[Hadoop-common-dev] [jira] Created: (HADOOP-3002) HDFS should not remove blocks while in safemode. - Grokbase

Grokbase › Groups › Hadoop › common-dev › March 2008

FAQ

[Hadoop-common-dev] [jira] Created: (HADOOP-3002) HDFS should not remove blocks while in safemode.

	Konstantin Shvachko (JIRA)
	Mar 12, 2008 at 12:52 am

HDFS should not remove blocks while in safemode.
------------------------------------------------

Key: HADOOP-3002
URL: https://issues.apache.org/jira/browse/HADOOP-3002
Project: Hadoop Core
Issue Type: Bug
Components: dfs
Reporter: Konstantin Shvachko
Priority: Critical
Fix For: 0.16.2, 0.17.0

I noticed that data-nodes are removing blocks during a rather prolonged distributed upgrade when the name-node is in safe mode.
This happened on my experimental cluster with accelerated block report rate.
By definition in safe mode the name-node should not
- accept client requests to change the namespace state, and
- schedule block replications and/or block removal for the data-nodes.

We don't want any unnecessary replications until all blocks are reported during startup.
We also don't want to remove blocks if safe mode is entered manually.
In heartbeat processing we explicitly verify that the name-node is in safe-mode and do not return any block commands to the data-nodes.
Block reports can also return block commands, which should be banned during safe mode.

--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.

Search Discussions

20 responses
Oldest
Nested

Sameer Paranjpye (JIRA)

at Mar 21, 2008 at 8:09 pm

⇧

[ https://issues.apache.org/jira/browse/HADOOP-3002?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Sameer Paranjpye updated HADOOP-3002:
-------------------------------------

Fix Version/s: (was: 0.16.2)

HDFS should not remove blocks while in safemode.
------------------------------------------------

Key: HADOOP-3002
URL: https://issues.apache.org/jira/browse/HADOOP-3002
Project: Hadoop Core
Issue Type: Bug
Components: dfs
Reporter: Konstantin Shvachko
Priority: Critical
Fix For: 0.17.0

I noticed that data-nodes are removing blocks during a rather prolonged distributed upgrade when the name-node is in safe mode.
This happened on my experimental cluster with accelerated block report rate.
By definition in safe mode the name-node should not
- accept client requests to change the namespace state, and
- schedule block replications and/or block removal for the data-nodes.
We don't want any unnecessary replications until all blocks are reported during startup.
We also don't want to remove blocks if safe mode is entered manually.
In heartbeat processing we explicitly verify that the name-node is in safe-mode and do not return any block commands to the data-nodes.
Block reports can also return block commands, which should be banned during safe mode.

--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.

reply | permalink

Sameer Paranjpye (JIRA)

at Mar 27, 2008 at 5:56 pm

⇧

[ https://issues.apache.org/jira/browse/HADOOP-3002?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Sameer Paranjpye updated HADOOP-3002:
-------------------------------------

Priority: Major (was: Blocker)

HDFS should not remove blocks while in safemode.
------------------------------------------------

Key: HADOOP-3002
URL: https://issues.apache.org/jira/browse/HADOOP-3002
Project: Hadoop Core
Issue Type: Bug
Components: dfs
Reporter: Konstantin Shvachko
Fix For: 0.17.0

I noticed that data-nodes are removing blocks during a rather prolonged distributed upgrade when the name-node is in safe mode.
This happened on my experimental cluster with accelerated block report rate.
By definition in safe mode the name-node should not
- accept client requests to change the namespace state, and
- schedule block replications and/or block removal for the data-nodes.
We don't want any unnecessary replications until all blocks are reported during startup.
We also don't want to remove blocks if safe mode is entered manually.
In heartbeat processing we explicitly verify that the name-node is in safe-mode and do not return any block commands to the data-nodes.
Block reports can also return block commands, which should be banned during safe mode.

