After upgrading to hub 7.x, some customers reported experiencing instability and connectivity issues between tunneled hubs, including queue connectivity failures and frequent tunnel disconnections.
The settings prescribed in this article have been found to remediate these issues and to also be applicable in improving connection stability in hub 8.x versions as well.
It is considered typical, and not necessarily a problem in and of itself, for tunnels to disconnect occasionally; hub 7.x architecture contained numerous improvements intended to reduce the impact of these minor events by ensuring that the hub and tunnels can recover quickly from such events, without any noticeable impact to the system or loss of data.
However, if the tunnels are found to be disconnecting so frequently that it is causing a business impact (such as difficulty configuring probes and a barrage of tunnel/queue alarms), then the following settings may help mitigate this behavior and stabilize the connectivity.
These settings should be applied to each hub in the environment.
Edit the hub.cfg and find the <hub> section. In that section, adjust the following keys (most should exist already, but if any do not, you may add them.)
postroute_interval = 120
postroute_reply_timeout = 180
postroute_passive_timeout = 300
hub_request_timeout = 120
tunnel_hang_timeout = 300
tunnel_hang_retries = 3
*** Please note that if you have a recent version of the hub, you may already have the settings above ***
After saving the modified hub.cg, restart the "hub" probe to pick up the modifications.
On systems with very low network latency and fast response between hubs, it may be beneficial to decrease tunnel_hang_timeout to 120, or even 60.
Also for improved response with the Infrastructure Manager, consider setting the 'probe request timeout" in "IM" Tools -> "Options" up to 60 which should improve performance (the default is 10).