You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@zookeeper.apache.org by "Flavio Paiva Junqueira (JIRA)" <ji...@apache.org> on 2008/11/20 15:38:44 UTC
[jira] Created: (ZOOKEEPER-230) Improvements to FLE
Improvements to FLE
-------------------
Key: ZOOKEEPER-230
URL: https://issues.apache.org/jira/browse/ZOOKEEPER-230
Project: Zookeeper
Issue Type: Improvement
Components: leaderElection
Affects Versions: 3.1.0
Reporter: Flavio Paiva Junqueira
Assignee: Flavio Paiva Junqueira
Fix For: 3.0.0
I'm about to attach a patch that implements the following modifications:
. Currently, if a server is on leader election and doesn't receive a notification for some amount of time t, then it sends a new set of notifications if at least one server has delivered a message from the previous set. With this patch, the amount of time a server waits for a notification before sending a new set increases exponentially;
. I have separated connecting to servers and queuing new notification messages. Before they were all in the same message. The advantage is that now I can tell to an instance of QuorumCnxManager to try to connect to other servers without generating new notification messages;
. I have changed the logging level of several messages on QuorumCnxManager. They were "warn", but they should really be either "info" or "debug". I've changed them to info.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (ZOOKEEPER-230) Improvements to FLE
Posted by "Mahadev konar (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/ZOOKEEPER-230?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Mahadev konar updated ZOOKEEPER-230:
------------------------------------
Attachment: ZOOKEEPER-230.patch
attaching patch from the top level of the source tree.
> Improvements to FLE
> -------------------
>
> Key: ZOOKEEPER-230
> URL: https://issues.apache.org/jira/browse/ZOOKEEPER-230
> Project: Zookeeper
> Issue Type: Improvement
> Components: leaderElection
> Affects Versions: 3.0.0
> Reporter: Flavio Paiva Junqueira
> Assignee: Flavio Paiva Junqueira
> Fix For: 3.1.0
>
> Attachments: ZOOKEEPER-230.patch, ZOOKEEPER-230.patch, ZOOKEEPER-230.patch
>
>
> I'm about to attach a patch that implements the following modifications:
> . Currently, if a server is on leader election and doesn't receive a notification for some amount of time t, then it sends a new set of notifications if at least one server has delivered a message from the previous set. With this patch, the amount of time a server waits for a notification before sending a new set increases exponentially;
> . I have separated connecting to servers and queuing new notification messages. Before they were all in the same message. The advantage is that now I can tell to an instance of QuorumCnxManager to try to connect to other servers without generating new notification messages;
> . I have changed the logging level of several messages on QuorumCnxManager. They were "warn", but they should really be either "info" or "debug". I've changed them to info.
>
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (ZOOKEEPER-230) Improvements to FLE
Posted by "Mahadev konar (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/ZOOKEEPER-230?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Mahadev konar updated ZOOKEEPER-230:
------------------------------------
Attachment: ZOOKEEPER-230.patch
ok fixed the comments i had, and also removed some unused variables in the code.
> Improvements to FLE
> -------------------
>
> Key: ZOOKEEPER-230
> URL: https://issues.apache.org/jira/browse/ZOOKEEPER-230
> Project: Zookeeper
> Issue Type: Improvement
> Components: leaderElection
> Affects Versions: 3.0.0
> Reporter: Flavio Paiva Junqueira
> Assignee: Flavio Paiva Junqueira
> Fix For: 3.1.0
>
> Attachments: ZOOKEEPER-230.patch, ZOOKEEPER-230.patch, ZOOKEEPER-230.patch, ZOOKEEPER-230.patch, ZOOKEEPER-230.patch
>
>
> I'm about to attach a patch that implements the following modifications:
> . Currently, if a server is on leader election and doesn't receive a notification for some amount of time t, then it sends a new set of notifications if at least one server has delivered a message from the previous set. With this patch, the amount of time a server waits for a notification before sending a new set increases exponentially;
> . I have separated connecting to servers and queuing new notification messages. Before they were all in the same message. The advantage is that now I can tell to an instance of QuorumCnxManager to try to connect to other servers without generating new notification messages;
> . I have changed the logging level of several messages on QuorumCnxManager. They were "warn", but they should really be either "info" or "debug". I've changed them to info.
