Istvan Fajth created HDDS-2695:
----------------------------------
Summary: SCM is not able to start under certain conditions
Key: HDDS-2695
URL: https://issues.apache.org/jira/browse/HDDS-2695
Project: Hadoop Distributed Data Store
Issue Type: Bug
Components: SCM
Reporter: Istvan Fajth
Given
- a cluster where RATIS-677 happened, and DataNodes are already failing to
start properly due to the issue
When
- I restart the cluster and start to see the exceptions as described in
RATIS-677
- I stop the 3 DN that has the failing pipeline
- remove the ratis metadata for the pipeline
- close the pipeline with scmcli
- restart the 3 DN
Then
- SCM is unable to come out of safe mode, the log shows the following possible
reason:
{code}
2019-12-09 01:13:38,437 INFO
org.apache.hadoop.hdds.scm.safemode.SCMSafeModeManager: SCM in safe mode.
Pipelines with at least one datanode reported count is 0, required at least one
datanode reported per pipeline count is 4
{code}
If after this I restart the SCM, it fails without logging any exception, and
the standard error contains the following message es the last one:
{code}
PipelineID=<id_of_pipeline_that_has_been_closed> not found
{code}
Also scmcli did not list the closed pipeline after I closed it and checked the
active pipelines.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: ozone-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: ozone-issues-h...@hadoop.apache.org
