Intermittently losing CAPWAP connections on BR200s connected by DSL

  • 1
  • Question
  • Updated 4 years ago
  • Answered
Some of the BR200s that we manage are connected by low-bandwidth ADSL connections. Maybe once a day or every two days several (2-3) in different geographic locations will drop CAPWAP and alarm:

Device XXXX: The CAPWAP connection with HiveManager was lost.

Presumably it only loses CAPWAP for just enough time to trigger the alarm and usually by the time we log in to HMOL it's back online.

Is it possible that the location is just saturating their DSL pipe intermittently, and therefore dropping CAPWAP UDP traffic briefly? If so, could this be resolved by forcing CAPWAP over TCP/HTTP?
Photo of Steven Bateman

Steven Bateman

  • 65 Posts
  • 12 Reply Likes

Posted 4 years ago

  • 1
Photo of Andrew MacTaggart

Andrew MacTaggart, Champ

  • 483 Posts
  • 86 Reply Likes
http://www.aerohive.com/330000/docs/h...

Dead Neighbor Interval: The length of time that the CAPWAP server and client wait for responses to their echo requests before determining that their neighbor is dead—that is, that the neighbor is no longer responding—and that it is safe to close the CAPWAP connection to that neighbor. The default neighbor dead interval is 105 seconds. The minimum setting for this is twice the CAPWAP echo request interval, and the maximum is 240 seconds.

http://www.aerohive.com/330000/docs/h...

Photo of Steven Bateman

Steven Bateman

  • 65 Posts
  • 12 Reply Likes
Of course, right in front of my nose the whole time...

Thanks!

Late edit: Well it seems like these settings are only available for on-prem HM users, if I'm not mistaken. I guess it's time for a support case as the problematic VHM is HM Online.
Photo of Steven Bateman

Steven Bateman

  • 65 Posts
  • 12 Reply Likes
Later in this thread it address the issue of CAPWAP settings being unmodifiable in HMOL instances.

http://community.aerohive.com/aerohiv...

Consider this a vote for a feature request. This would greatly help separate wheat from chaff when managing many BR200s. We are currently attempting to address the slow connectivity issues as well, but that's slow going...

Also, is my original question a possible workaround? Or should I work with support to attempt to optimize the location of our server?