I don't have anything in my database right now that is in a ERROR/DISABLED state, but would this report help you?
Here's that SQL code of the Interface Report from Orion Report Writer:
SELECT
Nodes.Caption AS NodeName, Interfaces.Caption AS Interface_Caption, Interfaces.Status AS Status, Interfaces.Severity AS Severity
FROM
Nodes INNER JOIN Interfaces ON (Nodes.NodeID = Interfaces.NodeID)
WHERE
(
NOT (Interfaces.Severity = 0)
)
I'm using "Severity" in this case because...
Definition of the SEVERITY variable:
${Severity}
A network health score providing 1 point for an interface in a warning state, 1000 points for a down interface, and 1 million points for a down node
hope this helps.
I decided to use SNMP Traps to let me know when an interface became error disabled. I still wouldn't mind having a report to run.
Could you tell me what command you issued to turn the traps on for err-disabled
Another option is to trigger alerts from Syslog messages generated from the switches when ports go into err-disable. It works well for me.
I personally would go with syslog. On top of solving your issue it would allow you to monitor for things that we all should be especially on core devices. I use it for just that on all my most important devices. It takes a little time but once you get the spam filtered out you will know when anything goes on in your devices that really need to be looked at.
In my experience major failures will be preceded by a long line of error messages. Seeing these gives you ample time to get a fix or a workaround in place before complete failure.
Hi
I used syslog and created a widget to show the erro disabled events for the NOC for the last hour.
SELECT Hostname, SUBSTRING(Message, CHARINDEX('putting ', Message) + 7, len(Message)) AS Expr1, count(messagetype) as Count
FROM Syslog
WHERE DateTime >= dateadd(hour,-1,getdate())and DateTime <= getdate()
and messagetype = 'PM-4-ERR_DISABLE'
group by hostname, messagetype, message order by count(messagetype) desc
Kind Regards
James
I did write a report;
This is a report writter report, and is scheduled to runs each night (automagically) and the server team uses it to find and fix potential issues.
Here is the SQL that I used to find interfaces that were not happy for Windows boxes..
Select
n.NodeID
,n.Caption
,n.IP_Address
,n.Cls
,n.ServerLocation
from netperfmon_10b.dbo.Nodes n
where (StatusDescription like '%one or more interfaces are in%') and (Vendor = 'windows')
and (serverlocation <> ' ' ) and (Caption Not Like '%complelent%') and (n.unmanaged = 0)
order by Caption
