You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@ozone.apache.org by "Ivan Andika (Jira)" <ji...@apache.org> on 2025/06/30 09:43:00 UTC

[jira] [Updated] (HDDS-13353) SCMNodeManager register should not crash handler thread

     [ https://issues.apache.org/jira/browse/HDDS-13353?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Ivan Andika updated HDDS-13353:
-------------------------------
    Description: 
Our cluster uses org.apache.hadoop.net.ScriptBasedMapping as our net.topology.node.switch.mapping.impl implementation. 

However, we encountered that when the net.topology.script.file.name is pointed to the file that the SCM has no access, the SCM register handler thread simply crash without responding to the Datanode. This causes SCM to be stuck in safe mode indefinitely. Furthermore, the datanode simply reports that there is a SocketTimeoutException, although the network is not the problem.

  was:
Our cluster uses org.apache.hadoop.net.ScriptBasedMapping as our net.topology.node.switch.mapping.impl implementation. 

However, we encountered that when the net.topology.script.file.name is pointed to the file that the SCM has no access, the SCM register handler thread 


> SCMNodeManager register should not crash handler thread
> -------------------------------------------------------
>
>                 Key: HDDS-13353
>                 URL: https://issues.apache.org/jira/browse/HDDS-13353
>             Project: Apache Ozone
>          Issue Type: Sub-task
>            Reporter: Ivan Andika
>            Assignee: Ivan Andika
>            Priority: Major
>
> Our cluster uses org.apache.hadoop.net.ScriptBasedMapping as our net.topology.node.switch.mapping.impl implementation. 
> However, we encountered that when the net.topology.script.file.name is pointed to the file that the SCM has no access, the SCM register handler thread simply crash without responding to the Datanode. This causes SCM to be stuck in safe mode indefinitely. Furthermore, the datanode simply reports that there is a SocketTimeoutException, although the network is not the problem.



--
This message was sent by Atlassian Jira
(v8.20.10#820010)

---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@ozone.apache.org
For additional commands, e-mail: issues-help@ozone.apache.org