iunderwo

The double-listing of the interface name was accidental. Oops. I didn't take that field into account when I wrote the script, mostly since the searches I do are usually done within interface descriptions. I also added a SortBy function which allows an arbitrary sort order to be applied to the interface search as well.…

Not to toss in a bit of sarcasm...but don't let Windows Update spoil you. // Ian Underwood - Service Management // Level 3 Communications

Are you looking at a traffic problem, or a routing problem (i.e. multiple RFC-1918 sites using overlapping space)? SW is really pretty light in its traffic requirements, so a proxy isn't really going to save you much. However, a proxy may be useful if you have a number of other applications which are polling the same…

For grins, I ran a MIB walk on one of the systems I manage, looking to find out where the process information is stored. Interestingly enough, it looks as if the App Monitor is just using the names as part of the hrSWRunName in the HOST-RESOURCES-MIB. On the plus side, Windows does report any parameters and the startup…

You need to disable and re-enable the alerts in the system manager. There are several other alerts that have no timeout or reset values, so once they've been set, they stay that way. A timeout value on these kinds of alerts has been requested before. // Ian Underwood - Service Management // Level 3 Communications

You'll need the account manager for 6.4. The user data wasn't included in the SQL database until 7.1, IIRC. You can download the utility from the Unsupported section in the website. // Ian Underwood - Service Management // Level 3 Communications

It sounds like you've explained the reasoning enough and they still don't get it. I'd go to a higher layer of management and explain that you're looking for 1 IP to go through. If that doesn't work, add in another NIC and do what you need to do behind the firewall with that instead. // Ian Underwood - Service Management…

You can always download an evaluation copy and make sure the application meets your needs. That's what it's there for. // Ian Underwood - Service Management // Level 3 Communications

You can select "Show SQL" in the report writer and it'll spit out the query that it's using. // Ian Underwood - Service Management // Level 3 Communications

That requires some Custom MIB support which is often requested. Hopefully, we'll see it soon. // Ian Underwood - Service Management // Level 3 Communications

Cindy, I was not able to get the auto-start to work properly for my SysLog server. I was only using SQL 2005 in a development environment. Since I use 2000 in production, I haven't had any problems. Once v8 comes out, I'll try tackling it again. // Ian Underwood - Network Engineering // Boston Stock Exchange

I found the SQL install not to be too bad ... especially since instances were a lot easier to understand. Outside of that, there are some problems with SQL 2005. I found that my syslog server and beta alert engine would not automatically start. I use SQL 2000 in production and it works very well. // Ian Underwood - Network…

Looking at this and thinking the same thing, would it be feasible to send a syslog when a node goes down in a specific format, and then let the syslog manager send an alert from there based on that format? The Syslog manager does have threshold alerts, which is nice. // Ian Underwood - Network Engineering // Boston Stock…

Depending on what kind of function you're looking at, it might be easier to do any of the console work via Remote Desktop. The only problem I'm aware of is the 8-bit video limition with Win2k servers. This might not be an issue with the Win2k3 Server. The remote manager is a per-seat license, I believe, which would make it…

That's excellent! I didn't think of using a function (not a coder by trade). Works like a charm! // Ian Underwood - Service Management // Level 3 Communications

That's true...and would probably work if you're writing an advanced SQL query. Otherwise "start with" will be faster to run...or at least that's what my experience has shown. // Ian Underwood - Service Management // Level 3 Communications

Wildcards are not applicable in SQL. You should change the condition to "Records where 'system name' starts with 'R'". // Ian Underwood - Service Management // Level 3 Communications

Orion manages networks using fairly common SNMP items. Compiling MIBs and working with enterprise MIBs are not options currently. Working with specific OIDs is also not part of the package. A MIB walker is a part of the engineer's toolset. While working with OIDs is a basic SNMP function, software which manipulates these…

The only workaround is to move your stats out to a 10-minute interval for collection. The alerting engine isn't really very intelligent. // Ian Underwood - Service Management // Level 3 Communications

All the monitored data is saved in the database. Since it's in SQL format, you can work with it any way you need to. // Ian Underwood - Service Management // Level 3 Communications

I'm going to need to side with JMP on this one...that's a heck of a price tag...and I thought wheeling and dealing was required to get my SW installation. // Ian Underwood - Service Management // Level 3 Communications

In the integrations, though...whose front-end is it supposed to use? // Ian Underwood - Service Management // Level 3 Communications

NTLM can work with Orion provided you're not doing anything funky with usernames. It's discussed in-depth in the Automatic Login section of the administrative manual. One thing I have observed is that the auto-logout is based on activity, so if you have a page auto-refresh every 5 minutes or so, the account should never…

There's a tool which you can use to assign different nodes to different pollers ... so it shouldn't be a problem. I think all the pollers need to be down when assignments are changed though ... but I could be wrong. Unrelated question though ... why not do 3 pollers and have a hot standby ready in case one of them decides…

The more elements that are alerting on your network, the more impractical having a sound is going to be. Several years ago, at another company, one group thought that would be a good idea which lasted about 6 hours until one full glass of water went sailing through the air. Sound gets annoying pretty quickly unless it's…

If you purchased the program, you should email support and get them involved. That's going to be a much faster turnaround time than this forum will be. // Ian Underwood - Network Engineering // Boston Stock Exchange

You might want to see what happens if you install Cirrus first. The two applications don't share the same database, so there is no preference to the order. That might work ... but I haven't tried it. Failing that, the only option available to you on a Trial basis is to install the two applications on two different machines…

IIRC, the interface type is taken from whatever SNMP says the interface type is. With Cisco devices, the interface type depends largely on the encapsulation used. HDLC shows up as Serial. PPP, Frame Relay, and ATM show up accordingly. This is regardless of the CSU/DSU type. NG is right though, you'll need a custom property…

There is the possiblity since the sysUptime is stored in the database now. However, I don't think they should make it an alert unless they have a reset condition for it. For example, alert on reboot, but reset alert after it's been up and stable for an hour or so. It's a useful alert, but not when it lingers as triggered.…

You should be able to get into graphs as granular as 5 minutes. Data taken at lower intervals will show up as a min/max/average point and bar graph instead of as a data point. One thing you need to bear in mind is that if you change any of the intervals from the system manager, you need to click the button which allows you…