You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@hbase.apache.org by "stack (JIRA)" <ji...@apache.org> on 2011/01/13 19:37:45 UTC
[jira] Created: (HBASE-3441) Add means of graceful decommission of
node where it sheds load slowly in a non-disruptive manner until its
carrying none
Add means of graceful decommission of node where it sheds load slowly in a non-disruptive manner until its carrying none
------------------------------------------------------------------------------------------------------------------------
Key: HBASE-3441
URL: https://issues.apache.org/jira/browse/HBASE-3441
Project: HBase
Issue Type: Bug
Reporter: stack
Fix For: 0.92.0
Needed decommissioning nodes. Currently we effectively crash them out.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] [Commented] (HBASE-3441) Add means of graceful decommission
of node where it sheds load slowly in a non-disruptive manner until its
carrying none
Posted by "stack (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-3441?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13019700#comment-13019700 ]
stack commented on HBASE-3441:
------------------------------
HBASE-3071 added a script to decommission a server. HBASE-3780 is an issue to enable the jvm's jmx mgmt bean so can invoke full gc from outside jvm so we can do Todd's suggestion above from the script added by hbase-3071.
@Nicolas Want to open new issue to implement your fancy double-writing?
I'll resolve this issue as near dupe of hbase-3071. Please reopen if I shouldn't have done this.
> Add means of graceful decommission of node where it sheds load slowly in a non-disruptive manner until its carrying none
> ------------------------------------------------------------------------------------------------------------------------
>
> Key: HBASE-3441
> URL: https://issues.apache.org/jira/browse/HBASE-3441
> Project: HBase
> Issue Type: Bug
> Reporter: stack
> Priority: Blocker
> Fix For: 0.92.0
>
>
> Needed decommissioning nodes. Currently we effectively crash them out.
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (HBASE-3441) Add means of graceful decommission of
node where it sheds load slowly in a non-disruptive manner until its
carrying none
Posted by "Jean-Daniel Cryans (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-3441?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Jean-Daniel Cryans updated HBASE-3441:
--------------------------------------
Priority: Blocker (was: Critical)
Moving to blocker, we need it.
> Add means of graceful decommission of node where it sheds load slowly in a non-disruptive manner until its carrying none
> ------------------------------------------------------------------------------------------------------------------------
>
> Key: HBASE-3441
> URL: https://issues.apache.org/jira/browse/HBASE-3441
> Project: HBase
> Issue Type: Bug
> Reporter: stack
> Priority: Blocker
> Fix For: 0.92.0
>
>
> Needed decommissioning nodes. Currently we effectively crash them out.
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] Commented: (HBASE-3441) Add means of graceful decommission
of node where it sheds load slowly in a non-disruptive manner until its
carrying none
Posted by "Todd Lipcon (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-3441?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12981419#action_12981419 ]
Todd Lipcon commented on HBASE-3441:
------------------------------------
Yes, this will be super helpful. I especially want a feature where we can put a node into a state where it sheds all of its load gracefully, then without actually restarting the JVM, put it back into normal state. We could use this to fix our memory fragmentation issue - on a rolling basis in a larger cluster, shed regions, System.gc(), then go back into normal operational state.
> Add means of graceful decommission of node where it sheds load slowly in a non-disruptive manner until its carrying none
> ------------------------------------------------------------------------------------------------------------------------
>
> Key: HBASE-3441
> URL: https://issues.apache.org/jira/browse/HBASE-3441
> Project: HBase
> Issue Type: Bug
> Reporter: stack
> Fix For: 0.92.0
>
>
> Needed decommissioning nodes. Currently we effectively crash them out.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (HBASE-3441) Add means of graceful decommission
of node where it sheds load slowly in a non-disruptive manner until its
carrying none
Posted by "Nicolas Spiegelberg (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-3441?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12981429#action_12981429 ]
Nicolas Spiegelberg commented on HBASE-3441:
--------------------------------------------
+1. An accompanying feature you might want to think about is the ability to do a hot standby, where you start the region on another server and double-write to HLogs until you're ready to decommission the old region location.
