You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@hbase.apache.org by "Michael Bieniosek (JIRA)" <ji...@apache.org> on 2008/08/14 19:23:45 UTC
[jira] Created: (HBASE-831) committing BatchUpdate with no row
should complain
committing BatchUpdate with no row should complain
--------------------------------------------------
Key: HBASE-831
URL: https://issues.apache.org/jira/browse/HBASE-831
Project: Hadoop HBase
Issue Type: Bug
Affects Versions: 0.2.0
Reporter: Michael Bieniosek
Priority: Minor
Running this code:
BatchUpdate update = new BatchUpdate();
update.put(key, value);
table.commit(update);
Down in getRegionServer, this triggers an NPE because the row is null (which I saw because I was running in a debugger); this NPE gets retried somewhere in the bowels of IPC. Instead, we should either remove the zero-arg BatchUpdate constructor, or have table.commit throw a runtimeexception if the row is null.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (HBASE-831) committing BatchUpdate with no row
should complain
Posted by "stack (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-831?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
stack updated HBASE-831:
------------------------
Attachment: 831-v2-shortertimeouts.patch
I think that the pause should be shorter still. Rather than 8 seconds, it should be 2 seconds. Attached file changes the pause to 2 seconds and removes the 64 off the end of RETRY_BACKOFF.
At 2 seconds all is 'snappier' -- creating, disablng, etc. -- and after ten retries we're at about 2 minutes which is long-enough in my opinion and is < TaskTracker timeout of ten minutes so we'll see error in MR logs rather than child killed messages.
At 8 seconds, with RETRY_BACKOFF as it was, we're waiting > 17 minutes (if I did my math right).
This patch passes all unit tests. Also tried it in a big MR job. More logging if client DEBUG is enabled but at least now you have a better clue whats going on.
> committing BatchUpdate with no row should complain
> --------------------------------------------------
>
> Key: HBASE-831
> URL: https://issues.apache.org/jira/browse/HBASE-831
> Project: Hadoop HBase
> Issue Type: Bug
> Affects Versions: 0.2.0
> Reporter: Michael Bieniosek
> Assignee: Andrew Purtell
> Priority: Minor
> Fix For: 0.2.1, 0.3.0
>
> Attachments: 831-v2-shortertimeouts.patch, 831.patch
>
>
> Running this code:
> BatchUpdate update = new BatchUpdate();
> update.put(key, value);
> table.commit(update);
> Down in getRegionServer, this triggers an NPE because the row is null (which I saw because I was running in a debugger); this NPE gets retried somewhere in the bowels of IPC. Instead, we should either remove the zero-arg BatchUpdate constructor, or have table.commit throw a runtimeexception if the row is null.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (HBASE-831) committing BatchUpdate with no row
should complain
Posted by "Michael Bieniosek (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-831?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12622622#action_12622622 ]
Michael Bieniosek commented on HBASE-831:
-----------------------------------------
Additionally, the retry logic shouldn't retry NPEs.
> committing BatchUpdate with no row should complain
> --------------------------------------------------
>
> Key: HBASE-831
> URL: https://issues.apache.org/jira/browse/HBASE-831
> Project: Hadoop HBase
> Issue Type: Bug
> Affects Versions: 0.2.0
> Reporter: Michael Bieniosek
> Priority: Minor
>
> Running this code:
> BatchUpdate update = new BatchUpdate();
> update.put(key, value);
> table.commit(update);
> Down in getRegionServer, this triggers an NPE because the row is null (which I saw because I was running in a debugger); this NPE gets retried somewhere in the bowels of IPC. Instead, we should either remove the zero-arg BatchUpdate constructor, or have table.commit throw a runtimeexception if the row is null.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (HBASE-831) committing BatchUpdate with no row
should complain
Posted by "Andrew Purtell (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-831?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Andrew Purtell updated HBASE-831:
---------------------------------
Assignee: Andrew Purtell
Status: Patch Available (was: Open)
> committing BatchUpdate with no row should complain
> --------------------------------------------------
>
> Key: HBASE-831
> URL: https://issues.apache.org/jira/browse/HBASE-831
> Project: Hadoop HBase
> Issue Type: Bug
> Affects Versions: 0.2.0
> Reporter: Michael Bieniosek
> Assignee: Andrew Purtell
> Priority: Minor
> Attachments: 831.patch
>
>
> Running this code:
> BatchUpdate update = new BatchUpdate();
> update.put(key, value);
> table.commit(update);