--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.

reply | permalink

Sameer Paranjpye (JIRA)

at Apr 3, 2008 at 6:04 pm

⇧

[ https://issues.apache.org/jira/browse/HADOOP-3002?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Sameer Paranjpye updated HADOOP-3002:
-------------------------------------

Affects Version/s: 0.16.0
Fix Version/s: (was: 0.17.0)

HDFS should not remove blocks while in safemode.
------------------------------------------------

Key: HADOOP-3002
URL: https://issues.apache.org/jira/browse/HADOOP-3002
Project: Hadoop Core
Issue Type: Bug
Components: dfs
Affects Versions: 0.16.0
Reporter: Konstantin Shvachko

I noticed that data-nodes are removing blocks during a rather prolonged distributed upgrade when the name-node is in safe mode.
This happened on my experimental cluster with accelerated block report rate.
By definition in safe mode the name-node should not
- accept client requests to change the namespace state, and
- schedule block replications and/or block removal for the data-nodes.
We don't want any unnecessary replications until all blocks are reported during startup.
We also don't want to remove blocks if safe mode is entered manually.
In heartbeat processing we explicitly verify that the name-node is in safe-mode and do not return any block commands to the data-nodes.
Block reports can also return block commands, which should be banned during safe mode.

--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.

reply | permalink

Lohit Vijayarenu (JIRA)

at Jun 19, 2008 at 6:10 pm

⇧

[ https://issues.apache.org/jira/browse/HADOOP-3002?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Lohit Vijayarenu updated HADOOP-3002:
-------------------------------------

Fix Version/s: 0.19.0

HDFS should not remove blocks while in safemode.
------------------------------------------------

Key: HADOOP-3002
URL: https://issues.apache.org/jira/browse/HADOOP-3002
Project: Hadoop Core
Issue Type: Bug
Components: dfs
Affects Versions: 0.16.0
Reporter: Konstantin Shvachko
Fix For: 0.19.0

I noticed that data-nodes are removing blocks during a rather prolonged distributed upgrade when the name-node is in safe mode.
This happened on my experimental cluster with accelerated block report rate.
By definition in safe mode the name-node should not
- accept client requests to change the namespace state, and
- schedule block replications and/or block removal for the data-nodes.
We don't want any unnecessary replications until all blocks are reported during startup.
We also don't want to remove blocks if safe mode is entered manually.
In heartbeat processing we explicitly verify that the name-node is in safe-mode and do not return any block commands to the data-nodes.
Block reports can also return block commands, which should be banned during safe mode.

--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.

reply | permalink

Robert Chansler (JIRA)

at Jul 3, 2008 at 6:32 pm

⇧

[ https://issues.apache.org/jira/browse/HADOOP-3002?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Robert Chansler updated HADOOP-3002:
------------------------------------

Priority: Blocker (was: Major)

If we fix this, 3677 can be demoted.

HDFS should not remove blocks while in safemode.
------------------------------------------------

Key: HADOOP-3002
URL: https://issues.apache.org/jira/browse/HADOOP-3002
Project: Hadoop Core
Issue Type: Bug
Components: dfs
Affects Versions: 0.16.0
Reporter: Konstantin Shvachko
Priority: Blocker
Fix For: 0.19.0

I noticed that data-nodes are removing blocks during a rather prolonged distributed upgrade when the name-node is in safe mode.
This happened on my experimental cluster with accelerated block report rate.
By definition in safe mode the name-node should not
- accept client requests to change the namespace state, and
- schedule block replications and/or block removal for the data-nodes.
We don't want any unnecessary replications until all blocks are reported during startup.
We also don't want to remove blocks if safe mode is entered manually.
In heartbeat processing we explicitly verify that the name-node is in safe-mode and do not return any block commands to the data-nodes.
Block reports can also return block commands, which should be banned during safe mode.

--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.

reply | permalink

Robert Chansler (JIRA)

at Jul 3, 2008 at 6:32 pm

⇧

[ https://issues.apache.org/jira/browse/HADOOP-3002?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Robert Chansler reassigned HADOOP-3002:
---------------------------------------

Assignee: Konstantin Shvachko

HDFS should not remove blocks while in safemode.
------------------------------------------------

Key: HADOOP-3002
URL: https://issues.apache.org/jira/browse/HADOOP-3002
Project: Hadoop Core
Issue Type: Bug
Components: dfs
Affects Versions: 0.16.0
Reporter: Konstantin Shvachko
Assignee: Konstantin Shvachko
Priority: Blocker
Fix For: 0.19.0

I noticed that data-nodes are removing blocks during a rather prolonged distributed upgrade when the name-node is in safe mode.
This happened on my experimental cluster with accelerated block report rate.
By definition in safe mode the name-node should not
- accept client requests to change the namespace state, and
- schedule block replications and/or block removal for the data-nodes.
We don't want any unnecessary replications until all blocks are reported during startup.
We also don't want to remove blocks if safe mode is entered manually.
In heartbeat processing we explicitly verify that the name-node is in safe-mode and do not return any block commands to the data-nodes.
Block reports can also return block commands, which should be banned during safe mode.

