[jira] Commented: (HDFS-86) Corrupted blocks get deleted but not replicated
[ https://issues.apache.org/jira/browse/HDFS-86?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12900342#action_12900342 ] Thanh Do commented on HDFS-86: -- i have a cluster of two nodes. Say a block with 2 replicas, and one of them get corrupted. The corrupted block is reported to NN, but it is never deleted or replicated, even after NN restarts. Not sure this is a bug or just a policy. I am playing the append-trunk Corrupted blocks get deleted but not replicated --- Key: HDFS-86 URL: https://issues.apache.org/jira/browse/HDFS-86 Project: Hadoop HDFS Issue Type: Bug Reporter: Hairong Kuang Assignee: Hairong Kuang Attachments: blockInvalidate.patch When I test the patch to HADOOP-1345 on a two node dfs cluster, I see that dfs correctly delete the corrupted replica and successfully retry reading from the other correct replica, but the block does not get replicated. The block remains with only 1 replica until the next block report comes in. In my testcase, since the dfs cluster has only 2 datanodes, the target of replication is the same as the target of block invalidation. After poking the logs, I found out that the namenode sent the replication request before the block invalidation request. This is because the namenode does not invalidate a block well. In FSNamesystem.invalidateBlock, it first puts the invalidate request in a queue and then immediately removes the replica from its state, which triggers the choosing a target for the block. When requests are sent back to the target datanode as a reply to a heartbeat message, the replication requests have higher priority than the invalidate requests. This problem could be solved if a namenode removes an invalidated replica from its state only after the invalidate request is sent to the datanode. -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
[jira] Commented: (HDFS-86) Corrupted blocks get deleted but not replicated
[ https://issues.apache.org/jira/browse/HDFS-86?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12900357#action_12900357 ] Hairong Kuang commented on HDFS-86: --- This jira is too old. It should be closed. Now HDFS has a different policy with corrupt replicas. A corrupt replica does not get deleted until a good replica gets replicated. The problem you have is caused by the 2-node cluster. Because it does not an extra node to place the good replica, the corrupt one never gets deleted. If you add one more node to the cluster, the problem will go away. Corrupted blocks get deleted but not replicated --- Key: HDFS-86 URL: https://issues.apache.org/jira/browse/HDFS-86 Project: Hadoop HDFS Issue Type: Bug Reporter: Hairong Kuang Assignee: Hairong Kuang Attachments: blockInvalidate.patch When I test the patch to HADOOP-1345 on a two node dfs cluster, I see that dfs correctly delete the corrupted replica and successfully retry reading from the other correct replica, but the block does not get replicated. The block remains with only 1 replica until the next block report comes in. In my testcase, since the dfs cluster has only 2 datanodes, the target of replication is the same as the target of block invalidation. After poking the logs, I found out that the namenode sent the replication request before the block invalidation request. This is because the namenode does not invalidate a block well. In FSNamesystem.invalidateBlock, it first puts the invalidate request in a queue and then immediately removes the replica from its state, which triggers the choosing a target for the block. When requests are sent back to the target datanode as a reply to a heartbeat message, the replication requests have higher priority than the invalidate requests. This problem could be solved if a namenode removes an invalidated replica from its state only after the invalidate request is sent to the datanode. -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
[jira] Resolved: (HDFS-86) Corrupted blocks get deleted but not replicated
[ https://issues.apache.org/jira/browse/HDFS-86?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Hairong Kuang resolved HDFS-86. --- Resolution: Invalid Corrupted blocks get deleted but not replicated --- Key: HDFS-86 URL: https://issues.apache.org/jira/browse/HDFS-86 Project: Hadoop HDFS Issue Type: Bug Reporter: Hairong Kuang Assignee: Hairong Kuang Attachments: blockInvalidate.patch When I test the patch to HADOOP-1345 on a two node dfs cluster, I see that dfs correctly delete the corrupted replica and successfully retry reading from the other correct replica, but the block does not get replicated. The block remains with only 1 replica until the next block report comes in. In my testcase, since the dfs cluster has only 2 datanodes, the target of replication is the same as the target of block invalidation. After poking the logs, I found out that the namenode sent the replication request before the block invalidation request. This is because the namenode does not invalidate a block well. In FSNamesystem.invalidateBlock, it first puts the invalidate request in a queue and then immediately removes the replica from its state, which triggers the choosing a target for the block. When requests are sent back to the target datanode as a reply to a heartbeat message, the replication requests have higher priority than the invalidate requests. This problem could be solved if a namenode removes an invalidated replica from its state only after the invalidate request is sent to the datanode. -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
