[
https://issues.apache.org/jira/browse/GEODE-8739?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17249381#comment-17249381
]
Bill Burcham commented on GEODE-8739:
-------------------------------------
[~upthewaterspout] wondered:
Do we have a plan for how we can safely delete the .dat file with geode-for-k8s
to work around the issue without introducing new bugs? Or should we be sticking
with our existing 30 second sleep workaround until we implement some new
network partition detection logic?
[~burcham] answers:
We have no plan for deleting that .dat file. We recognize the problem. For now
the 30-second sleep is the best workaround. A more fulsome solution to network
partition detection is needed.
> Split brain when locators exhaust join attempts on non existant servers
> -----------------------------------------------------------------------
>
> Key: GEODE-8739
> URL: https://issues.apache.org/jira/browse/GEODE-8739
> Project: Geode
> Issue Type: Bug
> Components: membership
> Reporter: Jason Huynh
> Priority: Major
> Attachments: exportedLogs_locator-0.zip, exportedLogs_locator-1.zip
>
>
> The hypothesis: "if there is a locator view .dat file with several
> non-existent servers then then locators will waste all of their join attempts
> on the servers instead of finding each other"
> Scenario is a test/user attempts to recreate a cluster with existing .dat and
> persistent files. The locators are spun in parallel and from the analysis,
> it looks like they are able to communicate with each other, but then end up
> forming their own ds.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
