Uploaded image for project: 'PUBLIC - Liferay Portal Community Edition'
  1. PUBLIC - Liferay Portal Community Edition
  2. LPS-70709

remote staging with cluster behind load balancer - updating of target group_ record is not atomic

    Details

    • Type: Bug
    • Status: Closed
    • Resolution: Inactive
    • Affects Version/s: 6.2.X EE
    • Fix Version/s: None
    • Component/s: Staging
    • Labels:

      Description

      Opening LPP as per escalation engineer's comment: https://web.liferay.com/group/customer/support/-/support/ticket/IBMNHSESR-37/comment/85950896

      Customer faces an issue where clustering and staging are in picture together. As they have a load balancer in front of their nodes, and publication/staging turn on hits the load balancer, there is a problem when the several request hits different nodes, while on the staging server these calls are within one transaction, on the live cluster it is in separate transactions, what cannot hit the same server, so data inconsistency happens.

      To avoid this, the relevant methods in GroupLocalServiceImpl should be annotated with @Clusterable(onMaster = true) as it is used in BackgroundTaskLocalServiceImpl. Please see that class as reference.

      Please provide a hotfix for this issue.

        Attachments

          Issue Links

            Activity

              People

              • Votes:
                0 Vote for this issue
                Watchers:
                0 Start watching this issue

                Dates

                • Created:
                  Updated:
                  Resolved:
                  Days since last comment:
                  2 years, 23 weeks, 4 days ago

                  Packages

                  Version Package