mesverrum · Observability Architect · ✭✭✭✭✭

in the database you will find the credential under apm_componentsettings IIRC, it'll have a key like "__credentialid" roughly, and an integer. I want to say that -4 is inherit from template, -3 is inherit from node, and any actual number would map to a credentialid on the credentials table where the credential is listed as…

It's been on my to do list to re-write a discovery script I have to make it look at the existing nodes and just build a rediscovery of existing nodes like "whatever filter i want" I'll update the thread with it when I get done, it's not terrible complex if you just rework any of the existing examples of a discovery script…

NCM saves a text copy of all the configs as it collects them to your polling engine by default, you could just copy that entire folder Where are the configs stored locally on the NCM server - SolarWinds Worldwide, LLC. Help and Support

There is a possible spot in NCM Settings > Config Settings > SNMP Config Transfer Timeout Default is 4 min

So the problem you are running into it kind of common with SNMP tables, the name/description you are looking for probably lives on some entirely different table than the one that has your attenuation values. The GUI for that tool really only lets you pick label columns that are in the same table, but if you can find the…

Did you try CPULoad > '90' and (Vendor Like 'Cisco' OR Caption Like '*SWNPM*' OR Caption Like '*UPS*')

All of the above. The way Windows handles interfaces with multiple ip addresses is to convert all the source ips and destination gateways to binary and then compare them to see which has the most bits in common and then pick that as the source. There's a post by aLTeReGo i believe where they explain how to specify the…

This was kind of a beast to wrangle up since the newer version of the alert tables are a little cryptic. This is was written to be used as a Custom Query resource SELECT ac.Name, ac.Description, ac.ObjectType, ac.CreatedBy, aa.CategoryType, a.ActionTypeID, a.DescriptionFROM Orion.AlertConfigurations acjoin…

All I can remember from playing with it in the past is that if I ran it as a non admin user I got errors with it not saving the connection data. Never dug into it to see why and how since I had admin on the machines I ran it on, you might want to open a support case to have them look into it.

You can't really stop it, that is the way directlink works, it auto logs people in without prompting for any credentials. You could set it up so automatic windows login is enabled and I think that would take precedence over getting automatically punted to the DL account, but in most cases I find directlink to be enough of…

Natively you cannot do custom percent bars and the like, for that sort of thing I either simplify my requirements down to using icons, which you can make happen with SWQL, or I would move to something like SSRS. In SWQL you can do something like this, with case sensitive icons to indicate breached thresholds. If you looked…

So the "other" option basically means you have to come up with your own method of updating the DNS record since there is no built in way to handle it. You may not directly manage the DNS server in this case, but I can assure you that if your IT team reaches out to them there is some method they would use to automatically…

The views and actions tables are relatively small in terms of SQL tables, so there is almost no direct performance gain to be had from cleaning them up, but in terms of general housekeeping there is a case to be made for keeping your instance neat and orderly. It's probably not a major priority though. There's no built in…

So to start off the troubleshooting, is your primary server up and running already? Can you log into the Web console there?

In SQL joining all these different kinds tables with different kinds of things on them can be kind of a mess with a lot of the data duplicating in columns, but this should get you the raw info you asked for as a custom swql query. It would be possible to clean this up and make the output a lot nicer but it is going to take…

The two queries stand alone, if you don't have the vlanid info tagged in ipam then just use the first one, if you do then the second one is a possible solution to finding out what subnets are on what vlan. As for the error, it's timing out. The syntax is good but I suspect you have too many devices/interfaces for it to…

Most of this stuff is easy to get to, the tricky part is going to be the subnet info. select distinct n.Vendor, n.Caption, n.IP, n.Location, n.CustomProperties.Building , n.NodeVlans.VlanId, n.NodeVlans.VlanName ,case when i.Caption is not null then i.Caption when map.IfIndex is not null then 'Assigned to Unmonitored…

Web interface typically displays in the local time of the machine you are browsing from, but lots of the raw data is stored in UTC. I'd assume that was the most likely cause here. -Marc Netterfield Loop1 Systems: SolarWinds Training and Professional Services * LinkedIN: Loop1 Systems * Facebook: Loop1 Systems * Twitter:…

Despite them telling you that tcpdump is a no-go they are going to need some similar type of deep packet inspection tool to get that granular, Wireshark or something off their firewall maybe. Netflow (referring to the protocol itself, not the Solarwinds application) does not collect data on packet size at all, so that's…

Ah I assumed that this tool was already working and you wanted to add additional columns of data to it. Going at it from that perspective my first thought is, when you try to execute that SQL query from the orion database manager tool or sql studio does it return the results you expect from the beginning? If you arent…

I use mine as the travel bag when i go on site, which is like 20-30 weeks a year. Right now I'm on a gig working from home for like 30 days in a row so my bag is tipped upside down with all my stuff scattered in the guest room until I get motivated enough to sort it out. It's starting to show just a bit of external wear…

You can see in the trigger action string what it is passing to your other tool. Add any additional columns of data to the SQL: Select ....} part and it will dump them all in. Should be pretty straightforward if you know how to find things in SQL.

The statistic threshold in this case would be the value you need to set, i don't have a tomcat server handy but I suspect it would be reporting in bytes, so something like critical less than 2000000000?

So there isn't a metric that diectly says "community failed" anywhere, which is why people have built lots of reports of the years on indirect indicators that the polling isn't working. The way I do it these days is this SWQL report select n.caption as [Node] ,n.detailsurl as [_linkfor_Node]…

So starting with the assumption that all your interfaces are monitored all the time, I'd do this one in SWQL: select count(t2.interfaceid) as Interfaces, t2.date from (SELECT distinct datetrunc('day',DateTime) as Date, interfaceid FROM Orion.NPM.InterfaceAvailability ia where datetime > addday(-30,GETDATE()) group by…

Look on the NCM summary page for a widget that has advanced search, I use that for what you are describing. It's not super obvious beside that one widget at the top of the out of the box view

Fairly certain that there are no verbs for creating ip sla operations, hence the lack of info on the topic. In cases like this I would probably re-create one of the objects by hand in the GUI, check the database to see what tables were impacted, compare that to what I had in the old database and develop some SQL scripts to…

It has been a while since i had one fail, but I believe if you put a hostname and it can't be resolved it fails so it appears to me to just be doing a dns lookup. I know that IP addresses always pass even if they don't exist on the network though. -Marc Netterfield Loop1 Systems: SolarWinds Training and Professional…

I'd wager your servers are in a -5 UTC time zone? When you build custom widgets it's very important to confirm if the data in the DB is server time or, more likely, UTC and adjust your script accordingly.

When you are checking for interfaces like this you should start with doing an snmp walk against the router when it generates the message to see which interface is associated with that . There is a table that should relate #448304 to an interface name and apparently that interface is not being monitored in NPM. That is the…