savell

On taking a quick look at the database, there is a Netflowsources table with a column 'enabled'. All the interfaces have a 1 in it, which I assume means don't filter from the list. Wonder if I could set all these to 0 except for the single active Netflow transmitter?? Is there a better way? Dave.

We use the ciscoEnvMonTemperatureStateTable (1.3.6.1.4.1.9.9.13.1.3.1.6). This returns 5 values (1-5) that indicate normal, warning, critical, shutdown or nonPresent. This way the device determines the alerted threshold values - and you can use the same alert rule for many device types (that all have different threshold…

Yes, all the servers are keyed from the same ntp source, and all are in sync (and in the same timezone). Support understand the poller status issue and have a fix in train (the keepalive for the poller status can't run while the topology calculation is updating - and ours is taking a very very long time...) Our…

The above link is a UnDP for a temp table. If you just want the power supply status then this is the one we use Fan status table is We also use this UnDP for the temp state table . Rather than the actual value, it returns six possible outcomes for each temperature probe on the device (i.e. normal, warning, critical,…

Eric, as far as I can determine there is not a good way to do this currently - unless you consider defining 24 rules, one for each interface (although even this doesn't help with the if/then/else logic). I had a similar issue with QOS compliance at an interface level and ended up creating a solution using Cirrus data, but…

Well, 32 hours later the db maintenance job ended. I have no idea why this occurred but will keep an eye on the next run. The maint log indicated that the bulk of the time was used 'Removing summarized older data from InterfaceTraffic' Dave.

Hi Denny, Funny what you miss - I didn't see the little cross to delete the alert fromn the UI (not that I use the RDP session much these days. Yes, I would like alerts removed automatically when a device is deleted. Dave.

An example of how this really hurts. For probably a good application design reason, NCM executes the following query every 20 seconds: select * from policycacheresults where cacheid = '0' Unfortunately, as the table doesn't have an index (or a key), the query takes between 22 and 29 seconds to run. To execute it requires a…

They look like resources added from the NCM integration module. Did the install of the integration module on the additional web server complete successfully? Don't forget that you will also need to copy the security certificate onto the additional web server (http://support.solarwinds.com/kbase/ProblemDetail.cfm?ID=1121).…

I think it's more likely to be included in NCM rather than NPM. Dave.

We also create a custom property, but utilise an SSIS job to parse the cached map objects and populate the field (a nightly process). I'll post a copy of the SSIS script later today when I have the chance. Dave.

We use the ciscoEnvMonTemperatureStateTable (1.3.6.1.4.1.9.9.13.1.3.1.6). This returns 5 values (1-5) that indicate normal, warning, critical, shutdown or nonPresent. This way the device determines the alerted threshold values - and you can use the same alert rule for many device types (that all have different threshold…

Think of a loopback address as a virtual interface. It doesn't have a physical presence, so this makes them quite reliable (i.e. a physical failure won't disable the loopback address). If you have multiple routes for example to a device, then it's possible that access can be retained to your loopback in a event of a…

I have posted this custom report here Please note, it also contains a DTS script to populate a custom property. Dave.

That one was killing us - thanks support! Sav.

I raised this as a bug under case #64126. Dave.

Yes, I think that this would be a useful way forward. On the upside, it would give great flexibility as to how the pollers could be deployed without concern for placement, optimum poller hardware, location etc. It may also allow SolarWinds to be more flexible concerning the number of APM component licences for example, and…

Kind of worked out how to deal with this situation. Basically I created an SQL Server DTS script to populate the Cirrus comments field with the appropriate interface function types for each device (from the detail already captured in the Cirrus Interfaces table). This runs nightly to update any changes, and then each…

Upgrade was smooth and painless for me (approx 4000 nodes over three pollers). Didn't need to backup alerts, reports etc. I also needed to install Application Monitor, Netflow and Additional Web Server - all OK. Small issue outstanding with Netflow but this pre-existed 8.1 (talking with Support regarding this one). I…

I wouldn't have thought that a trial version would have any impact upon something like this. If you are keen to research the cause before logging a case, you can run profiler on the SQL server and capture the SQL statement issued to produce this table (I'm assuming it's just a fairly straight forward count on the Nodes…

As far as I know, 9.5.1. contains bug fixes and some performance enhancements only. Dave.

I'm also having an issue with the APM Additional website (although my error is different again). Here is the first few lines of what I receive when selecting Application Performance Monitor from the additional web site... System.ServiceModel.CommunicationException: The socket connection was aborted. This could be caused by…

I would do this via SQL rather than a script. Try a select first... select caption, sysname, ip_address from nodes where sysname <> caption If you are then sure that the update is required (and the correct updates are listed) - turn it into an update... update nodes set caption = sysname where sysname <> caption Make sure…

We find we have duplicates every now and then. I setup an SQL report to run which shows me any duplicate entries - a quick check every couple of weeks keeps things straight. Dave. Edit: Slightly more helpful if I also post the SQL we use... SELECT Caption, COUNT(Caption) AS NumberOfDups FROM Nodes GROUP BY Caption HAVING…

We are running the latest 9.1 release. I might run a query to tomorrow just to check that there are no TrapID's in the TrapVarbinds table that don't also exist in the Traps table (this should this be the case given the way you described the tables above). If this is a problem, it existed prior to verion 9 I think. I only…

Mmmm... I didn't realise that this was the flow used. It seems a very poor design in terms of providing a usable web interface from a performance standpoint. Anyone from SolarWinds want to comment regarding any changes proposed in this space? I'm thinking that for us the integration module is pointless. I wonder if I…

The best solution I have been able to come up with, is a pattern match across multiple lines (i.e. block of text). For an acl list, this means that you can include the remark at the top of the access list, and the deny any at the bottom. So for an acl like this... access-list 49 remark Management Server access-list 49…

Thanks for the response Andy. Makes it slightly easier on us anyway doing it in a two step approach. Dave.

It's an open enhancement request (). As suggested in that thread, even appending the user name to the Event log messages (add node, delete node etc) would be a great start. Dave.

I'm not in the office to check, but I don't think that you have the correct OID for the temperature status table. Try ciscoEnvMonTemperatureStatusTable (1.3.6.1.4.1.9.9.13.1.3). Dave.