--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.

reply | permalink

Konstantin Shvachko (JIRA)

at Jul 4, 2008 at 2:25 am

⇧

[ https://issues.apache.org/jira/browse/HADOOP-3002?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Konstantin Shvachko updated HADOOP-3002:
----------------------------------------

Fix Version/s: (was: 0.19.0)
0.17.0
0.18.0
Status: Patch Available (was: Open)

HDFS should not remove blocks while in safemode.
------------------------------------------------

Key: HADOOP-3002
URL: https://issues.apache.org/jira/browse/HADOOP-3002
Project: Hadoop Core
Issue Type: Bug
Components: dfs
Affects Versions: 0.16.0
Reporter: Konstantin Shvachko
Assignee: Konstantin Shvachko
Priority: Blocker
Fix For: 0.18.0, 0.17.0

Attachments: DelBlocksInSafeMode.patch

I noticed that data-nodes are removing blocks during a rather prolonged distributed upgrade when the name-node is in safe mode.
This happened on my experimental cluster with accelerated block report rate.
By definition in safe mode the name-node should not
- accept client requests to change the namespace state, and
- schedule block replications and/or block removal for the data-nodes.
We don't want any unnecessary replications until all blocks are reported during startup.
We also don't want to remove blocks if safe mode is entered manually.
In heartbeat processing we explicitly verify that the name-node is in safe-mode and do not return any block commands to the data-nodes.
Block reports can also return block commands, which should be banned during safe mode.

--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.

reply | permalink

Konstantin Shvachko (JIRA)

at Jul 4, 2008 at 2:25 am

⇧

[ https://issues.apache.org/jira/browse/HADOOP-3002?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Konstantin Shvachko updated HADOOP-3002:
----------------------------------------

Attachment: DelBlocksInSafeMode.patch

This is the patch that postpones removal of blocks until the safe mode is off.
The main reason for delition was that block report processing was removing blocks that do not belong
to any file directly ignoring the regular mechanism that first adds invalid blocks into recentInvalidateSets
and then schedules them for deletion via heartbeats.
# I changed block report processing to just placing invalid blocks to recentInvalidateSets
and not returning any commands to data-nodes. This optimized processReport() because now it
does not scan the block report once again looking for invalid blocks.
# I changed heartbeat processing because it never checked the safe mode and would schedule
replications or deletions if there were any in the pending lists.
During startup the pending lists are empty but in manual safe mode it may not be the case.
So now the only commands that are allowed when safe mode is on are requests for block reports
and distributed upgrade commands.
It is not clear why some code in handleHeartbeat() is inside the synchronized section and some is not.
Placed everything inside.

HDFS should not remove blocks while in safemode.
------------------------------------------------

Key: HADOOP-3002
URL: https://issues.apache.org/jira/browse/HADOOP-3002
Project: Hadoop Core
Issue Type: Bug
Components: dfs
Affects Versions: 0.16.0
Reporter: Konstantin Shvachko
Assignee: Konstantin Shvachko
Priority: Blocker
Fix For: 0.17.0, 0.18.0

Attachments: DelBlocksInSafeMode.patch

I noticed that data-nodes are removing blocks during a rather prolonged distributed upgrade when the name-node is in safe mode.
This happened on my experimental cluster with accelerated block report rate.
By definition in safe mode the name-node should not
- accept client requests to change the namespace state, and
- schedule block replications and/or block removal for the data-nodes.
We don't want any unnecessary replications until all blocks are reported during startup.
We also don't want to remove blocks if safe mode is entered manually.
In heartbeat processing we explicitly verify that the name-node is in safe-mode and do not return any block commands to the data-nodes.
Block reports can also return block commands, which should be banned during safe mode.