> Add means of graceful decommission of node where it sheds load slowly in a non-disruptive manner until its carrying none
> ------------------------------------------------------------------------------------------------------------------------
>
> Key: HBASE-3441
> URL: https://issues.apache.org/jira/browse/HBASE-3441
> Project: HBase
> Issue Type: Bug
> Reporter: stack
> Fix For: 0.92.0
>
>
> Needed decommissioning nodes. Currently we effectively crash them out.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] [Commented] (HBASE-3441) Add means of graceful decommission
of node where it sheds load slowly in a non-disruptive manner until its
carrying none
Posted by "Alex Newman (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-3441?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13015639#comment-13015639 ]
Alex Newman commented on HBASE-3441:
------------------------------------
Jean, where are we on nicolas's suggestion? Do they have to move in lockstep? What's the guarantee btwn the two wals?1`
> Add means of graceful decommission of node where it sheds load slowly in a non-disruptive manner until its carrying none
> ------------------------------------------------------------------------------------------------------------------------
>
> Key: HBASE-3441
> URL: https://issues.apache.org/jira/browse/HBASE-3441
> Project: HBase
> Issue Type: Bug
> Reporter: stack
> Priority: Blocker
> Fix For: 0.92.0
>
>
> Needed decommissioning nodes. Currently we effectively crash them out.
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (HBASE-3441) Add means of graceful decommission of
node where it sheds load slowly in a non-disruptive manner until its
carrying none
Posted by "stack (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-3441?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
stack updated HBASE-3441:
-------------------------
Priority: Critical (was: Major)
> Add means of graceful decommission of node where it sheds load slowly in a non-disruptive manner until its carrying none
> ------------------------------------------------------------------------------------------------------------------------
>
> Key: HBASE-3441
> URL: https://issues.apache.org/jira/browse/HBASE-3441
> Project: HBase
> Issue Type: Bug
> Reporter: stack
> Priority: Critical
> Fix For: 0.92.0
>
>
> Needed decommissioning nodes. Currently we effectively crash them out.
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HBASE-3441) Add means of graceful decommission
of node where it sheds load slowly in a non-disruptive manner until its
carrying none
Posted by "Jean-Daniel Cryans (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-3441?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13015645#comment-13015645 ]
Jean-Daniel Cryans commented on HBASE-3441:
-------------------------------------------
(I guess Jean is me so I'm going to answer)
The region moving stuff is orthogonal to decommissioning IMO, and IIRC Benoit either opened an issue about it or at least he described on the dev list a way to make it much less disruptive to move a region.
> Add means of graceful decommission of node where it sheds load slowly in a non-disruptive manner until its carrying none
> ------------------------------------------------------------------------------------------------------------------------
>
> Key: HBASE-3441
> URL: https://issues.apache.org/jira/browse/HBASE-3441
> Project: HBase
> Issue Type: Bug
> Reporter: stack
> Priority: Blocker
> Fix For: 0.92.0
>
>
> Needed decommissioning nodes. Currently we effectively crash them out.
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Resolved] (HBASE-3441) Add means of graceful decommission
of node where it sheds load slowly in a non-disruptive manner until its
carrying none
Posted by "stack (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-3441?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
stack resolved HBASE-3441.
--------------------------
Resolution: Duplicate
Closing as near-dupe of hbase-3071
> Add means of graceful decommission of node where it sheds load slowly in a non-disruptive manner until its carrying none
> ------------------------------------------------------------------------------------------------------------------------
>
> Key: HBASE-3441
> URL: https://issues.apache.org/jira/browse/HBASE-3441
> Project: HBase
> Issue Type: Bug
> Reporter: stack
> Priority: Blocker
> Fix For: 0.92.0
>
>
> Needed decommissioning nodes. Currently we effectively crash them out.
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira