netlogix

Another question would be, why does it show ICMP polling for 10014 (662+8020+1332)? How can you ICMP status poll an interface or volume? Should ICMP polling only be 662? You can kinda poll an interface by ping, but you don't know for sure what interface was used by the ping, just that a node with that IP replied.

For the way to use the _, it would be [_] (i.e. like '/a1[_]test' (that is a [ _ ] with out the spaces) I don't how to do a group for 'like', but, why are you using like? Could you use VolumeDescription in ('/a1_test','/tsmtmp','/tsmalog'). I don't see any wildcards, so you don't need to use 'like'. Also, you could add a…

Might be able to pull the alert SQL query into the message, but it would probably be ugly. A better way could be to put information in the Alert description and include that. I am not positive what the variable is, but you could try "${AlertMessage}". When you are creating the email there is an insert variable wizard that…

I just started working on this myself, so far, I have a "Rule" setup that does a similar thing. Open NCM Policy Reporter Create a report i.e. "Switch Config Audit" under that, create a Policy i.e. "End user Switch" Assign Switches to that (i.e. selection criteria where Name contains "IDF") Create rule like - Check off…

(Edit: create custom sql alert - subtype node) JOIN CPULoadByDays ON CPULoadByDays.NodeID = Nodes.NodeID WHERE (avgload > 25) AND CPULoadByDays.DateTime >= dateadd(dd, -5, datediff(dd, 0, getdate())) GROUP by Nodes.NodeID, Nodes.Caption In the alert, use ${SQL:SELECT Avg(CPULoadByDays.AvgLoad) from CPULoadByDays where…

I would also like this feature. I have some common SQL servers that house multiple App's databases, so I have an app template that is for multiple apps on that sql server, but the services and processes are in another app template assigned to the app server. I have worked out other ways to get things looking right, but…

I am able to create an alert that will only be checked between 5 am and 4 am (23 hours) and an email alert that will only run between 7 am and 6 am (23 hours), and it works... Maybe some specific combination in your setup. Can you share a little more?

Some of the distributions Net-SNMP conf file is so complicated, you have to enable any and every MIB you want to poll, try to reduce the complexity of the snmpd.conf file. i.e. /etc/snmp/snmpd.conf agentAddress udp:161 rwcommunity private then restart snmpd If it works after that, then you would need to either use the…

I think I get what is going on, this is left over stuff that is to help when the server reboots to keep things working well, but now it's a pain. To get rid of service dependencies, open the registry and go to: HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\services\ and look for all the SolarWinds... Keys and find ones that…

I think you are monitoring the switch ports not the server's nics, is that correct, if so, you need to think of it from the switches perspective. The switch is transmitting the unicast data out through the port to the server (Transmit). If you thought about it in the port or server's perspective, it would be receiving it,…

Why not use Custom Properties? If you can't do that, you could ${SQL: Select Blah from [OTHERSQLSOURCE].[OTHERDATABASE].dbo.[Table] where [OTHERDATABASEKEY] = ${Nodes.OTHERDATABASEKEY}} (if you are using HTML email, replace the inner {} with @) But you need to do all the linking in the SQL server and test it through Visual…

Is this a unused forum/"space"

yes, but if you are still on Server 2003, I would suggest building a new server 2008 R2 machine - if you use the same hostname and IP, then there is nearly nothing to do. If you are just P2V'ing, then it should be a piece of cake (as long as the OS is happy).

you could mark those interfaces as unplugable. That might take care of your alerting part if the alert is for status = Down because status will change to unplugged instead of down.

also, you may want to try it from the NPM server (the https://IP), because the could be some type of access list protecting the vCenter server. Or maybe the vCenter server was installed using non-standard ports. (ooh how would solarwinds handle that!)

IVM - part of NPM for polling VMware 1) I am not sure what you mean with this, but with 5.0 you must have NPM installed to integrate APM/SAM in the web console now. In that last release or two, there was a way to have APM and NPM separate, but share a database and therefore a web console (I have no idea how, but I guess…

it would need to be a custom property or a subnet or a caption/hostname pattern or a list of captions. How do you know it's a SQL or exchange server?

Yes, this something I would like too, but from what I know, right now SAM/APM only accepts numerics , so you could select Select ABS(Count(Name)-<INSERT_COUNT_OF_DBs>) From sysdatabases, then set the warning and critical at > 0. Not pretty... but that doesn't get renames either or a drop then create new. But it is something

you can export it many different way, but in any case, even if you left the data in Orion's syslog table, you might have some issues getting the data into your next syslog server. One way would be to just backup the whole orion database, then restore to a new name when you get to the need to do something about importing…

The dependencies setup is the best way, but if you do that, make sure your alert says to either alert when down, or if it says alert when not up, then include not up and not unreachable (and not unmanaged). Be careful with suppression, suppression suppresses the alert if the suppression is true at all in the database. So…

I am pretty sure it is the SQL thing, adatole wrote something like it in this. he used "is empty": Where ANY of the following are true Where ALL of the following are true ALERT_CPU is empty CPU_LOAD is greater than 90 Where ALL of the following are true ALERT_CPU is not empty the field CPU_LOAD is greater than the field…

mdriskell - I gave it a shot - that was interesting - I think I got it though. I don't know if the report writer will like it, but try this: Select EventTypeName, isnull([Shift1],0) as Shift1, isnull([Shift2],0) as Shift2, isnull([Shift3],0) as Shift3, (isnull([Shift1],0) + isnull([Shift2],0) + isnull([Shift3],0)) as Total…

Yeah, or like the way the SNMP string works. You don't need to re-invent the tech, just copy to another place.

From what I understand, basic alerts are still available only for backward compatibility. Every new alert should be done through advanced alerts. If you have anything in basic alerts, you should figure out how to convert them to advanced alerts or start getting info on how to do it that. If you can't do them with advanced…

The "Test Alert" pays no attention to the trigger tab, if just fires of the actions tab, can you post a screen shot of the trigger tab?

I think it is binary data stored in that table as "Screenshot". here is a post about how to write a script to save it. asp.net - How to Download A file stored in SQL DB in Binary Format - Stack Overflow

Check this post out: TIPS & TRICKS: Stop The Madness: How to set alert thresholds per-device He did another pretty cool one here: TIPS & TRICKS: Stop the madness! Avoiding alerts but continuing to pull statistics.

You get 2? That is weird, I don't. Are you using the dynamic IP option in the node properties? I only have 1 polling engine, maybe it has something to do with that...

I had an issue like this every time the server lost access to the DNS server even for only a second or two. I fixed it by changing to an IP in the SNMP config instead of a host name. But of course if you are using IPs, then it doesn't apply. If you want, you could just try it on one or two servers that are having issue…