--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.

reply | permalink

Nigel Daley (JIRA)

at Jul 7, 2008 at 5:46 pm

⇧

[ https://issues.apache.org/jira/browse/HADOOP-3002?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Nigel Daley updated HADOOP-3002:
--------------------------------

Fix Version/s: (was: 0.17.0)
0.17.2

HDFS should not remove blocks while in safemode.
------------------------------------------------

Key: HADOOP-3002
URL: https://issues.apache.org/jira/browse/HADOOP-3002
Project: Hadoop Core
Issue Type: Bug
Components: dfs
Affects Versions: 0.16.0
Reporter: Konstantin Shvachko
Assignee: Konstantin Shvachko
Priority: Blocker
Fix For: 0.17.2, 0.18.0

Attachments: DelBlocksInSafeMode.patch

I noticed that data-nodes are removing blocks during a rather prolonged distributed upgrade when the name-node is in safe mode.
This happened on my experimental cluster with accelerated block report rate.
By definition in safe mode the name-node should not
- accept client requests to change the namespace state, and
- schedule block replications and/or block removal for the data-nodes.
We don't want any unnecessary replications until all blocks are reported during startup.
We also don't want to remove blocks if safe mode is entered manually.
In heartbeat processing we explicitly verify that the name-node is in safe-mode and do not return any block commands to the data-nodes.
Block reports can also return block commands, which should be banned during safe mode.

--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.

reply | permalink

Hadoop QA (JIRA)

at Jul 7, 2008 at 6:24 pm

⇧

[ https://issues.apache.org/jira/browse/HADOOP-3002?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12611264#action_12611264 ]

Hadoop QA commented on HADOOP-3002:
-----------------------------------

-1 overall. Here are the results of testing the latest attachment
http://issues.apache.org/jira/secure/attachment/12385265/DelBlocksInSafeMode.patch
against trunk revision 674442.

+1 @author. The patch does not contain any @author tags.

-1 tests included. The patch doesn't appear to include any new or modified tests.
Please justify why no tests are needed for this patch.

+1 javadoc. The javadoc tool did not generate any warning messages.

+1 javac. The applied patch does not increase the total number of javac compiler warnings.

-1 findbugs. The patch appears to introduce 1 new Findbugs warnings.

+1 release audit. The applied patch does not increase the total number of release audit warnings.

-1 core tests. The patch failed core unit tests.

+1 contrib tests. The patch passed contrib unit tests.

Test results: http://hudson.zones.apache.org/hudson/job/Hadoop-Patch/2799/testReport/
Findbugs warnings: http://hudson.zones.apache.org/hudson/job/Hadoop-Patch/2799/artifact/trunk/build/test/findbugs/newPatchFindbugsWarnings.html
Checkstyle results: http://hudson.zones.apache.org/hudson/job/Hadoop-Patch/2799/artifact/trunk/build/test/checkstyle-errors.html
Console output: http://hudson.zones.apache.org/hudson/job/Hadoop-Patch/2799/console

This message is automatically generated.

HDFS should not remove blocks while in safemode.
------------------------------------------------

Key: HADOOP-3002
URL: https://issues.apache.org/jira/browse/HADOOP-3002
Project: Hadoop Core
Issue Type: Bug
Components: dfs
Affects Versions: 0.16.0
Reporter: Konstantin Shvachko
Assignee: Konstantin Shvachko
Priority: Blocker
Fix For: 0.17.2, 0.18.0

Attachments: DelBlocksInSafeMode.patch

I noticed that data-nodes are removing blocks during a rather prolonged distributed upgrade when the name-node is in safe mode.
This happened on my experimental cluster with accelerated block report rate.
By definition in safe mode the name-node should not
- accept client requests to change the namespace state, and
- schedule block replications and/or block removal for the data-nodes.
We don't want any unnecessary replications until all blocks are reported during startup.
We also don't want to remove blocks if safe mode is entered manually.
In heartbeat processing we explicitly verify that the name-node is in safe-mode and do not return any block commands to the data-nodes.
Block reports can also return block commands, which should be banned during safe mode.

