You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@cassandra.apache.org by "Chris Goffinet (JIRA)" <ji...@apache.org> on 2009/08/15 02:04:14 UTC

[jira] Created: (CASSANDRA-367) CommitLog does not flush on writes

CommitLog does not flush on writes
----------------------------------

                 Key: CASSANDRA-367
                 URL: https://issues.apache.org/jira/browse/CASSANDRA-367
             Project: Cassandra
          Issue Type: Bug
          Components: Core
    Affects Versions: 0.4
            Reporter: Chris Goffinet
            Assignee: Jonathan Ellis
            Priority: Blocker
             Fix For: 0.4


When you write data to CommitLog, we are not calling a flush() in class LogRecordAdder.

Is this acceptable? We added flush() and its working now. This bug was introduced when things were consolidated in r799942

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


[jira] Updated: (CASSANDRA-367) CommitLog does not flush on writes

Posted by "Jonathan Ellis (JIRA)" <ji...@apache.org>.
     [ https://issues.apache.org/jira/browse/CASSANDRA-367?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Jonathan Ellis updated CASSANDRA-367:
-------------------------------------

    Attachment: 367-v2.patch

v2 w/ config option actually used :)

> CommitLog does not flush on writes
> ----------------------------------
>
>                 Key: CASSANDRA-367
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-367
>             Project: Cassandra
>          Issue Type: Bug
>          Components: Core
>    Affects Versions: 0.4
>            Reporter: Chris Goffinet
>            Assignee: Jonathan Ellis
>            Priority: Blocker
>             Fix For: 0.4
>
>         Attachments: 0001-CASSANDRA-367-Added-flush.patch, 367-v2.patch, 367.patch
>
>
> When you write data to CommitLog, we are not calling a flush() in class LogRecordAdder.
> Is this acceptable? We added flush() and its working now. This bug was introduced when things were consolidated in r799942

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


[jira] Commented: (CASSANDRA-367) CommitLog does not flush on writes

Posted by "Chris Goffinet (JIRA)" <ji...@apache.org>.
    [ https://issues.apache.org/jira/browse/CASSANDRA-367?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12744172#action_12744172 ] 

Chris Goffinet commented on CASSANDRA-367:
------------------------------------------

Thread.sleep(600) should be pulling in getCommitLogSyncPeriod()

> CommitLog does not flush on writes
> ----------------------------------
>
>                 Key: CASSANDRA-367
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-367
>             Project: Cassandra
>          Issue Type: Bug
>          Components: Core
>    Affects Versions: 0.4
>            Reporter: Chris Goffinet
>            Assignee: Jonathan Ellis
>            Priority: Blocker
>             Fix For: 0.4
>
>         Attachments: 0001-CASSANDRA-367-Added-flush.patch, 367-v2.patch, 367.patch
>
>
> When you write data to CommitLog, we are not calling a flush() in class LogRecordAdder.
> Is this acceptable? We added flush() and its working now. This bug was introduced when things were consolidated in r799942

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


[jira] Updated: (CASSANDRA-367) CommitLog does not flush on writes

Posted by "Jonathan Ellis (JIRA)" <ji...@apache.org>.
     [ https://issues.apache.org/jira/browse/CASSANDRA-367?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Jonathan Ellis updated CASSANDRA-367:
-------------------------------------

    Attachment: 367-3.patch

forgot while loop :)

> CommitLog does not flush on writes
> ----------------------------------
>
>                 Key: CASSANDRA-367
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-367
>             Project: Cassandra
>          Issue Type: Bug
>          Components: Core
>    Affects Versions: 0.4
>            Reporter: Chris Goffinet
>            Assignee: Jonathan Ellis
>            Priority: Blocker
>             Fix For: 0.4
>
>         Attachments: 0001-CASSANDRA-367-Added-flush.patch, 367-3.patch, 367-v2.patch, 367.patch
>
>
> When you write data to CommitLog, we are not calling a flush() in class LogRecordAdder.
> Is this acceptable? We added flush() and its working now. This bug was introduced when things were consolidated in r799942

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


[jira] Updated: (CASSANDRA-367) CommitLog does not flush on writes

Posted by "Jonathan Ellis (JIRA)" <ji...@apache.org>.
     [ https://issues.apache.org/jira/browse/CASSANDRA-367?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Jonathan Ellis updated CASSANDRA-367:
-------------------------------------

    Attachment: 367.patch

patch to always sync commitlog, either as a batch pre-ack (current CLSync of true) or a new periodic mode defaulting to ever 1000ms

