smargh

It's not the only limitation of monitoring volumes. Orion's requirement to monitor [Windows] volumes by only SNMP causes great frustration for me. Even changing a volume label on the server causes it to become 'unknown' to Orion, requiring a deletion of the volume, re-discover and re-typing of the custom properties.…

The problem is the custom property integer type is can't contain above 2GB when specified as bytes. So, I use megabytes for that field and then use a custom SQL alert for volumes. I have two alerts - one is a 5GB default for all volumes that don't have a specified minimum free space, and another for those that have the…

For the trigger definition, did you do [status] [has changed] ? That won't work - "has changed", as far as I'm aware, is for uptime only. This is what the TriggerQuery SQL will be in your AlertDefinitions table for this alert: SELECT DISTINCT CustomPollerAssignment.CustomPollerAssignmentID AS NetObjectID,…

Windows (particularly 2008 and newer) includes lots of unnecessary stuff in the SNMP interfaces table, and Orion doesn't know that anything needs to be filtered out. If me or you were coding Orion instead of SolarWinds, we would put in some logic to exclude things like the Miniport virtual adapters and the Teredo ipv6…

The "Node Added" criteria works fine, but "Node Removed" either won't work or will be unreliable depending on deletion timescales because the row in the Nodes table for the deleted node will no longer exist.

To be Frank, I haven't done anything with the NetFlow module - I've only done basic research. Perhaps you could expand reporting into the areas of traffic analysis at a lower level than just looking at the top talkers or highest traffic levels. From what I've seen of the features of ManageEngine's NetFlow Analyzer, it…

SELECT (SELECT COUNT(DISTINCT Nodes.Caption) AS COUNT_of_NodeName FROM Nodes WHERE ((Nodes.Status = '1') AND (Nodes.Link = 1))) + (SELECT COUNT(Interfaces.InterfaceName) AS COUNT_of_Interface_Name FROM Interfaces WHERE ((Interfaces.Link = 1) AND (Interfaces.Status = '1'))) AS ColumnTitleHere

The value is just an integer, with a maximum value equivalent to around 2 gigabytes when converted to bytes. This makes it unsuitable for directly comparing available & used space, because they are in the database as bytes only. I use custom fields to store limits for megabytes. Then, use a custom SQL volume alert but…

I do this with built-in Windows feature & Orion trap alerts. On the server where the events will be logged: Configure the SNMP service (properties of the service) in the "Traps" tab to tell the OS where to send the traps. A service restart is sometimes necessary after this (especially on 2003), but often not on 2008 R2.…

Orion sends email via SMTP, so Exchange isn't aware that the distribution list name is the intended recipient. You need to use one of the SMTP email addresses that you've assigned to the distribution list. Alternatively, if you've specified multiple recipients, be sure to use the format blah@blah.com,blah2@blah.com

Do this in the "Manage Nodes" section, not within "list resources" In the Manage Nodes window, find the node, click the + to expand, tick the box for the interface, choose delete.

I use SQL Server Agent to check custom pollers. Perhaps you can modify this to suit your own requirements, or it might give you ideas for a built-in alert for checking specific custom pollers etc. -- -- Check SolarWinds custom pollers are updating correctly. -- DECLARE @MessageCount INT DECLARE @Messagetext VARCHAR(2000)…

I'm not sure what you mean precisely by linking to "the specific alert", but I include these links in my alert emails for volumes & nodes: Orion Volume: yourservername:port/.../VolumeDetails.aspx Orion Node: yourservername:port/.../NodeDetails.aspx

Just RDP to it as usual, then run: tscon %sessionname% /dest:console That will disconnect the RDP session and show the user's session on the physical console. I've done this before, when when pre-profiling laptops for remote users.

NPM does this on Windows whenever the label of the volume is changed on the server itself. Perhaps it's the same on Linux in that the SNMP OIDs of the volume change after it's named or otherwise altered in some way? I opened a ticket for this with Support a week or two ago - they confirmed that there's no workaround except…

At the moment, yes. I use a mail server of "mail.server.set.by.local.hosts.file" :) Alternatively, add an alias for your mail server to your DNS system and use that in all your alerts. This is a known change or feature request.

Check the email options in all your alerts (trigger & reset areas) for any unexpected spaces or commas etc. Check the To:, CC: and BCC: fields. I have seen your error appear in the logs when there is a space or other unexpected character in one of the recipient fields, and it may also happen with other fields also.

I use a custom SQL node alert for this purpose: Trigger condition: WHERE ( (DATEDIFF(mi, Nodes.LastSystemUpTimePollUtc, getutcdate()) > 5) AND (Nodes.Status = '1') ) Reset condition: WHERE ( (DATEDIFF(mi, Nodes.LastSystemUpTimePollUtc, getutcdate()) < 30) AND (Nodes.Status = '1') )

Don't use LastSync. I tried it for this exact scenario, but it's not reliable and I think is only intended to show the last DB sync rather than the "last successful poll" time. I use this custom SQL node alert. It will only work on recent versions of NPM. I've only been testing it for a few days, but it seems to work okay…

You can get temp/humidity monitoring add-ons for APC UPSes, which can be polled via SNMP rather than the sending of traps. As a bonus, the device & required internet devices would remain up if, for example, the power failed so that the air con stopped which could cause the high temperatures :) Of course, as with all…

What version of NPM are you using? I think you're on 10.0 or earlier. Edit: From your previous posts, perhaps you're on 9.5.1. I have a suspicition that those volumes have all been renamed on the server. If you rename a drive on your Windows server, Orion will lose track of it. With Orion 9.5.1, and perhaps earlier…

This may apply to you: blogs.technet.com/.../source-ip-address-selection-on-a-multi-homed-windows-computer.aspx If so, you'll need to ensure you're using a suitably up-to-date OS (2003/2008 with hotfix, or 2008 R2 with hotfix/SP1) then delete the additional IP and add it again using the netsh command with the skipassource…

I do HSRP state change alerts, which do appear as active alerts on the homepage as you would expect. The bad aspect is that universal pollers only poll every 10 minutes. I use the cHsrpGrpStandbyState OID, and have two custom properties: HSRP_Active_IP, which contains the normal active IP address, and HSRP_Standby_IP,…

Does the router have a default gateway set? I've seen that before when the router has rebooted but the existing settings hadn't been saved.

This seems to work okay. Note that I've used NodeID to avoid having to use joins and make the subqueries longer. There's probably a better way of doing this, but this is perhaps the easiest to follow & expand on for people who aren't familiar with SQL.You might want to add additional conditions with RESPONDING = 'Y' to…

I believe this would need to be a custom SQL alert which includes a select subquery to get data from two volumes for a given node. This would, I think, not work if you anyone else wanted to check more than one volume on two different servers, but it should be do-able on just one node.

It's doable, but would involve several steps, starting with telling the Windows server's SNMP service where to send traps, then using evntwin.exe on the servers to tell them to send event ID 1074 to Orion as traps. Orion can then be configured to run an action when receiving that trap, probably sending an instant email &…

What did the BSODs say?

Perhaps you deleted a node recently and some things in the database haven't caught up yet. Run a manual database maintenance. For me, doing that normally fixes the error.

1) Could you post a screenshot of the Orion pages which shows the table for this access point connection status? I find they're useful to visualise what information I need to alert on. 2) Do you have one alert per node, or one alert covering lots of nodes?