--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.

reply | permalink

Konstantin Shvachko (JIRA)

at Jul 7, 2008 at 10:19 pm

⇧

[ https://issues.apache.org/jira/browse/HADOOP-3002?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Konstantin Shvachko updated HADOOP-3002:
----------------------------------------

Attachment: DelBlocksInSafeMode-018.patch

HDFS should not remove blocks while in safemode.
------------------------------------------------

Key: HADOOP-3002
URL: https://issues.apache.org/jira/browse/HADOOP-3002
Project: Hadoop Core
Issue Type: Bug
Components: dfs
Affects Versions: 0.16.0
Reporter: Konstantin Shvachko
Assignee: Konstantin Shvachko
Priority: Blocker
Fix For: 0.17.2, 0.18.0

Attachments: DelBlocksInSafeMode-018.patch, DelBlocksInSafeMode.patch

I noticed that data-nodes are removing blocks during a rather prolonged distributed upgrade when the name-node is in safe mode.
This happened on my experimental cluster with accelerated block report rate.
By definition in safe mode the name-node should not
- accept client requests to change the namespace state, and
- schedule block replications and/or block removal for the data-nodes.
We don't want any unnecessary replications until all blocks are reported during startup.
We also don't want to remove blocks if safe mode is entered manually.
In heartbeat processing we explicitly verify that the name-node is in safe-mode and do not return any block commands to the data-nodes.
Block reports can also return block commands, which should be banned during safe mode.

--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.

reply | permalink

dhruba borthakur (JIRA)

at Jul 7, 2008 at 10:53 pm

⇧

[ https://issues.apache.org/jira/browse/HADOOP-3002?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12611387#action_12611387 ]

dhruba borthakur commented on HADOOP-3002:
------------------------------------------

Hi Konstantin, I opened HADOOP-3709 to document the lock-hierarchy violation in processing heartbeats. The goal is to not acquire the global FSNamesystem lock to process every heartbeat. Maybe the patch you provide in this patch already fixes HADOOP-3709.

HDFS should not remove blocks while in safemode.
------------------------------------------------

Key: HADOOP-3002
URL: https://issues.apache.org/jira/browse/HADOOP-3002
Project: Hadoop Core
Issue Type: Bug
Components: dfs
Affects Versions: 0.16.0
Reporter: Konstantin Shvachko
Assignee: Konstantin Shvachko
Priority: Blocker
Fix For: 0.17.2, 0.18.0

Attachments: DelBlocksInSafeMode-018.patch, DelBlocksInSafeMode.patch

I noticed that data-nodes are removing blocks during a rather prolonged distributed upgrade when the name-node is in safe mode.
This happened on my experimental cluster with accelerated block report rate.
By definition in safe mode the name-node should not
- accept client requests to change the namespace state, and
- schedule block replications and/or block removal for the data-nodes.
We don't want any unnecessary replications until all blocks are reported during startup.
We also don't want to remove blocks if safe mode is entered manually.
In heartbeat processing we explicitly verify that the name-node is in safe-mode and do not return any block commands to the data-nodes.
Block reports can also return block commands, which should be banned during safe mode.

--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.

reply | permalink

Konstantin Shvachko (JIRA)

at Jul 7, 2008 at 11:52 pm

⇧

[ https://issues.apache.org/jira/browse/HADOOP-3002?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12611405#action_12611405 ]

Konstantin Shvachko commented on HADOOP-3002:
---------------------------------------------

I reverted changes that has been committed. The global lock leads to a potential deadlock.

Thanks Dhruba I overlooked the global lock, which we did not have before. It was introduced in 0.18 by HADOOP-1985.
I'll submit another patch.

HDFS should not remove blocks while in safemode.
------------------------------------------------

Key: HADOOP-3002
URL: https://issues.apache.org/jira/browse/HADOOP-3002
Project: Hadoop Core
Issue Type: Bug
Components: dfs
Affects Versions: 0.16.0
Reporter: Konstantin Shvachko
Assignee: Konstantin Shvachko
Priority: Blocker
Fix For: 0.17.2, 0.18.0

Attachments: DelBlocksInSafeMode-018.patch, DelBlocksInSafeMode.patch

I noticed that data-nodes are removing blocks during a rather prolonged distributed upgrade when the name-node is in safe mode.
This happened on my experimental cluster with accelerated block report rate.
By definition in safe mode the name-node should not
- accept client requests to change the namespace state, and
- schedule block replications and/or block removal for the data-nodes.
We don't want any unnecessary replications until all blocks are reported during startup.
We also don't want to remove blocks if safe mode is entered manually.
In heartbeat processing we explicitly verify that the name-node is in safe-mode and do not return any block commands to the data-nodes.
Block reports can also return block commands, which should be banned during safe mode.

