DX Unified Infrastructure Management

In the past when a hub became unavailable we received an alarm for that. Now on 8.20 we do not see anything! What do I miss or what has changed?

On the central hub is shows the missing as RED. Why do I not get a message on such an major event!

Any suggestion or clues here?

regards

Ashley Pietersen

Have you verified that it is actually a case of hub going or down, or just an issue in IM? I've seen this happen a lot in IM on 8.2, especially if there are multiple paths to the hub. Also sometimes IM just reports them as red and as soon as I click them it refreshes them alright. Maybe some timeout have been changed that leads to this or something.

-jon

Yes it is completely gone for some reason. Our remote desktop service to that hub has lost also it connection.

I have configured a sms-alert on any hub-/tunnel alarms. But when it does not fire an alarm I stay in the blue. Which makes me very nervous.

regards

Ashley

This is what I find sofar : (we are running hub 7.80)

https://na4.salesforce.com/articles/FrequentlyAskedQuestions/Hub-7-61-not-generating-tunnel-disconnect-alarms?popup=true

Due to major architectural changes in the way tunnels work in hub 7.61 compared to previous versions, the tunnel disconnection alarms were no longer reliable - false disconnect alarms would be generated very frequently as a result of these architectural changes.   Therefore, product management made the decision to have the development team remove these alarms entirely from the hub -- alarms will no longer be generated upon simple tunnel disconnection.

In the future, our intention is to implement a new, more intelligent form of tunnel alarming, which will be user-configurable and generate messages such as "Tunnel to hub X has been disconnected for Y minutes" where Y will be configurable by the user.   The first step toward implementing these alarms was to remove the portions of the code which were responsible for generating the old alarms.

Currently, in hub 7.61, no alarms will be generated from a tunnel disconnection, however, an alarm will be generated when a GET queue fails to connect or receive data for 2 minutes.  These queue alarms should be relied upon for now instead of the tunnel client alarms.

The severity of these alarms can be adjusted by a configuration key called "queue_connect_severity" which can be placed in the <hub> section of the hub.cfg.  Valid values are 1 (information) through 5 (Critical.)

regards

Ashley

Unfortunately I could not detect any alarms after disabling a remote controller/hub with the above mentioned addition in the hub.cfg .

regards

Ashley

The way we handle these tunnel disconnect situations is the following, we have a nas rule that looks at the alarms that generated for the "Failed to get the queue from hub XXXX". When the count of this alert hits 5x, we then generate a new alarm that says: "Check client hub at: &msg" and our NOC then knows to then check the hub at that site mentioned in the new nas alarm that gets generated. That way we know that if it hits a count of 5 its been down for ~10min and should be checked out to know whats going on.

Thanks fro the response Daniel,

Not sure I understand what you mean. NAS rules do apply on alarms that are raised. In this case no alarms are raised. What do I miss?

regards

Ashley

Today in hub version 7.71 if the hub goes out and tries to "GET" a queue from a remove hub and fails it will trigger an Informational alarm in the following format:

Queue 'get_Alarms_from_XXXX' failed to connect to hub /DOMAIN/HUB/Robot/hub

We have a nas rule that says, if you see this 5x, then generate a new alarm at a Critical Level saying basically:  "Go Check out this hub in IM and if you cannot expand it its DOWN!!! &message"

Understood!