Uploaded image for project: 'Couchbase Server'
  1. Couchbase Server
  2. MB-7537

Rabalance button should change status depending on node statuses

    Details

    • Type: Bug
    • Status: Closed
    • Priority: Major
    • Resolution: Cannot Reproduce
    • Affects Version/s: 2.0.1
    • Fix Version/s: 3.0
    • Component/s: UI
    • Security Level: Public
    • Labels:
    • Triage:
      Untriaged

      Description

      2.0.1-129-rel
      steps:
      1. cluster with 2 nodes: 10.3.121.112 & 10.3.121.113
      2. click Remove for 10.3.121.113
      3. stop cb on 10.3.121.113
      result:
      on UI 10.3.121.113 node is Down, but Rebalance button is enabled

      only after cache refresh(F5 or Ctrl+F5) the button becomes disabled.

      Alk, and another question related to rebalance with nodes down:

      So we can run rebalance( through UI or Rest) when any node is not ready to rebalance( down or warmup). If I run a rebalance I see progress on a minute on UI. Does it make sense to start it at all, if any of the servers is certainly not ready yet?
      Because now we can't trigger rebalance on UI after F5) but can via rest ( rebalance failure after 1 min)

      Rebalance exited with reason

      {not_all_nodes_are_ready_yet,['ns_1@10.3.121.113']}

      ns_orchestrator002 ns_1@10.3.121.112 14:26:30 - Wed Jan 16, 2013
      Shutting down bucket "defaullt" on 'ns_1@10.3.121.113' for server shutdown ns_memcached002 ns_1@10.3.121.113 14:26:09 - Wed Jan 16, 2013
      Bucket "defaullt" loaded on node 'ns_1@10.3.121.113' in 0 seconds. ns_memcached001 ns_1@10.3.121.113 14:25:51 - Wed Jan 16, 2013
      Couchbase Server has started on web port 8091 on node 'ns_1@10.3.121.113'. menelaus_sup001 ns_1@10.3.121.113 14:25:49 - Wed Jan 16, 2013
      Node 'ns_1@10.3.121.113' synchronized otp cookie pozzoodiircndvke from cluster ns_cookie_manager002 ns_1@10.3.121.113 14:25:49 - Wed Jan 16, 2013
      Started rebalancing bucket defaullt ns_rebalancer000 ns_1@10.3.121.112 14:25:30 - Wed Jan 16, 2013
      Starting rebalance, KeepNodes = ['ns_1@10.3.121.112'], EjectNodes = ['ns_1@10.3.121.113'] ns_orchestrator004 ns_1@10.3.121.112 14:25:30 - Wed Jan 16, 2013

      1. 10.3.121.112-8091-diag.txt.gz
        512 kB
        Andrei Baranouski
      2. 10.3.121.113-8091-diag.txt.gz
        810 kB
        Andrei Baranouski
      1. cleean_browser_cash_F5.png
        62 kB
      2. rebalance_enabled.png
        63 kB
      3. rebalance_enabled.png
        62 kB
      No reviews matched the request. Check your Options in the drop-down menu of this sections header.

        Activity

        Hide
        alkondratenko Aleksey Kondratenko (Inactive) added a comment -

        REST is harmless. I mean if someone really wants that that somebody can check node statuses prior to sending request. And naturally there's some inherent race here. I.e. just after confirming that all nodes are up, one of them can go down or pending but rebalance request will be sent. Which ok IMHO.

        Regarding UI bug, thanks. Looks like bug indeed.

        Show
        alkondratenko Aleksey Kondratenko (Inactive) added a comment - REST is harmless. I mean if someone really wants that that somebody can check node statuses prior to sending request. And naturally there's some inherent race here. I.e. just after confirming that all nodes are up, one of them can go down or pending but rebalance request will be sent. Which ok IMHO. Regarding UI bug, thanks. Looks like bug indeed.
        Hide
        alkondratenko Aleksey Kondratenko (Inactive) added a comment -

        Hm. I was unable to reproduce. Can you double check it's still there ?

        Show
        alkondratenko Aleksey Kondratenko (Inactive) added a comment - Hm. I was unable to reproduce. Can you double check it's still there ?
        Hide
        andreibaranouski Andrei Baranouski added a comment -

        yes, I can reproduce it on 2.2.0-738

        button 'Rebalance' is enabled when node is down that had been removed(button 'Remove') before
        see screenshot

        Show
        andreibaranouski Andrei Baranouski added a comment - yes, I can reproduce it on 2.2.0-738 button 'Rebalance' is enabled when node is down that had been removed(button 'Remove') before see screenshot
        Hide
        maria Maria McDuff (Inactive) added a comment -

        Andrei,

        Are you still seeing this in 3.0? pls re-assign to Alk for fixing on 3.0 if that is still the case.

        Show
        maria Maria McDuff (Inactive) added a comment - Andrei, Are you still seeing this in 3.0? pls re-assign to Alk for fixing on 3.0 if that is still the case.
        Hide
        anil Anil Kumar added a comment -

        Andrie - Retest this issue in 3.0 and open a new bug if still happens.

        Triage - June 04 2014 Alk, Wayne, Parag, Anil

        Show
        anil Anil Kumar added a comment - Andrie - Retest this issue in 3.0 and open a new bug if still happens. Triage - June 04 2014 Alk, Wayne, Parag, Anil

          People

          • Assignee:
            andreibaranouski Andrei Baranouski
            Reporter:
            andreibaranouski Andrei Baranouski
          • Votes:
            0 Vote for this issue
            Watchers:
            4 Start watching this issue

            Dates

            • Created:
              Updated:
              Resolved:

              Gerrit Reviews

              There are no open Gerrit changes