretty anal about putting the ILO drivers in the OS (both Windows and Linux). If the sysadmins didn't do…

I've started down the APM documentation path in my spare time. I'm working on building up the following sql query: Note: I don't like joins. select nodes . caption 'NodeName' , APM_Application . Name 'AppName' , apm_component . name 'ComponentName' , APM_ComponentDefinition . Name 'ComponentType'from APM_Application ,…

Try this sql in a custom table resource: NOTE: Change some_custom_property to what you need in the where clause and pick your EventTypes select top 25 EventTime, NetObjectID, Events.EventType, '<img src=/NetPerfMon/images/Event-' + LTRIM(STR(Events.EventType,10)) + '.gif> ' + '<a…

Could you post an example of how you handle the filesystems in HPUX with custom pollers? Chris

This is not currently possible today. I had to chime in about the answer above. It always makes me smile and quietly laugh. If a function is not "currently" available : 1. Was it available in the past and deprecated out at sometime ? 2. Is it scheduled for inclusion in a future release ? 3. Are you waiting to see if there…

We've been sending alerts to Remedy via rmcreate.exe which creates a ticket in Remedy since 2005. Using a workflow to match the nodes in Remedy, enabling us to autoclose/update tickets, when Orion sends updates/alert resets. I've never heard of rmcreate.exe. A quick google of it shows nothing (at least the way I search).…

It appears to be a mix of connections. Some go from the web server to the database, some from the webserver to the main polling engine. The ones to the main polling engine are SWIS queries over port 17777 but I don't know what data is being queried. chris

You can separate the web load off the main server with a webserver license. We've got 3 running like that and it helps.

HAS ANYONE ELSE NOTICED THAT TICKET SUBMISSIONS DON'T GIVE EMAIL ACKS ANYMORE? http://thwack.solarwinds.com/thread/57893 Ticket submitted via web 2 days ago and no email, no phone call, no nothing. I'm gonna call. I'm annoyed. this used to work quite nicely.

Whoops. didn't notice the 300Mb / 6 hours. That would make the final number ~7K / minute / server

Here's my thought process: 1. Node 1 is detected down. NCM puts it into the work queue with an aggressive ICMP poll. 2. Node 2 goes down and it gets put into the work queue AFTER node 1. The queue is processed sequentially .... Time goes by .... people get older ... network / sysadmins fret .... 3. The problem is resolved…

That was the way my thinking was leaning. I realized that a JobID is not the same as a uniqueid (like the IPAddressN column in the NodeIPAddresses table). I'll cobble together something that merges the profileid with a homegrown id for the NPM instance. Thanks for the input.