[jira] Commented: (HDFS-1346) DFSClient receives out of order packet ack
[ https://issues.apache.org/jira/browse/HDFS-1346?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12900465#action_12900465 ] Hairong Kuang commented on HDFS-1346: - Todd, yours is missing this patch: https://issues.apache.org/jira/secure/attachment/12439379/pipelineHeartbeat.patch. HDFS-101 says that it fixes a bug of incorrect handle of pipeline heartbeat in yahoo's hadoop security branch 0.20. But I did not put the bug description there. Koji, do you still remember what exact problem that pipelineHeartbeat.patch is fixed? DFSClient receives out of order packet ack -- Key: HDFS-1346 URL: https://issues.apache.org/jira/browse/HDFS-1346 Project: Hadoop HDFS Issue Type: Bug Components: data-node, hdfs client Affects Versions: 0.20-append Reporter: Hairong Kuang Assignee: Hairong Kuang Fix For: 0.20-append Attachments: blockrecv-diff.txt, outOfOrder.patch When running 0.20 patched with HDFS-101, we sometimes see an error as follow: WARN hdfs.DFSClient: DFSOutputStream ResponseProcessor exception for block blk_-2871223654872350746_21421120java.io.IOException: Responseprocessor: Expecting seq no for block blk_-2871223654872350746_21421120 10280 but received 10281 at org.apache.hadoop.hdfs.DFSClient$DFSOutputStream$ResponseProcessor.run(DFSClient.java:2570) This indicates that DFS client expects an ack for packet N, but receives an ack for packet N+1. -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
[jira] Commented: (HDFS-1347) TestDelegationToken uses mortbay.log for logging
[ https://issues.apache.org/jira/browse/HDFS-1347?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12900467#action_12900467 ] Boris Shkolnik commented on HDFS-1347: -- ran tests manually, ant test - passed. TestDelegationToken uses mortbay.log for logging Key: HDFS-1347 URL: https://issues.apache.org/jira/browse/HDFS-1347 Project: Hadoop HDFS Issue Type: Bug Components: test Affects Versions: 0.22.0 Reporter: Boris Shkolnik Assignee: Boris Shkolnik Fix For: 0.22.0 Attachments: HDFS-1347.patch needs to be changed to commons.log -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
[jira] Commented: (HDFS-1320) Add LOG.isDebugEnabled() guard for each LOG.debug(...)
[ https://issues.apache.org/jira/browse/HDFS-1320?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12900484#action_12900484 ] Tsz Wo (Nicholas), SZE commented on HDFS-1320: -- does the JVM not optimize for this case in the fast-path? Hi Ryan, from the benchmark results [here|https://issues.apache.org/jira/browse/HADOOP-6884?focusedCommentId=12900087page=com.atlassian.jira.plugin.system.issuetabpanels%3Acomment-tabpanel#action_12900087], it does not seem JVM optimized this. I think JVM cannot do anything in general since parameter evaluation may have side-effect. It is hard for the JVM to determine whether it is safe to skip those instructions. Add LOG.isDebugEnabled() guard for each LOG.debug(...) Key: HDFS-1320 URL: https://issues.apache.org/jira/browse/HDFS-1320 Project: Hadoop HDFS Issue Type: Improvement Affects Versions: 0.22.0 Reporter: Erik Steffl Assignee: Erik Steffl Fix For: 0.22.0 Attachments: HDFS-1320-0.22-1.patch, HDFS-1320-0.22-2.patch, HDFS-1320-0.22.patch Each LOG.debug(...) should be executed only if LOG.isDebugEnabled() is true, in some cases it's expensive to construct the string that is being printed to log. It's much easier to always use LOG.isDebugEnabled() because it's easier to check (rather than in each case reason wheather it's neccessary or not). -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
[jira] Created: (HDFS-1348) DecommissionManager holds fsnamesystem lock during the whole process of checking if decomissioning DataNodes are finished or not
DecommissionManager holds fsnamesystem lock during the whole process of checking if decomissioning DataNodes are finished or not Key: HDFS-1348 URL: https://issues.apache.org/jira/browse/HDFS-1348 Project: Hadoop HDFS Issue Type: Improvement Components: name-node Reporter: Hairong Kuang Assignee: Hairong Kuang Fix For: 0.22.0 NameNode normally is busy all the time. Its log is full of activities every second. But once for a while, NameNode seems to pause for more than 10 seconds without doing anything, leaving a blank in its log even though no garbage collection is happening. One culprit is DecommionManager. Its monitor holds the fsynamesystem lock during the whole process of checking if decomissioning DataNodes are finished or not, during which it checks every block of up to a default of 5 datanodes. -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
[jira] Commented: (HDFS-1073) Simpler model for Namenode's fs Image and edit Logs
