Uploaded image for project: 'Couchbase Kubernetes'
  1. Couchbase Kubernetes
  2. K8S-2342

Helm configuration can create an invalid time interval for auto compaction

    XMLWordPrintable

Details

    • Bug
    • Status: Closed
    • Minor
    • Resolution: Cannot Reproduce
    • 2.2.0
    • 2.3.0
    • helm, kubernetes, operator
    • None
    • Kubernetes Autonomous Operator 2.2 on EKS
    • 2

    Description

      Using the values.yaml provided within the operator helm chart without modifying the auto compaction settings, a cluster can be created with the time interval of auto compaction turned on and set to 0 for start time and end time as shown in the image.
      It is unlikely this was set manually by a user by accident since turning this option on requires all 4 fields for the start and end times to be manually entered.
      Upon turning it off, the operator does not seem to turn it back on so it seems like the option isn't managed but it was still somehow turned on.

      Attachments

        Issue Links

          For Gerrit Dashboard: K8S-2342
          # Subject Branch Project Status CR V

          Activity

            Further investigation has shown that there is nothing wrong with the default Helm configuration:

            helm upgrade --install test couchbase/couchbase-operator --version=2.2.102
            

            This leaves us a cluster without the time period configured:

            curl http://localhost:8091/settings/autoCompaction -u Administrator:dWVjV3h0     
            {"autoCompactionSettings":\{"parallelDBAndViewCompaction":false,"databaseFragmentationThreshold":{"percentage":30,"size":"undefined"},"viewFragmentationThreshold":\{"percentage":30,"size":"undefined"},"indexCompactionMode":"circular","indexCircularCompaction":\{"daysOfWeek":"Sunday,Monday,Tuesday,Wednesday,Thursday,Friday,Saturday","interval":{"fromHour":0,"toHour":0,"fromMinute":0,"toMinute":0,"abortOutside":false}},"indexFragmentationThreshold":\{"percentage":30}},"purgeInterval":3}
            

            Interestingly there is no way to tell that the interval is not set but the UI shows it as unticked:

            patrick.stephens Patrick Stephens (Inactive) added a comment - Further investigation has shown that there is nothing wrong with the default Helm configuration: helm upgrade --install test couchbase/couchbase-operator --version=2.2.102 This leaves us a cluster without the time period configured: curl http://localhost:8091/settings/autoCompaction -u Administrator:dWVjV3h0      {"autoCompactionSettings":\{"parallelDBAndViewCompaction":false,"databaseFragmentationThreshold":{"percentage":30,"size":"undefined"},"viewFragmentationThreshold":\{"percentage":30,"size":"undefined"},"indexCompactionMode":"circular","indexCircularCompaction":\{"daysOfWeek":"Sunday,Monday,Tuesday,Wednesday,Thursday,Friday,Saturday","interval":{"fromHour":0,"toHour":0,"fromMinute":0,"toMinute":0,"abortOutside":false}},"indexFragmentationThreshold":\{"percentage":30}},"purgeInterval":3} Interestingly there is no way to tell that the interval is not set but the UI shows it as unticked:

            Sulemaan Yasin please can you confirm what values file you are using so I can test against it?

            patrick.stephens Patrick Stephens (Inactive) added a comment - Sulemaan Yasin please can you confirm what values file you are using so I can test against it?

            I am using a values.yaml with none of the auto compaction settings changed from the defaults available in the chart repo. I have a second cluster using the exact same settings which doesn't have this issue so i don't think it is the final config i am using. It could be related to K8S-2263 where i managed to break the helm config and create phantom backup jobs. I can try following those steps again to see if that has the issue.

            sulemaan.yasin Sulemaan Yasin added a comment - I am using a values.yaml with none of the auto compaction settings changed from the defaults available in the chart repo. I have a second cluster using the exact same settings which doesn't have this issue so i don't think it is the final config i am using. It could be related to K8S-2263 where i managed to break the helm config and create phantom backup jobs. I can try following those steps again to see if that has the issue.

            OK, I'll mark as resolved but feel free to re-open if you can repeat it.

            patrick.stephens Patrick Stephens (Inactive) added a comment - OK, I'll mark as resolved but feel free to re-open if you can repeat it.

            Roo Thorp Please verify and close this issue

            arunkumar Arunkumar Senthilnathan (Inactive) added a comment - Roo Thorp Please verify and close this issue

            People

              roo.thorp Roo Thorp
              sulemaan.yasin Sulemaan Yasin
              Votes:
              0 Vote for this issue
              Watchers:
              3 Start watching this issue

              Dates

                Created:
                Updated:
                Resolved:

                Gerrit Reviews

                  There are no open Gerrit changes

                  PagerDuty