Announcement

Collapse
No announcement yet.

AD replication failover and MaxFailureTimeForIntersiteLink (secs)

Collapse
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • AD replication failover and MaxFailureTimeForIntersiteLink (secs)

    We are planning/testing a hub and spoke site topology with 2 hubs. We are
    also controlling AD replication by disabling "Bridge All Sites" i.e.
    disabling the default transitivity of all Site Links.

    We are using Site Link Bridges to control the generation of new connection
    objects when one of the hub sites goes down so that a branch office will
    start replicating with the alternate hub if it's primary hub goes down.

    This works but the default failover time seems to be 2 hours per bridgehead
    server in the hub i.e. after 2 hours of failing to contact a bridgehead DC in
    the failed hub site the ISTG then creates a second connection object to pull
    from the second bridgehead DC in the failed site. After another 2 hours (i.e.
    4 hours after the hub went down) then the ISTG creates a third connection
    object to a bridgehead DC in the alternate hub and AD replication is then
    restored. Once the primary hub comes back up then the ISTG deletes the 2 new
    connection objects it had created and everything is back to normal.

    Has anyone used the "MaxFailureTimeForIntersiteLink (secs)" registry setting
    to change the intersite failure time from the default 2 hours? We don't seem
    to be able to get it to work - it always seems to take 2 hours whatever we
    set it to.

    We use "dcdiag /test:intersite /e /q" to see how long it's going to be
    before it considers the bridgehead server to be ineligible to be a
    bridgehead. This is how I saw that it was taking 2 hours per bridgehead
    server in the failed site to generate a connection object to the alternate
    hub.

    Thanks
    David

  • #2
    Re: AD replication failover and MaxFailureTimeForIntersiteLink (secs)

    Interesting case. Found this:

    http://www.informit.com/discussion/i...f-18abc57247c4

    Comment

    Working...
    X