>
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (ZOOKEEPER-230) Improvements to FLE
Posted by "Mahadev konar (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/ZOOKEEPER-230?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12655312#action_12655312 ]
Mahadev konar commented on ZOOKEEPER-230:
-----------------------------------------
the patch looks good except for a few nits:
- this is just a documentation bug
{noformat}
+ * notification checks. This impacts the amount of time to get
+ * the system up again after long partitions. Currently 6 seconds.
*/
{noformat}
this should be 60 seconds.
- the other one is
{noformat}
-
+ LOG.info("About to leave instance:" + proposedLeader + ", " + proposedZxid + ", " + self.getId() + ", " + self.getPeerState());
{noformat}
we should wrap these lines around to confirm to 80 character columns. We dont have to be pedantic about the column length but 80-90 column is good enough i think. Their are a few more lines I think that are overly long and should be wrapped arnd.
> Improvements to FLE
> -------------------
>
> Key: ZOOKEEPER-230
> URL: https://issues.apache.org/jira/browse/ZOOKEEPER-230
> Project: Zookeeper
> Issue Type: Improvement
> Components: leaderElection
> Affects Versions: 3.0.0
> Reporter: Flavio Paiva Junqueira
> Assignee: Flavio Paiva Junqueira
> Fix For: 3.1.0
>
> Attachments: ZOOKEEPER-230.patch, ZOOKEEPER-230.patch, ZOOKEEPER-230.patch, ZOOKEEPER-230.patch
>
>
> I'm about to attach a patch that implements the following modifications:
> . Currently, if a server is on leader election and doesn't receive a notification for some amount of time t, then it sends a new set of notifications if at least one server has delivered a message from the previous set. With this patch, the amount of time a server waits for a notification before sending a new set increases exponentially;
> . I have separated connecting to servers and queuing new notification messages. Before they were all in the same message. The advantage is that now I can tell to an instance of QuorumCnxManager to try to connect to other servers without generating new notification messages;
> . I have changed the logging level of several messages on QuorumCnxManager. They were "warn", but they should really be either "info" or "debug". I've changed them to info.
>
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (ZOOKEEPER-230) Improvements to FLE
Posted by "Flavio Paiva Junqueira (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/ZOOKEEPER-230?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Flavio Paiva Junqueira updated ZOOKEEPER-230:
---------------------------------------------
Status: Patch Available (was: Open)
Forgot to start the workflow...
> Improvements to FLE
> -------------------
>
> Key: ZOOKEEPER-230
> URL: https://issues.apache.org/jira/browse/ZOOKEEPER-230
> Project: Zookeeper
> Issue Type: Improvement
> Components: leaderElection
> Affects Versions: 3.0.0
> Reporter: Flavio Paiva Junqueira
> Assignee: Flavio Paiva Junqueira
> Fix For: 3.1.0
>
> Attachments: ZOOKEEPER-230.patch, ZOOKEEPER-230.patch, ZOOKEEPER-230.patch, ZOOKEEPER-230.patch
>
>
> I'm about to attach a patch that implements the following modifications:
> . Currently, if a server is on leader election and doesn't receive a notification for some amount of time t, then it sends a new set of notifications if at least one server has delivered a message from the previous set. With this patch, the amount of time a server waits for a notification before sending a new set increases exponentially;
> . I have separated connecting to servers and queuing new notification messages. Before they were all in the same message. The advantage is that now I can tell to an instance of QuorumCnxManager to try to connect to other servers without generating new notification messages;
> . I have changed the logging level of several messages on QuorumCnxManager. They were "warn", but they should really be either "info" or "debug". I've changed them to info.