--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.

reply | permalink

Konstantin Shvachko (JIRA)

at Jul 8, 2008 at 12:14 am

⇧

[ https://issues.apache.org/jira/browse/HADOOP-3002?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Konstantin Shvachko updated HADOOP-3002:
----------------------------------------

Status: Open (was: Patch Available)

HDFS should not remove blocks while in safemode.
------------------------------------------------

Key: HADOOP-3002
URL: https://issues.apache.org/jira/browse/HADOOP-3002
Project: Hadoop Core
Issue Type: Bug
Components: dfs
Affects Versions: 0.16.0
Reporter: Konstantin Shvachko
Assignee: Konstantin Shvachko
Priority: Blocker
Fix For: 0.17.2, 0.18.0

Attachments: DelBlocksInSafeMode-018.patch, DelBlocksInSafeMode.patch

I noticed that data-nodes are removing blocks during a rather prolonged distributed upgrade when the name-node is in safe mode.
This happened on my experimental cluster with accelerated block report rate.
By definition in safe mode the name-node should not
- accept client requests to change the namespace state, and
- schedule block replications and/or block removal for the data-nodes.
We don't want any unnecessary replications until all blocks are reported during startup.
We also don't want to remove blocks if safe mode is entered manually.
In heartbeat processing we explicitly verify that the name-node is in safe-mode and do not return any block commands to the data-nodes.
Block reports can also return block commands, which should be banned during safe mode.

--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.

reply | permalink

Konstantin Shvachko (JIRA)

at Jul 8, 2008 at 12:41 am

⇧

[ https://issues.apache.org/jira/browse/HADOOP-3002?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Konstantin Shvachko updated HADOOP-3002:
----------------------------------------

Attachment: DelBlocksInSafeMode.patch

This is a new patch, which does not change heartbeat processing.
The global lock issue will be taken care of by HADOOP-3620.

HDFS should not remove blocks while in safemode.
------------------------------------------------

Key: HADOOP-3002
URL: https://issues.apache.org/jira/browse/HADOOP-3002
Project: Hadoop Core
Issue Type: Bug
Components: dfs
Affects Versions: 0.16.0
Reporter: Konstantin Shvachko
Assignee: Konstantin Shvachko
Priority: Blocker
Fix For: 0.17.2, 0.18.0

Attachments: DelBlocksInSafeMode-018.patch, DelBlocksInSafeMode.patch, DelBlocksInSafeMode.patch

I noticed that data-nodes are removing blocks during a rather prolonged distributed upgrade when the name-node is in safe mode.
This happened on my experimental cluster with accelerated block report rate.
By definition in safe mode the name-node should not
- accept client requests to change the namespace state, and
- schedule block replications and/or block removal for the data-nodes.
We don't want any unnecessary replications until all blocks are reported during startup.
We also don't want to remove blocks if safe mode is entered manually.
In heartbeat processing we explicitly verify that the name-node is in safe-mode and do not return any block commands to the data-nodes.
Block reports can also return block commands, which should be banned during safe mode.

--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.

reply | permalink

Konstantin Shvachko (JIRA)

at Jul 8, 2008 at 3:41 am

⇧

[ https://issues.apache.org/jira/browse/HADOOP-3002?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Konstantin Shvachko updated HADOOP-3002:
----------------------------------------

Status: Patch Available (was: Open)

HDFS should not remove blocks while in safemode.
------------------------------------------------

Key: HADOOP-3002
URL: https://issues.apache.org/jira/browse/HADOOP-3002
Project: Hadoop Core
Issue Type: Bug
Components: dfs
Affects Versions: 0.16.0
Reporter: Konstantin Shvachko
Assignee: Konstantin Shvachko
Priority: Blocker
Fix For: 0.17.2, 0.18.0

Attachments: DelBlocksInSafeMode-018.patch, DelBlocksInSafeMode.patch, DelBlocksInSafeMode.patch

I noticed that data-nodes are removing blocks during a rather prolonged distributed upgrade when the name-node is in safe mode.
This happened on my experimental cluster with accelerated block report rate.
By definition in safe mode the name-node should not
- accept client requests to change the namespace state, and
- schedule block replications and/or block removal for the data-nodes.
We don't want any unnecessary replications until all blocks are reported during startup.
We also don't want to remove blocks if safe mode is entered manually.
In heartbeat processing we explicitly verify that the name-node is in safe-mode and do not return any block commands to the data-nodes.
Block reports can also return block commands, which should be banned during safe mode.

