Uploaded image for project: 'Couchbase Server'
  1. Couchbase Server
  2. MB-23036

Autofaliover of node taking more than 5 sec when a node's firewall is enabled

    XMLWordPrintable

Details

    • Bug
    • Resolution: Incomplete
    • Major
    • 5.0.0
    • 5.0.0
    • ns_server
    • Centos
    • Untriaged
    • Unknown

    Description

      1. Create a cluster with 3 nodes and 1 bucket with data in it
      2. Enable autofailover and set timeout to 5
      3. On on of the nodes (non orchestrator) enable the firewall

      We expect the failover to occur after 5 sec (8 sec giving around 3 sec buffer?). But the actual failover is reported to be started after excess of 10 sec in the UI logs.

      Following are the UI logs from one of the test. The settings was enabled on 7:11:13 AM and the failure injected on 7:11:18 AM. We expect the auto failover to be initiated on 7:11:23 AM but as can be seen below, it was started on 7:11:28 AM

      Failed over 'ns_1@172.23.98.80': ok ns_rebalancer 000 ns_1@172.23.98.79 7:11:28 AM Mon Feb 27, 2017
      Starting failing over 'ns_1@172.23.98.80' ns_rebalancer 000 ns_1@172.23.98.79 7:11:28 AM Mon Feb 27, 2017
      Enabled auto-failover with timeout 5 auto_failover 000 ns_1@172.23.98.79 7:11:13 AM Mon Feb 27, 2017

      Attachments

        No reviews matched the request. Check your Options in the drop-down menu of this sections header.

        Activity

          People

            bharath.gp Bharath G P
            bharath.gp Bharath G P
            Votes:
            0 Vote for this issue
            Watchers:
            7 Start watching this issue

            Dates

              Created:
              Updated:
              Resolved:

              Gerrit Reviews

                There are no open Gerrit changes

                PagerDuty