[jira] Updated: (ZOOKEEPER-335) zookeeper servers should commit the new leader txn to their logs.
[ https://issues.apache.org/jira/browse/ZOOKEEPER-335?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Travis Crawford updated ZOOKEEPER-335: -- Attachment: ZOOKEEPER-790.travis.log.bz2 Please see my most recent comment for a summary of where the interesting lines are in this log file. zookeeper servers should commit the new leader txn to their logs. - Key: ZOOKEEPER-335 URL: https://issues.apache.org/jira/browse/ZOOKEEPER-335 Project: Zookeeper Issue Type: Bug Components: server Affects Versions: 3.1.0 Reporter: Mahadev konar Assignee: Mahadev konar Priority: Blocker Fix For: 3.4.0 Attachments: faultynode-vishal.txt, zk.log.gz, zklogs.tar.gz, ZOOKEEPER-790.travis.log.bz2 currently the zookeeper followers do not commit the new leader election. This will cause problems in a failure scenarios with a follower acking to the same leader txn id twice, which might be two different intermittent leaders and allowing them to propose two different txn's of the same zxid. -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
[jira] Updated: (ZOOKEEPER-335) zookeeper servers should commit the new leader txn to their logs.
[ https://issues.apache.org/jira/browse/ZOOKEEPER-335?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Vishal K updated ZOOKEEPER-335: --- Attachment: faultynode-vishal.txt Apologies for multiple attachments. zookeeper servers should commit the new leader txn to their logs. - Key: ZOOKEEPER-335 URL: https://issues.apache.org/jira/browse/ZOOKEEPER-335 Project: Zookeeper Issue Type: Bug Components: server Affects Versions: 3.1.0 Reporter: Mahadev konar Assignee: Mahadev konar Priority: Blocker Fix For: 3.4.0 Attachments: faultynode-vishal.txt, zk.log.gz, zklogs.tar.gz currently the zookeeper followers do not commit the new leader election. This will cause problems in a failure scenarios with a follower acking to the same leader txn id twice, which might be two different intermittent leaders and allowing them to propose two different txn's of the same zxid. -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
[jira] Updated: (ZOOKEEPER-335) zookeeper servers should commit the new leader txn to their logs.
[ https://issues.apache.org/jira/browse/ZOOKEEPER-335?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Vishal K updated ZOOKEEPER-335: --- Attachment: zklogs.tar.gz Hi, I have attached the logs. The log entries are similar to those reported by others. We are testing with 3 nodes. Each node is run in a VM running SLES 11. All 3 VMs are run on the same host. VMs are sharing the same disk. cpuinfo and meminfo for VM is in the attached file. I have also tried to collect more info with -verbose:gc -Xloggc:/../ -Xprof options to java. gc.lg contains the gc output rest of the info should be in msgs.log Default java heap size was used. java version 1.6.0_18 was used. One point to note - In my case, on the misbehaving node could not joing the cluster. Rest of the cluster was stable (except for the flood of log messages on the leader because the misbehaving follower kept terminating session). Hope this helps. Thanks. zookeeper servers should commit the new leader txn to their logs. - Key: ZOOKEEPER-335 URL: https://issues.apache.org/jira/browse/ZOOKEEPER-335 Project: Zookeeper Issue Type: Bug Components: server Affects Versions: 3.1.0 Reporter: Mahadev konar Assignee: Mahadev konar Priority: Blocker Fix For: 3.4.0 Attachments: zk.log.gz, zklogs.tar.gz currently the zookeeper followers do not commit the new leader election. This will cause problems in a failure scenarios with a follower acking to the same leader txn id twice, which might be two different intermittent leaders and allowing them to propose two different txn's of the same zxid. -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
[jira] Updated: (ZOOKEEPER-335) zookeeper servers should commit the new leader txn to their logs.