>
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (ZOOKEEPER-230) Improvements to FLE
Posted by "Mahadev konar (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/ZOOKEEPER-230?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12650663#action_12650663 ]
Mahadev konar commented on ZOOKEEPER-230:
-----------------------------------------
flavio,
- Currently, if a server is on leader election and doesn't receive a notification for some amount of time t, then it sends a new set of notifications if at least one server has delivered a message from the previous set. With this patch, the amount of time a server waits for a notification before sending a new set increases exponentially;
i dont unserstand the reason behind this. do we generate too many messages for the leader election notification? I think the traffic generated should be small. Also this does not seem to have a limit on exponential backoff. It will increase to insanely large amounts in case of network partition right?
> Improvements to FLE
> -------------------
>
> Key: ZOOKEEPER-230
> URL: https://issues.apache.org/jira/browse/ZOOKEEPER-230
> Project: Zookeeper
> Issue Type: Improvement
> Components: leaderElection
> Affects Versions: 3.0.0
> Reporter: Flavio Paiva Junqueira
> Assignee: Flavio Paiva Junqueira
> Fix For: 3.1.0
>
> Attachments: ZOOKEEPER-230.patch
>
>
> I'm about to attach a patch that implements the following modifications:
> . Currently, if a server is on leader election and doesn't receive a notification for some amount of time t, then it sends a new set of notifications if at least one server has delivered a message from the previous set. With this patch, the amount of time a server waits for a notification before sending a new set increases exponentially;
> . I have separated connecting to servers and queuing new notification messages. Before they were all in the same message. The advantage is that now I can tell to an instance of QuorumCnxManager to try to connect to other servers without generating new notification messages;
> . I have changed the logging level of several messages on QuorumCnxManager. They were "warn", but they should really be either "info" or "debug". I've changed them to info.
>
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (ZOOKEEPER-230) Improvements to FLE
Posted by "Flavio Paiva Junqueira (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/ZOOKEEPER-230?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Flavio Paiva Junqueira updated ZOOKEEPER-230:
---------------------------------------------
Attachment: ZOOKEEPER-230.patch
Upper limits the amount of time between two consecutive checks. With this patch, the maximum interval between two consecutive checks for new notifications is 1 minute.
> Improvements to FLE
> -------------------
>
> Key: ZOOKEEPER-230
> URL: https://issues.apache.org/jira/browse/ZOOKEEPER-230
> Project: Zookeeper
> Issue Type: Improvement
> Components: leaderElection
> Affects Versions: 3.0.0
> Reporter: Flavio Paiva Junqueira
> Assignee: Flavio Paiva Junqueira
> Fix For: 3.1.0
>
> Attachments: ZOOKEEPER-230.patch, ZOOKEEPER-230.patch
>
>
> I'm about to attach a patch that implements the following modifications:
> . Currently, if a server is on leader election and doesn't receive a notification for some amount of time t, then it sends a new set of notifications if at least one server has delivered a message from the previous set. With this patch, the amount of time a server waits for a notification before sending a new set increases exponentially;
> . I have separated connecting to servers and queuing new notification messages. Before they were all in the same message. The advantage is that now I can tell to an instance of QuorumCnxManager to try to connect to other servers without generating new notification messages;
> . I have changed the logging level of several messages on QuorumCnxManager. They were "warn", but they should really be either "info" or "debug". I've changed them to info.
