Solarwinds HA - default and recommended settings for failover time?

robert777

Greetings.  Question about Solarwinds HA:

From the deployment documentation:

------------------------------------

Default interval to consider a member as down in a pool

Define how long the active pool member can be down before a failover occurs. Provide the interval in seconds.

------------------------------------

Digging through this, I am not able to locate the default # of seconds for this setting.

Can someone let me know what  is the default setting for this value? 

Full disclosure: I don't have direct access to the Solarwinds system in question - doing some research for another engineer on site. 

Having trouble with too many failovers for transient conditions.  Can anyone recommend a threshold level that may help with this problem (double the default?).

Thanks for any advice from the community.

  

robert777

Greetings again. 

I found this thread that looks like it addresses my question:

thwack.solarwinds.com/.../increase-the-maximum-value-in-ha-pool-default-interval-to-consider-a-member-as-down-in-a-pool

default value: 32 seconds 

maximum value: 60 seconds

Thank you @HerrDoktor  for raising the issue and putting in a feature request for a higher maximum value.

60 seconds will not be long enough for the particular situation that prompted my question.

Does anyone know of a workaround for the 60 second maximum on timeout threshold?  Like a direct database edit, or something?  Please post if any ideas.

In a previous job I always wondered why they used an external load balancer to monitor SW for failover rather than native HA functionality.  Now I think the very-short maximum threshold had something to do with it.   In that setting the load balancer was configured for 180 seconds Information Service down or unavailable on the Main Polling Engine for a failover.

I welcome any input from the community on how to configure a high-than-60-second threshold for Solarwinds HA failover. 

Thanks!