> CommitLog does not flush on writes
> ----------------------------------
>
>                 Key: CASSANDRA-367
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-367
>             Project: Cassandra
>          Issue Type: Bug
>          Components: Core
>    Affects Versions: 0.4
>            Reporter: Chris Goffinet
>            Assignee: Jonathan Ellis
>            Priority: Blocker
>             Fix For: 0.4
>
>         Attachments: 0001-CASSANDRA-367-Added-flush.patch, 367.patch
>
>
> When you write data to CommitLog, we are not calling a flush() in class LogRecordAdder.
> Is this acceptable? We added flush() and its working now. This bug was introduced when things were consolidated in r799942

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


[jira] Commented: (CASSANDRA-367) CommitLog does not flush on writes

Posted by "Jonathan Ellis (JIRA)" <ji...@apache.org>.
    [ https://issues.apache.org/jira/browse/CASSANDRA-367?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12743518#action_12743518 ] 

Jonathan Ellis commented on CASSANDRA-367:
------------------------------------------

I'd rather not call flush every add if possible.  What problems are you seeing?

(To be pedantic: the change in the revision in question moved from using an unbuffered writer, to using a buffered one.  Not removing a pre-existing flush.)

> CommitLog does not flush on writes
> ----------------------------------
>
>                 Key: CASSANDRA-367
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-367
>             Project: Cassandra
>          Issue Type: Bug
>          Components: Core
>    Affects Versions: 0.4
>            Reporter: Chris Goffinet
>            Assignee: Jonathan Ellis
>            Priority: Blocker
>             Fix For: 0.4
>
>         Attachments: 0001-CASSANDRA-367-Added-flush.patch
>
>
> When you write data to CommitLog, we are not calling a flush() in class LogRecordAdder.
> Is this acceptable? We added flush() and its working now. This bug was introduced when things were consolidated in r799942

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


[jira] Commented: (CASSANDRA-367) CommitLog does not flush on writes

Posted by "Chris Goffinet (JIRA)" <ji...@apache.org>.
    [ https://issues.apache.org/jira/browse/CASSANDRA-367?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12744767#action_12744767 ] 

Chris Goffinet commented on CASSANDRA-367:
------------------------------------------

Oops! Yes, this looks better now. +1

> CommitLog does not flush on writes
> ----------------------------------
>
>                 Key: CASSANDRA-367
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-367
>             Project: Cassandra
>          Issue Type: Bug
>          Components: Core
>    Affects Versions: 0.4
>            Reporter: Chris Goffinet
>            Assignee: Jonathan Ellis
>            Priority: Blocker
>             Fix For: 0.4
>
>         Attachments: 0001-CASSANDRA-367-Added-flush.patch, 367-3.patch, 367-v2.patch, 367.patch
>
>
> When you write data to CommitLog, we are not calling a flush() in class LogRecordAdder.
> Is this acceptable? We added flush() and its working now. This bug was introduced when things were consolidated in r799942

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


[jira] Updated: (CASSANDRA-367) CommitLog does not flush on writes

Posted by "Chris Goffinet (JIRA)" <ji...@apache.org>.
     [ https://issues.apache.org/jira/browse/CASSANDRA-367?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Chris Goffinet updated CASSANDRA-367:
-------------------------------------

    Attachment:     (was: 0001-CASSANDRA-367-Added-flush.patch)

> CommitLog does not flush on writes
> ----------------------------------
>
>                 Key: CASSANDRA-367
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-367
>             Project: Cassandra
>          Issue Type: Bug
>          Components: Core
>    Affects Versions: 0.4
>            Reporter: Chris Goffinet
>            Assignee: Jonathan Ellis
>            Priority: Blocker
>             Fix For: 0.4
>
>         Attachments: 0001-CASSANDRA-367-Added-flush.patch
>
>
> When you write data to CommitLog, we are not calling a flush() in class LogRecordAdder.
> Is this acceptable? We added flush() and its working now. This bug was introduced when things were consolidated in r799942

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


[jira] Resolved: (CASSANDRA-367) CommitLog does not flush on writes

Posted by "Jonathan Ellis (JIRA)" <ji...@apache.org>.
     [ https://issues.apache.org/jira/browse/CASSANDRA-367?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Jonathan Ellis resolved CASSANDRA-367.
--------------------------------------

    Resolution: Fixed

committed

> CommitLog does not flush on writes
> ----------------------------------
>
>                 Key: CASSANDRA-367
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-367
>             Project: Cassandra
>          Issue Type: Bug
>          Components: Core
>    Affects Versions: 0.4
>            Reporter: Chris Goffinet
>            Assignee: Jonathan Ellis
>            Priority: Blocker
>             Fix For: 0.4
>
>         Attachments: 0001-CASSANDRA-367-Added-flush.patch, 367-3.patch, 367-v2.patch, 367.patch
>
>
> When you write data to CommitLog, we are not calling a flush() in class LogRecordAdder.
> Is this acceptable? We added flush() and its working now. This bug was introduced when things were consolidated in r799942