>
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (ZOOKEEPER-230) Improvements to FLE
Posted by "Flavio Paiva Junqueira (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/ZOOKEEPER-230?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Flavio Paiva Junqueira updated ZOOKEEPER-230:
---------------------------------------------
Status: Patch Available (was: Open)
> Improvements to FLE
> -------------------
>
> Key: ZOOKEEPER-230
> URL: https://issues.apache.org/jira/browse/ZOOKEEPER-230
> Project: Zookeeper
> Issue Type: Improvement
> Components: leaderElection
> Affects Versions: 3.1.0
> Reporter: Flavio Paiva Junqueira
> Assignee: Flavio Paiva Junqueira
> Fix For: 3.0.0
>
> Attachments: ZOOKEEPER-230.patch
>
>
> I'm about to attach a patch that implements the following modifications:
> . Currently, if a server is on leader election and doesn't receive a notification for some amount of time t, then it sends a new set of notifications if at least one server has delivered a message from the previous set. With this patch, the amount of time a server waits for a notification before sending a new set increases exponentially;
> . I have separated connecting to servers and queuing new notification messages. Before they were all in the same message. The advantage is that now I can tell to an instance of QuorumCnxManager to try to connect to other servers without generating new notification messages;
> . I have changed the logging level of several messages on QuorumCnxManager. They were "warn", but they should really be either "info" or "debug". I've changed them to info.
>
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (ZOOKEEPER-230) Improvements to FLE
Posted by "Hudson (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/ZOOKEEPER-230?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12656952#action_12656952 ]
Hudson commented on ZOOKEEPER-230:
----------------------------------
Integrated in ZooKeeper-trunk #174 (See [http://hudson.zones.apache.org/hudson/job/ZooKeeper-trunk/174/])
> Improvements to FLE
> -------------------
>
> Key: ZOOKEEPER-230
> URL: https://issues.apache.org/jira/browse/ZOOKEEPER-230
> Project: Zookeeper
> Issue Type: Improvement
> Components: leaderElection
> Affects Versions: 3.0.0
> Reporter: Flavio Paiva Junqueira
> Assignee: Flavio Paiva Junqueira
> Fix For: 3.1.0
>
> Attachments: ZOOKEEPER-230.patch, ZOOKEEPER-230.patch, ZOOKEEPER-230.patch, ZOOKEEPER-230.patch, ZOOKEEPER-230.patch
>
>
> I'm about to attach a patch that implements the following modifications:
> . Currently, if a server is on leader election and doesn't receive a notification for some amount of time t, then it sends a new set of notifications if at least one server has delivered a message from the previous set. With this patch, the amount of time a server waits for a notification before sending a new set increases exponentially;
> . I have separated connecting to servers and queuing new notification messages. Before they were all in the same message. The advantage is that now I can tell to an instance of QuorumCnxManager to try to connect to other servers without generating new notification messages;
> . I have changed the logging level of several messages on QuorumCnxManager. They were "warn", but they should really be either "info" or "debug". I've changed them to info.
>
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (ZOOKEEPER-230) Improvements to FLE
Posted by "Mahadev konar (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/ZOOKEEPER-230?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Mahadev konar updated ZOOKEEPER-230:
------------------------------------
Resolution: Fixed
Hadoop Flags: [Reviewed]
Status: Resolved (was: Patch Available)
I just committed this. thanks flavio.
> Improvements to FLE
> -------------------
>
> Key: ZOOKEEPER-230
> URL: https://issues.apache.org/jira/browse/ZOOKEEPER-230
> Project: Zookeeper
> Issue Type: Improvement
> Components: leaderElection
> Affects Versions: 3.0.0
> Reporter: Flavio Paiva Junqueira
> Assignee: Flavio Paiva Junqueira
> Fix For: 3.1.0
>
> Attachments: ZOOKEEPER-230.patch, ZOOKEEPER-230.patch, ZOOKEEPER-230.patch, ZOOKEEPER-230.patch, ZOOKEEPER-230.patch
>
>
> I'm about to attach a patch that implements the following modifications:
> . Currently, if a server is on leader election and doesn't receive a notification for some amount of time t, then it sends a new set of notifications if at least one server has delivered a message from the previous set. With this patch, the amount of time a server waits for a notification before sending a new set increases exponentially;
> . I have separated connecting to servers and queuing new notification messages. Before they were all in the same message. The advantage is that now I can tell to an instance of QuorumCnxManager to try to connect to other servers without generating new notification messages;
> . I have changed the logging level of several messages on QuorumCnxManager. They were "warn", but they should really be either "info" or "debug". I've changed them to info.