--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.

reply | permalink

Hadoop QA (JIRA)

at Jul 8, 2008 at 10:24 pm

⇧

[ https://issues.apache.org/jira/browse/HADOOP-3002?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12611830#action_12611830 ]

Hadoop QA commented on HADOOP-3002:
-----------------------------------

-1 overall. Here are the results of testing the latest attachment
http://issues.apache.org/jira/secure/attachment/12385454/DelBlocksInSafeMode.patch
against trunk revision 674932.

+1 @author. The patch does not contain any @author tags.

-1 tests included. The patch doesn't appear to include any new or modified tests.
Please justify why no tests are needed for this patch.

+1 javadoc. The javadoc tool did not generate any warning messages.

+1 javac. The applied patch does not increase the total number of javac compiler warnings.

+1 findbugs. The patch does not introduce any new Findbugs warnings.

+1 release audit. The applied patch does not increase the total number of release audit warnings.

+1 core tests. The patch passed core unit tests.

+1 contrib tests. The patch passed contrib unit tests.

Test results: http://hudson.zones.apache.org/hudson/job/Hadoop-Patch/2816/testReport/
Findbugs warnings: http://hudson.zones.apache.org/hudson/job/Hadoop-Patch/2816/artifact/trunk/build/test/findbugs/newPatchFindbugsWarnings.html
Checkstyle results: http://hudson.zones.apache.org/hudson/job/Hadoop-Patch/2816/artifact/trunk/build/test/checkstyle-errors.html
Console output: http://hudson.zones.apache.org/hudson/job/Hadoop-Patch/2816/console

This message is automatically generated.

HDFS should not remove blocks while in safemode.
------------------------------------------------

Key: HADOOP-3002
URL: https://issues.apache.org/jira/browse/HADOOP-3002
Project: Hadoop Core
Issue Type: Bug
Components: dfs
Affects Versions: 0.16.0
Reporter: Konstantin Shvachko
Assignee: Konstantin Shvachko
Priority: Blocker
Fix For: 0.17.2, 0.18.0

Attachments: DelBlocksInSafeMode-018.patch, DelBlocksInSafeMode.patch, DelBlocksInSafeMode.patch

I noticed that data-nodes are removing blocks during a rather prolonged distributed upgrade when the name-node is in safe mode.
This happened on my experimental cluster with accelerated block report rate.
By definition in safe mode the name-node should not
- accept client requests to change the namespace state, and
- schedule block replications and/or block removal for the data-nodes.
We don't want any unnecessary replications until all blocks are reported during startup.
We also don't want to remove blocks if safe mode is entered manually.
In heartbeat processing we explicitly verify that the name-node is in safe-mode and do not return any block commands to the data-nodes.
Block reports can also return block commands, which should be banned during safe mode.

--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.

reply | permalink

Konstantin Shvachko (JIRA)

at Jul 8, 2008 at 11:34 pm

⇧

[ https://issues.apache.org/jira/browse/HADOOP-3002?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Konstantin Shvachko updated HADOOP-3002:
----------------------------------------

Attachment: DelBlocksInSafeMode-017.patch
DelBlocksInSafeMode-018.patch

HDFS should not remove blocks while in safemode.
------------------------------------------------

Key: HADOOP-3002
URL: https://issues.apache.org/jira/browse/HADOOP-3002
Project: Hadoop Core
Issue Type: Bug
Components: dfs
Affects Versions: 0.16.0
Reporter: Konstantin Shvachko
Assignee: Konstantin Shvachko
Priority: Blocker
Fix For: 0.17.2, 0.18.0

Attachments: DelBlocksInSafeMode-017.patch, DelBlocksInSafeMode-018.patch, DelBlocksInSafeMode-018.patch, DelBlocksInSafeMode.patch, DelBlocksInSafeMode.patch

I noticed that data-nodes are removing blocks during a rather prolonged distributed upgrade when the name-node is in safe mode.
This happened on my experimental cluster with accelerated block report rate.
By definition in safe mode the name-node should not
- accept client requests to change the namespace state, and
- schedule block replications and/or block removal for the data-nodes.
We don't want any unnecessary replications until all blocks are reported during startup.
We also don't want to remove blocks if safe mode is entered manually.
In heartbeat processing we explicitly verify that the name-node is in safe-mode and do not return any block commands to the data-nodes.
Block reports can also return block commands, which should be banned during safe mode.