[ https://issues.apache.org/jira/browse/ZOOKEEPER-335?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Mike Solomon updated ZOOKEEPER-335: --- Attachment: zk.log.gz This is the log file of my second attempt to restart a wedged zookeeper server. zookeeper servers should commit the new leader txn to their logs. - Key: ZOOKEEPER-335 URL: https://issues.apache.org/jira/browse/ZOOKEEPER-335 Project: Zookeeper Issue Type: Bug Components: server Affects Versions: 3.1.0 Reporter: Mahadev konar Assignee: Mahadev konar Priority: Blocker Fix For: 3.4.0 Attachments: zk.log.gz currently the zookeeper followers do not commit the new leader election. This will cause problems in a failure scenarios with a follower acking to the same leader txn id twice, which might be two different intermittent leaders and allowing them to propose two different txn's of the same zxid. -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
[jira] Updated: (ZOOKEEPER-335) zookeeper servers should commit the new leader txn to their logs.
[ https://issues.apache.org/jira/browse/ZOOKEEPER-335?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Patrick Hunt updated ZOOKEEPER-335: --- Priority: Blocker (was: Major) Raising to blocker level - we've seen this reported by users a couple times now. zookeeper servers should commit the new leader txn to their logs. - Key: ZOOKEEPER-335 URL: https://issues.apache.org/jira/browse/ZOOKEEPER-335 Project: Zookeeper Issue Type: Bug Components: server Affects Versions: 3.1.0 Reporter: Mahadev konar Assignee: Mahadev konar Priority: Blocker Fix For: 3.4.0 currently the zookeeper followers do not commit the new leader election. This will cause problems in a failure scenarios with a follower acking to the same leader txn id twice, which might be two different intermittent leaders and allowing them to propose two different txn's of the same zxid. -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
[jira] Updated: (ZOOKEEPER-335) zookeeper servers should commit the new leader txn to their logs.
[ https://issues.apache.org/jira/browse/ZOOKEEPER-335?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Mahadev konar updated ZOOKEEPER-335: Fix Version/s: (was: 3.3.0) 3.4.0 given the change is non trivial as mentioned above, I am moving this to 3.4 for now. zookeeper servers should commit the new leader txn to their logs. - Key: ZOOKEEPER-335 URL: https://issues.apache.org/jira/browse/ZOOKEEPER-335 Project: Zookeeper Issue Type: Bug Components: server Affects Versions: 3.1.0 Reporter: Mahadev konar Assignee: Mahadev konar Fix For: 3.4.0 currently the zookeeper followers do not commit the new leader election. This will cause problems in a failure scenarios with a follower acking to the same leader txn id twice, which might be two different intermittent leaders and allowing them to propose two different txn's of the same zxid. -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
[jira] Updated: (ZOOKEEPER-335) zookeeper servers should commit the new leader txn to their logs.
[ https://issues.apache.org/jira/browse/ZOOKEEPER-335?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Mahadev konar updated ZOOKEEPER-335: Fix Version/s: (was: 3.2.0) 3.3.0 to fix this issue we require server -server protocol change. Thsi protocol change will break backwards compatibility. To maintain backwards compatibility the code becomes quite complex and tricky. Instead of making a last minute change and having to do al lthe testing to check if backwards compatibily for servers is maintained, I am moving it to 3.3 to see if we want to fix it in a backwards compatible manner or fix it in 4.0 and break backwards compatibility. zookeeper servers should commit the new leader txn to their logs. - Key: ZOOKEEPER-335 URL: https://issues.apache.org/jira/browse/ZOOKEEPER-335 Project: Zookeeper Issue Type: Bug Components: server Affects Versions: 3.1.0 Reporter: Mahadev konar Assignee: Mahadev konar Fix For: 3.3.0 currently the zookeeper followers do not commit the new leader election. This will cause problems in a failure scenarios with a follower acking to the same leader txn id twice, which might be two different intermittent leaders and allowing them to propose two different txn's of the same zxid. -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