>
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (ZOOKEEPER-230) Improvements to FLE
Posted by "Mahadev konar (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/ZOOKEEPER-230?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Mahadev konar updated ZOOKEEPER-230:
------------------------------------
Affects Version/s: (was: 3.1.0)
3.0.0
Fix Version/s: (was: 3.0.0)
3.1.0
i think you interchanged the fix and affects versions :) .
> Improvements to FLE
> -------------------
>
> Key: ZOOKEEPER-230
> URL: https://issues.apache.org/jira/browse/ZOOKEEPER-230
> Project: Zookeeper
> Issue Type: Improvement
> Components: leaderElection
> Affects Versions: 3.0.0
> Reporter: Flavio Paiva Junqueira
> Assignee: Flavio Paiva Junqueira
> Fix For: 3.1.0
>
> Attachments: ZOOKEEPER-230.patch
>
>
> I'm about to attach a patch that implements the following modifications:
> . Currently, if a server is on leader election and doesn't receive a notification for some amount of time t, then it sends a new set of notifications if at least one server has delivered a message from the previous set. With this patch, the amount of time a server waits for a notification before sending a new set increases exponentially;
> . I have separated connecting to servers and queuing new notification messages. Before they were all in the same message. The advantage is that now I can tell to an instance of QuorumCnxManager to try to connect to other servers without generating new notification messages;
> . I have changed the logging level of several messages on QuorumCnxManager. They were "warn", but they should really be either "info" or "debug". I've changed them to info.
>
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (ZOOKEEPER-230) Improvements to FLE
Posted by "Flavio Paiva Junqueira (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/ZOOKEEPER-230?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Flavio Paiva Junqueira updated ZOOKEEPER-230:
---------------------------------------------
Attachment: ZOOKEEPER-230.patch
> Improvements to FLE
> -------------------
>
> Key: ZOOKEEPER-230
> URL: https://issues.apache.org/jira/browse/ZOOKEEPER-230
> Project: Zookeeper
> Issue Type: Improvement
> Components: leaderElection
> Affects Versions: 3.1.0
> Reporter: Flavio Paiva Junqueira
> Assignee: Flavio Paiva Junqueira
> Fix For: 3.0.0
>
> Attachments: ZOOKEEPER-230.patch
>
>
> I'm about to attach a patch that implements the following modifications:
> . Currently, if a server is on leader election and doesn't receive a notification for some amount of time t, then it sends a new set of notifications if at least one server has delivered a message from the previous set. With this patch, the amount of time a server waits for a notification before sending a new set increases exponentially;
> . I have separated connecting to servers and queuing new notification messages. Before they were all in the same message. The advantage is that now I can tell to an instance of QuorumCnxManager to try to connect to other servers without generating new notification messages;
> . I have changed the logging level of several messages on QuorumCnxManager. They were "warn", but they should really be either "info" or "debug". I've changed them to info.
>
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (ZOOKEEPER-230) Improvements to FLE
Posted by "Flavio Paiva Junqueira (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/ZOOKEEPER-230?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12650733#action_12650733 ]
Flavio Paiva Junqueira commented on ZOOKEEPER-230:
--------------------------------------------------
This patch is supposed to be an improvement, not a fix to a critical problem. I have seen applications that because of misconfiguration keep running even though the servers are not performing any useful work. In such cases, servers wake up frequently to send new notifications (also generating log messages) without the increase on the amount of time they wait until the next check. So, all I'm trying to do is reduce the amount of work a server performs in such cases by reducing the number of times it wakes up to check the queue of notifications over time. It doesn't hurt the performance of leader election, though, because if a server is partitioned, then messages will remain queued until the partition heals. You're right that the amount of time increases without bounds, and it might be important for long partitions, so I'm ok with having a an upper bound.