--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.

reply | permalink

Konstantin Shvachko (JIRA)

at Jul 8, 2008 at 11:37 pm

⇧

[ https://issues.apache.org/jira/browse/HADOOP-3002?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Konstantin Shvachko updated HADOOP-3002:
----------------------------------------

Resolution: Fixed
Fix Version/s: (was: 0.18.0)
Status: Resolved (was: Patch Available)

I just committed this.

HDFS should not remove blocks while in safemode.
------------------------------------------------

Key: HADOOP-3002
URL: https://issues.apache.org/jira/browse/HADOOP-3002
Project: Hadoop Core
Issue Type: Bug
Components: dfs
Affects Versions: 0.16.0
Reporter: Konstantin Shvachko
Assignee: Konstantin Shvachko
Priority: Blocker
Fix For: 0.17.2

Attachments: DelBlocksInSafeMode-017.patch, DelBlocksInSafeMode-018.patch, DelBlocksInSafeMode-018.patch, DelBlocksInSafeMode.patch, DelBlocksInSafeMode.patch

I noticed that data-nodes are removing blocks during a rather prolonged distributed upgrade when the name-node is in safe mode.
This happened on my experimental cluster with accelerated block report rate.
By definition in safe mode the name-node should not
- accept client requests to change the namespace state, and
- schedule block replications and/or block removal for the data-nodes.
We don't want any unnecessary replications until all blocks are reported during startup.
We also don't want to remove blocks if safe mode is entered manually.
In heartbeat processing we explicitly verify that the name-node is in safe-mode and do not return any block commands to the data-nodes.
Block reports can also return block commands, which should be banned during safe mode.

--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.

reply | permalink

Hudson (JIRA)

at Aug 22, 2008 at 12:40 pm

⇧

[ https://issues.apache.org/jira/browse/HADOOP-3002?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12624763#action_12624763 ]

Hudson commented on HADOOP-3002:
--------------------------------

Integrated in Hadoop-trunk #581 (See [http://hudson.zones.apache.org/hudson/job/Hadoop-trunk/581/])

HDFS should not remove blocks while in safemode.
------------------------------------------------

Key: HADOOP-3002
URL: https://issues.apache.org/jira/browse/HADOOP-3002
Project: Hadoop Core
Issue Type: Bug
Components: dfs
Affects Versions: 0.16.0
Reporter: Konstantin Shvachko
Assignee: Konstantin Shvachko
Priority: Blocker
Fix For: 0.17.2

Attachments: DelBlocksInSafeMode-017.patch, DelBlocksInSafeMode-018.patch, DelBlocksInSafeMode-018.patch, DelBlocksInSafeMode.patch, DelBlocksInSafeMode.patch

I noticed that data-nodes are removing blocks during a rather prolonged distributed upgrade when the name-node is in safe mode.
This happened on my experimental cluster with accelerated block report rate.
By definition in safe mode the name-node should not
- accept client requests to change the namespace state, and
- schedule block replications and/or block removal for the data-nodes.
We don't want any unnecessary replications until all blocks are reported during startup.
We also don't want to remove blocks if safe mode is entered manually.
In heartbeat processing we explicitly verify that the name-node is in safe-mode and do not return any block commands to the data-nodes.
Block reports can also return block commands, which should be banned during safe mode.

--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.

reply | permalink

Related Discussions

Discussion Navigation

view	thread \| post

Discussion Overview

group	common-dev @
categories	hadoop
posted	Mar 12, '08 at 12:52a
active	Aug 22, '08 at 12:40p
posts	21
users	1
website	hadoop.apache.org...
irc	#hadoop

1 user in discussion

Content

People

Users
Badges

Support

Translate

site design / logo © 2022 Grokbase