[ 
https://issues.apache.org/jira/browse/HDFS-13609?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17278488#comment-17278488
 ] 

xuzq commented on HDFS-13609:
-----------------------------

Thanks [~xkrogen].
{quote}Why JN1 is lagging: You're saying this is happening because JN1 wrote 
some txns to its cache, but not onto disk. Can you elaborate on why this causes 
it to lag?
{quote}
JN1 lagging, because the running JN1 is restarted with wrong 
_dfs.journalnode.edits.dir._

 

On this question, i think there are some bugs :(:
 # Cache is not reflective of what eventually written to disk in Journal.
 # _onlyDurableTxns_ is true in _selectRpcInputStreams_ is not correctness. 
Because maybe here is only have quorum responses, not all journal's response. 
And the first responses may not contain the full edits. It will cause can't 
tail any edits from journal.
 # _onlyDurableTxns_ is true in _editLogTailer.catchupDuringFailover()_, but is 
false in _getFSImage().editLog.openForWrite(getEffectiveLayoutVersion())_ in 
FSNamesystem#startActiveServices(). It maybe caused NameNode crash when 
failover it to active.

 

[~vagarychen] and [~shv],  thanks.

 

> [Edit Tail Fast Path Pt 3] NameNode-side changes to support tailing edits via 
> RPC
> ---------------------------------------------------------------------------------
>
>                 Key: HDFS-13609
>                 URL: https://issues.apache.org/jira/browse/HDFS-13609
>             Project: Hadoop HDFS
>          Issue Type: Sub-task
>          Components: ha, namenode
>            Reporter: Erik Krogen
>            Assignee: Erik Krogen
>            Priority: Major
>             Fix For: HDFS-12943, 3.3.0
>
>         Attachments: HDFS-13609-HDFS-12943.000.patch, 
> HDFS-13609-HDFS-12943.001.patch, HDFS-13609-HDFS-12943.002.patch, 
> HDFS-13609-HDFS-12943.003.patch, HDFS-13609-HDFS-12943.004.patch
>
>
> See HDFS-13150 for the full design.
> This JIRA is targetted at the NameNode-side changes to enable tailing 
> in-progress edits via the RPC mechanism added in HDFS-13608. Most changes are 
> in the QuorumJournalManager.



--
This message was sent by Atlassian Jira
(v8.3.4#803005)

---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to