> Improvements to FLE
> -------------------
>
> Key: ZOOKEEPER-230
> URL: https://issues.apache.org/jira/browse/ZOOKEEPER-230
> Project: Zookeeper
> Issue Type: Improvement
> Components: leaderElection
> Affects Versions: 3.0.0
> Reporter: Flavio Paiva Junqueira
> Assignee: Flavio Paiva Junqueira
> Fix For: 3.1.0
>
> Attachments: ZOOKEEPER-230.patch
>
>
> I'm about to attach a patch that implements the following modifications:
> . Currently, if a server is on leader election and doesn't receive a notification for some amount of time t, then it sends a new set of notifications if at least one server has delivered a message from the previous set. With this patch, the amount of time a server waits for a notification before sending a new set increases exponentially;
> . I have separated connecting to servers and queuing new notification messages. Before they were all in the same message. The advantage is that now I can tell to an instance of QuorumCnxManager to try to connect to other servers without generating new notification messages;
> . I have changed the logging level of several messages on QuorumCnxManager. They were "warn", but they should really be either "info" or "debug". I've changed them to info.
>
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (ZOOKEEPER-230) Improvements to FLE
Posted by "Mahadev konar (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/ZOOKEEPER-230?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12652927#action_12652927 ]
Mahadev konar commented on ZOOKEEPER-230:
-----------------------------------------
the patch loooks good. here are some comments :
- 60000 should be declared as a final static int at the beginning of the class and used as a variable inside the code rather than using the value itself. Its easier to look for and replace.
- notimteout is updated twice - once when no new notifications are there and other time when its processing some notifications. Why is it incremented in the second case? also, its not bounded in the second case when notification != null
- LOG.info("Create new connection"); in QuorumCnxnManager does not give any info on which host its creating the connection to. The log message should be more informative so that its easy to find out what happened late. ( I understand its been the same earlier but it will be good if we can make it more informative).
- public void connect() . does this method need to be public? Also, some javadoc comments for developeers, on what the method does would be useful.
> Improvements to FLE
> -------------------
>
> Key: ZOOKEEPER-230
> URL: https://issues.apache.org/jira/browse/ZOOKEEPER-230
> Project: Zookeeper
> Issue Type: Improvement
> Components: leaderElection
> Affects Versions: 3.0.0
> Reporter: Flavio Paiva Junqueira
> Assignee: Flavio Paiva Junqueira
> Fix For: 3.1.0
>
> Attachments: ZOOKEEPER-230.patch, ZOOKEEPER-230.patch, ZOOKEEPER-230.patch
>
>
> I'm about to attach a patch that implements the following modifications:
> . Currently, if a server is on leader election and doesn't receive a notification for some amount of time t, then it sends a new set of notifications if at least one server has delivered a message from the previous set. With this patch, the amount of time a server waits for a notification before sending a new set increases exponentially;
> . I have separated connecting to servers and queuing new notification messages. Before they were all in the same message. The advantage is that now I can tell to an instance of QuorumCnxManager to try to connect to other servers without generating new notification messages;
> . I have changed the logging level of several messages on QuorumCnxManager. They were "warn", but they should really be either "info" or "debug". I've changed them to info.
>
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (ZOOKEEPER-230) Improvements to FLE
Posted by "Flavio Paiva Junqueira (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/ZOOKEEPER-230?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Flavio Paiva Junqueira updated ZOOKEEPER-230:
---------------------------------------------
Attachment: ZOOKEEPER-230.patch
Addressed Mahadev's comments. Thanks, Mahadev.