> Down in getRegionServer, this triggers an NPE because the row is null (which I saw because I was running in a debugger); this NPE gets retried somewhere in the bowels of IPC. Instead, we should either remove the zero-arg BatchUpdate constructor, or have table.commit throw a runtimeexception if the row is null.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (HBASE-831) committing BatchUpdate with no row
should complain
Posted by "stack (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-831?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
stack updated HBASE-831:
------------------------
Comment: was deleted
> committing BatchUpdate with no row should complain
> --------------------------------------------------
>
> Key: HBASE-831
> URL: https://issues.apache.org/jira/browse/HBASE-831
> Project: Hadoop HBase
> Issue Type: Bug
> Affects Versions: 0.2.0
> Reporter: Michael Bieniosek
> Assignee: Andrew Purtell
> Priority: Minor
> Fix For: 0.2.1, 0.3.0
>
> Attachments: 831-v2-shortertimeouts.patch, 831.patch
>
>
> Running this code:
> BatchUpdate update = new BatchUpdate();
> update.put(key, value);
> table.commit(update);
> Down in getRegionServer, this triggers an NPE because the row is null (which I saw because I was running in a debugger); this NPE gets retried somewhere in the bowels of IPC. Instead, we should either remove the zero-arg BatchUpdate constructor, or have table.commit throw a runtimeexception if the row is null.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (HBASE-831) committing BatchUpdate with no row
should complain
Posted by "Andrew Purtell (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-831?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Andrew Purtell updated HBASE-831:
---------------------------------
Attachment: 831.patch
Patch addresses invalid BatchUpdates. Checking happens on both the client and server side.
> committing BatchUpdate with no row should complain
> --------------------------------------------------
>
> Key: HBASE-831
> URL: https://issues.apache.org/jira/browse/HBASE-831
> Project: Hadoop HBase
> Issue Type: Bug
> Affects Versions: 0.2.0
> Reporter: Michael Bieniosek
> Priority: Minor
> Attachments: 831.patch
>
>
> Running this code:
> BatchUpdate update = new BatchUpdate();
> update.put(key, value);
> table.commit(update);
> Down in getRegionServer, this triggers an NPE because the row is null (which I saw because I was running in a debugger); this NPE gets retried somewhere in the bowels of IPC. Instead, we should either remove the zero-arg BatchUpdate constructor, or have table.commit throw a runtimeexception if the row is null.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (HBASE-831) committing BatchUpdate with no row
should complain
Posted by "Michael Bieniosek (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-831?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12622612#action_12622612 ]
Michael Bieniosek commented on HBASE-831:
-----------------------------------------
Also, hbase shouldn't catch and retry NPEs.
> committing BatchUpdate with no row should complain
> --------------------------------------------------
>
> Key: HBASE-831
> URL: https://issues.apache.org/jira/browse/HBASE-831
> Project: Hadoop HBase
> Issue Type: Bug
> Affects Versions: 0.2.0
> Reporter: Michael Bieniosek
> Priority: Minor
>
> Running this code:
> BatchUpdate update = new BatchUpdate();
> update.put(key, value);
> table.commit(update);
> Down in getRegionServer, this triggers an NPE because the row is null (which I saw because I was running in a debugger); this NPE gets retried somewhere in the bowels of IPC. Instead, we should either remove the zero-arg BatchUpdate constructor, or have table.commit throw a runtimeexception if the row is null.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (HBASE-831) committing BatchUpdate with no row
should complain
Posted by "Jim Kellerman (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-831?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Jim Kellerman updated HBASE-831:
--------------------------------
Resolution: Fixed
Fix Version/s: 0.3.0
0.2.1
Status: Resolved (was: Patch Available)
Committed to 0.2 branch and trunk. Thanks for the patch Andrew!
> committing BatchUpdate with no row should complain
> --------------------------------------------------
>
> Key: HBASE-831
> URL: https://issues.apache.org/jira/browse/HBASE-831
> Project: Hadoop HBase
> Issue Type: Bug
> Affects Versions: 0.2.0
> Reporter: Michael Bieniosek
> Assignee: Andrew Purtell
> Priority: Minor
> Fix For: 0.2.1, 0.3.0
>
> Attachments: 831.patch
>
>
> Running this code:
> BatchUpdate update = new BatchUpdate();
> update.put(key, value);
> table.commit(update);
> Down in getRegionServer, this triggers an NPE because the row is null (which I saw because I was running in a debugger); this NPE gets retried somewhere in the bowels of IPC. Instead, we should either remove the zero-arg BatchUpdate constructor, or have table.commit throw a runtimeexception if the row is null.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.