Hi Alexandr Shapkin
With master branch codebase also I'm able to re-produce this issue
(reproducible on 2.7.6, 2.9 and master)
Looks like there is a major bug in the way older WAL segment clean-up is
implemented during checkpoint.
Not sure how connecting to visor is causing this issue, any chance t
Thanks for your reply Alexandr Shapkin
The ticket you shared https://issues.apache.org/jira/browse/IGNITE-13373
more related to preloading
releaseHistoryForPreloading()
But the issue am facing is somehow WAL segments which are not needed for
crash recovery after checkpoint are not getting clea
Hi, Could you share the full logs with the exception? There is an interesting JIRA ticket that could be the reason for this [1] [1] - https://issues.apache.org/jira/browse/IGNITE-13373 From: shivakumarSent: Thursday, December 3, 2020 10:51 AMTo: user@ignite.apache.orgSubject: Could not clear his