bluefunelemental

Are you looking for the report to be generated at a specific time, x hour on each Monday, or when a specific event occurs, when count(x) > 1 ? if the former then a SQL report and the scheduler is fine- though I would recommend the new web-based ones if the later then you could create an alert to trigger based on your sql…

‌what I did in a similar experiment was to apply a SAM WMI process monitor against winmgmt service and an snmp process monitor against the snmp process. That allowed me to see the amount of CPU monitoring was contributing to the overall CPU and memory load. In my case it was always 0. How about yours?

I thought it was the endpoint resrouce?

Following along- I'm interested to see if you could trigger a script to pull the results of the oa cli command show syslog iLO all back into Orion? not sure if you could just get the same data by trap or syslog?

JJust to confirm you do not have any rows with status=available?

Travel day so pardon the short response. So when adding a node Orion uses DNS name if it can get one. If you have a bunch of IP's for caption then did you add DNS after adding into Orion or are there no DNS at all? I would query through SDK nodes whose caption <> DNS. Let this fill DNS and URI variables. Start there. Then…

Sign me up. Twice.

Old post but trying this as well and it seems to work for me - any issue with this approach besides being SQL vs SWQL? Add- report with datasource Custom SQL and use SELECT TimeStamp,Name, AlertObjectID, EntityCaption, Message, FROM Orion.dbo.AlertHistoryView WHERE [Message] LIKE 'Execute%' ORDER BY TimeStamp desc I don't…

sounds like this will be a large sql join report or several smaller ones. - either way start smaller and combine as you validate them. I would think a left outer join of nodes and each of the tables you want, interface, apm_hardwareinfo,etc where the right table column is null would work. I did something for your orphan…

MIght be best to post your swql or even better- do it first in sql to confirm your compliant then try to translate to SWQL- I find this removes a lot of trouble where SWQL just doesn't Like my functions But the script is fine. Rules to union data: * Each query must have the same number of columns * Each column must have…

I‌nterest piqued e-mail sent

There is no concept of alert assignment and solarwinds but you can set up assignments and custom properties and you can create alerts specific unfiltered for those groups Are you using web help desk or another incident mgmt system?

Mr Adato is right about your triggering logic. When faced with the same issue I quit my job and became a hermit decided to use a daily report instead. Web based reports are working great to pull me daily event log scheduled to email me every am .

Agents are much better till they are not. We have a preference for agents for all servers but find * they hang more frequently then WMI/SNMP would (still less work maintaining then adding them by hand) * they don't play nicely with Network Sonar Discovery AT ALL. Yeah you can check the box but try having one discovery and…

Not to be flippant but I simply set the sqlserver application to 0 in Nodes.ChildStatusParticipation and walked away. We will really need the ability to disable and fine-tune rollups from individual app insight components in a much better way in future versions - just highly fraged msdb tiny indexes is enough to make…

Ah yes. I've had a few calls with the PM's to improve the data they pull from UCS but no improvement yet. Approach I am taking is too pull in the data I want through UCS Powershell API then stuffing what I want into node custom properties in SWIS SDK. Maybe a field marked serial number to associate blade to node then…

I am tagging along here- I would be interested in filtering the report real time by nodeid and putting it on its own left side navigation page off the node details pages. Alert history for node. Thanks, Christian

Yeah this would need to be a custom SQL report in report writer as the friendly premade reports limit you to one SQL view at a time. Database manager is the best place to go reading tables, right click to see available columns, then back in report writer joins will be your friends.

Having a lab system, 30day eval is always free, to turn on all the alerts and see how they work is a great way to learn. These are the ones I use or have built and rely on the most - looking forward to others input... Node reboots hardware health is critical (lists items in email) interface, node, application down I have…

we looked into it and found only a SQL trigger would work. Our two options appeared to be base our groups' dynamic queries on a custom property whose drop down list we managed (as Orion admins knew to create groups when adding new values) or just create a query to keep up with what engineers selected. Did the later with a…

Late to the post but beside being able to search by VLAN in the UDT search field you can also make a report for it - here's a quick one I created that you can add a where vlanid = 'xxx' at the end to limit your response. SELECT n.caption, up.portifdescr,up.portdescription,up.speed,case up.duplex when 0 then 'None' when 1…

heck you can drop your netflow tables from NPM after upgrading to NTA4.0!

Sorry to veer off the thread here but having just done this why not have agents pushed into the server images or post-build installed with GPO and/or config management? The only reason I use discovery anymore is for on prem snmp, vcenter, and network gear.

‌not sure if others are doing this better then this approach but I make my groups by dynamic queries. I add a node custom Property for something like system then run the dynamic query based on node.system begins with [enter your text here] GO back and add the system value to your nodes you want. Now they will be added to…

I just checked and was surprised to see we are just getting temperature for Hardware Health on our 7k's running 6.1 and 6.2 - anyone seeing powersupplies etc?

I've done this off and on for the last few years. I think the first place to start would be building groups for your services- even if just manually adding the nodes and/or apps at first. The idea is to get the service group icon to indicate the RAG status you expect. It's a combination of which objects are primary in the…

What about using custom properties which are easy to update via API and can be used in alert actions?

‌other then nats and port forwarding you could deploy either additional polling engines which would be local or the new SAM agents if they are windows servers.

I have preferred to use an application view instead which I like much better but here's something that should work for you: SELECT e.serverName as Node , '/Orion/NetPerfMon/NodeDetails.aspx?NetObject=N%3a' + ToString(n.nodeid) AS [_LinkFor_Node] , '/Orion/images/StatusIcons/Small-' + n.StatusLED AS [_IconFor_Node] ,…

As an aside- the new webbased report tool includes time summarization without needing to group by.