[
https://issues.apache.org/jira/browse/KNOX-1423?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Sandor Molnar updated KNOX-1423:
--------------------------------
+Apache Knox 3.0.0 release preparation+
We are about to cut the 3.0.0 release branch. master will move on to 3.1.0.
This issue is currently unresolved and is not automatically included in 3.0.0.
Please choose one of the following:
# If this belongs in 3.0.0: once the branch is cut, cherry-pick your commit(s)
onto the new v3.0.0 branch in addition to master. Leave Fix Version = 3.0.0 so
we can track it.
# Do nothing: we will move this issue to the next release (3.1.0) as part of
release cleanup.
If we don't hear back and see no cherry-pick, the issue will be re-targeted to
3.1.0. Thanks!
> Document Zookeeper URL Manager behaviour for HA services
> --------------------------------------------------------
>
> Key: KNOX-1423
> URL: https://issues.apache.org/jira/browse/KNOX-1423
> Project: Apache Knox
> Issue Type: Bug
> Components: Site
> Reporter: Sandeep More
> Assignee: Sandeep More
> Priority: Major
> Fix For: 3.0.0
>
>
> For services that use Zookeeper based HA there is no documentation about how
> the failover scenario works.
> The failover code is a simple round-robin. All URLs are tried once on
> failover. When the list is exhausted the entire list of URLs is re-fetched
> from Zookeeper and the new list is tried again the same way (if the
> configured number of failover attempts has not been reached).
> Upon failure, the url is marked failed and next is tried and used until it
> fails.
>
> This needs to be explicitly documented for clear understanding of HA workings.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