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


[jira] Commented: (CASSANDRA-367) CommitLog does not flush on writes

Posted by "Hudson (JIRA)" <ji...@apache.org>.
    [ https://issues.apache.org/jira/browse/CASSANDRA-367?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12745065#action_12745065 ] 

Hudson commented on CASSANDRA-367:
----------------------------------

Integrated in Cassandra #172 (See [http://hudson.zones.apache.org/hudson/job/Cassandra/172/])
    add missing while loop on periodic commitlog sync thread.
patch by jbellis; reviewed by Chris Goffinet for 


> CommitLog does not flush on writes
> ----------------------------------
>
>                 Key: CASSANDRA-367
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-367
>             Project: Cassandra
>          Issue Type: Bug
>          Components: Core
>    Affects Versions: 0.4
>            Reporter: Chris Goffinet
>            Assignee: Jonathan Ellis
>            Priority: Blocker
>             Fix For: 0.4
>
>         Attachments: 0001-CASSANDRA-367-Added-flush.patch, 367-3.patch, 367-v2.patch, 367.patch
>
>
> When you write data to CommitLog, we are not calling a flush() in class LogRecordAdder.
> Is this acceptable? We added flush() and its working now. This bug was introduced when things were consolidated in r799942

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


[jira] Commented: (CASSANDRA-367) CommitLog does not flush on writes

Posted by "Chris Goffinet (JIRA)" <ji...@apache.org>.
    [ https://issues.apache.org/jira/browse/CASSANDRA-367?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12744283#action_12744283 ] 

Chris Goffinet commented on CASSANDRA-367:
------------------------------------------

+1

> CommitLog does not flush on writes
> ----------------------------------
>
>                 Key: CASSANDRA-367
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-367
>             Project: Cassandra
>          Issue Type: Bug
>          Components: Core
>    Affects Versions: 0.4
>            Reporter: Chris Goffinet
>            Assignee: Jonathan Ellis
>            Priority: Blocker
>             Fix For: 0.4
>
>         Attachments: 0001-CASSANDRA-367-Added-flush.patch, 367-v2.patch, 367.patch
>
>
> When you write data to CommitLog, we are not calling a flush() in class LogRecordAdder.
> Is this acceptable? We added flush() and its working now. This bug was introduced when things were consolidated in r799942

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


[jira] Updated: (CASSANDRA-367) CommitLog does not flush on writes

Posted by "Chris Goffinet (JIRA)" <ji...@apache.org>.
     [ https://issues.apache.org/jira/browse/CASSANDRA-367?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Chris Goffinet updated CASSANDRA-367:
-------------------------------------

    Attachment: 0001-CASSANDRA-367-Added-flush.patch

> CommitLog does not flush on writes
> ----------------------------------
>
>                 Key: CASSANDRA-367
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-367
>             Project: Cassandra
>          Issue Type: Bug
>          Components: Core
>    Affects Versions: 0.4
>            Reporter: Chris Goffinet
>            Assignee: Jonathan Ellis
>            Priority: Blocker
>             Fix For: 0.4
>
>         Attachments: 0001-CASSANDRA-367-Added-flush.patch
>
>
> When you write data to CommitLog, we are not calling a flush() in class LogRecordAdder.
> Is this acceptable? We added flush() and its working now. This bug was introduced when things were consolidated in r799942

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


[jira] Reopened: (CASSANDRA-367) CommitLog does not flush on writes

Posted by "Jonathan Ellis (JIRA)" <ji...@apache.org>.
     [ https://issues.apache.org/jira/browse/CASSANDRA-367?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Jonathan Ellis reopened CASSANDRA-367:
--------------------------------------


> CommitLog does not flush on writes
> ----------------------------------
>
>                 Key: CASSANDRA-367
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-367
>             Project: Cassandra
>          Issue Type: Bug
>          Components: Core
>    Affects Versions: 0.4
>            Reporter: Chris Goffinet
>            Assignee: Jonathan Ellis
>            Priority: Blocker
>             Fix For: 0.4
>
>         Attachments: 0001-CASSANDRA-367-Added-flush.patch, 367-v2.patch, 367.patch
>
>
> When you write data to CommitLog, we are not calling a flush() in class LogRecordAdder.
> Is this acceptable? We added flush() and its working now. This bug was introduced when things were consolidated in r799942

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