> Improvements to FLE
> -------------------
>
> Key: ZOOKEEPER-230
> URL: https://issues.apache.org/jira/browse/ZOOKEEPER-230
> Project: Zookeeper
> Issue Type: Improvement
> Components: leaderElection
> Affects Versions: 3.0.0
> Reporter: Flavio Paiva Junqueira
> Assignee: Flavio Paiva Junqueira
> Fix For: 3.1.0
>
> Attachments: ZOOKEEPER-230.patch, ZOOKEEPER-230.patch, ZOOKEEPER-230.patch, ZOOKEEPER-230.patch
>
>
> I'm about to attach a patch that implements the following modifications:
> . Currently, if a server is on leader election and doesn't receive a notification for some amount of time t, then it sends a new set of notifications if at least one server has delivered a message from the previous set. With this patch, the amount of time a server waits for a notification before sending a new set increases exponentially;
> . I have separated connecting to servers and queuing new notification messages. Before they were all in the same message. The advantage is that now I can tell to an instance of QuorumCnxManager to try to connect to other servers without generating new notification messages;
> . I have changed the logging level of several messages on QuorumCnxManager. They were "warn", but they should really be either "info" or "debug". I've changed them to info.
>
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (ZOOKEEPER-230) Improvements to FLE
Posted by "Mahadev konar (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/ZOOKEEPER-230?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12650664#action_12650664 ]
Mahadev konar commented on ZOOKEEPER-230:
-----------------------------------------
2 and 3 seem like a good change.
> Improvements to FLE
> -------------------
>
> Key: ZOOKEEPER-230
> URL: https://issues.apache.org/jira/browse/ZOOKEEPER-230
> Project: Zookeeper
> Issue Type: Improvement
> Components: leaderElection
> Affects Versions: 3.0.0
> Reporter: Flavio Paiva Junqueira
> Assignee: Flavio Paiva Junqueira
> Fix For: 3.1.0
>
> Attachments: ZOOKEEPER-230.patch
>
>
> I'm about to attach a patch that implements the following modifications:
> . Currently, if a server is on leader election and doesn't receive a notification for some amount of time t, then it sends a new set of notifications if at least one server has delivered a message from the previous set. With this patch, the amount of time a server waits for a notification before sending a new set increases exponentially;
> . I have separated connecting to servers and queuing new notification messages. Before they were all in the same message. The advantage is that now I can tell to an instance of QuorumCnxManager to try to connect to other servers without generating new notification messages;
> . I have changed the logging level of several messages on QuorumCnxManager. They were "warn", but they should really be either "info" or "debug". I've changed them to info.
>
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (ZOOKEEPER-230) Improvements to FLE
Posted by "Mahadev konar (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/ZOOKEEPER-230?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Mahadev konar updated ZOOKEEPER-230:
------------------------------------
Status: Open (was: Patch Available)
cacelling the patch for now..
> Improvements to FLE
> -------------------
>
> Key: ZOOKEEPER-230
> URL: https://issues.apache.org/jira/browse/ZOOKEEPER-230
> Project: Zookeeper
> Issue Type: Improvement
> Components: leaderElection
> Affects Versions: 3.0.0
> Reporter: Flavio Paiva Junqueira
> Assignee: Flavio Paiva Junqueira
> Fix For: 3.1.0
>
> Attachments: ZOOKEEPER-230.patch, ZOOKEEPER-230.patch, ZOOKEEPER-230.patch
>
>
> I'm about to attach a patch that implements the following modifications:
> . Currently, if a server is on leader election and doesn't receive a notification for some amount of time t, then it sends a new set of notifications if at least one server has delivered a message from the previous set. With this patch, the amount of time a server waits for a notification before sending a new set increases exponentially;
> . I have separated connecting to servers and queuing new notification messages. Before they were all in the same message. The advantage is that now I can tell to an instance of QuorumCnxManager to try to connect to other servers without generating new notification messages;
> . I have changed the logging level of several messages on QuorumCnxManager. They were "warn", but they should really be either "info" or "debug". I've changed them to info.
>
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.