[ https://issues.apache.org/jira/browse/HDFS-1073?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12900499#action_12900499 ] Todd Lipcon commented on HDFS-1073: --- Hey Sanjay, Thanks for reviving this. The notes you wrote above seem accurate. Couple of questions: bq. while writing edit logs to multiple files, a failure of the th system can result in different amounts of data written to each file - the tid allows one to pick one with the most tranasactions. Isn't this also doable by just seeing which as more non-zero bytes? ie seek to the end of the file, scan backwards through the 0 bytes, and stop. Whichever valid log is longer wins. Even in the case with the transaction-id, you have to do something like this for a few reasons: a) we'd rather scan backward from the end of the edit log than forward from the beginning, since it's going to be a faster startup, and b) even if we see a higher transaction id header on the last entry, that entry might have been incompletely written to the file, so we still have to verify that it deserializes correctly. bq. Main disadvantage is that the editlogs will be little bigger. So are you suggesting that each edit will include a header with the transaction ID in it? Isn't this redundant if the header of the whole edit file has the starting txid -- ie is there ever a case where we'd skip a txid? bq. In order to do an offline fsck one can needs to dump the block map; clearly one does not want to the local the system to do an atomic dump. The transaction id of when the dump is started can be written in the dump to allow the fsck to report consistently. Sorry, can you elaborate a little bit here? In order to get a consistent dump of the block map don't we need to take the FSN lock and thus stall all operations? Is the idea that the BackupNode would do the blockmap dump offline since it can hold a lock for some time without stalling clients? If that's the case, what's the purpose of the offline nature of the fsck instead of just having BackupNode allow fsck to point directly at it and access memory under the same lock? Mahadev said: bq. Is it the minimum set of code changes that is making you guys reject on the txn based snapshots and logging? I don't think either way has been decided/rejected yet. What you're saying has been my view - that doing txid based is a bigger change, since we have to introduce the txid concept and add extra code that allows replaying partial edit log files (ie a subrange of the edits within). But it's certainly doable and Sanjay has presented some good advantages. Simpler model for Namenode's fs Image and edit Logs Key: HDFS-1073 URL: https://issues.apache.org/jira/browse/HDFS-1073 Project: Hadoop HDFS Issue Type: Improvement Reporter: Sanjay Radia Assignee: Todd Lipcon Attachments: hdfs1073.pdf The naming and handling of NN's fsImage and edit logs can be significantly improved resulting simpler and more robust code. -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
[jira] Updated: (HDFS-1347) TestDelegationToken uses mortbay.log for logging
[ https://issues.apache.org/jira/browse/HDFS-1347?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Boris Shkolnik updated HDFS-1347: - Status: Resolved (was: Patch Available) Resolution: Fixed committed to trunk TestDelegationToken uses mortbay.log for logging Key: HDFS-1347 URL: https://issues.apache.org/jira/browse/HDFS-1347 Project: Hadoop HDFS Issue Type: Bug Components: test Affects Versions: 0.22.0 Reporter: Boris Shkolnik Assignee: Boris Shkolnik Fix For: 0.22.0 Attachments: HDFS-1347.patch needs to be changed to commons.log -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
[jira] Updated: (HDFS-535) TestFileCreation occasionally fails because of an exception in DataStreamer.
[ https://issues.apache.org/jira/browse/HDFS-535?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Jakob Homan updated HDFS-535: - Status: Open (was: Patch Available) This patch does indeed cause testFsCloseAfterClusterShutdown to fail for me: {noformat}4893 Testcase: testFsClose took 3.137 sec 4894 Testcase: testFsCloseAfterClusterShutdown took 2.751 sec 4895 FAILED 4896 Failed to close file after cluster shutdown 4897 junit.framework.AssertionFailedError: Failed to close file after cluster shutdown 4898 at org.apache.hadoop.hdfs.TestFileCreation.testFsCloseAfterClusterShutdown(TestFileCreation.java:851){noformat} Canceling patch for Konstanin to update, although I don't believe we've seen this problem for a while, so may we can just close this issue? TestFileCreation occasionally fails because of an exception in DataStreamer. Key: HDFS-535 URL: https://issues.apache.org/jira/browse/HDFS-535 Project: Hadoop HDFS Issue Type: Bug Components: hdfs client, test Affects Versions: 0.20.1 Reporter: Konstantin Shvachko Assignee: Konstantin Shvachko Attachments: TestFileCreate.patch One of test cases, namely {{testFsCloseAfterClusterShutdown()}}, of {{TestFileCreation}} fails occasionally. -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
[jira] Commented: (HDFS-718) configuration parameter to prevent accidental formatting of HDFS filesystem
[ https://issues.apache.org/jira/browse/HDFS-718?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12900580#action_12900580 ] Jakob Homan commented on HDFS-718: -- I'm +1. Once an Hadoop cluster is up and running in production it can potentially hold very critical and valuable information. An extra, optional safeguard that saves one such cluster and doesn't add any serious complexity to the code is worth it. A steadystate cluster is a very valuable thing... configuration parameter to prevent accidental formatting of HDFS filesystem --- Key: HDFS-718 URL: https://issues.apache.org/jira/browse/HDFS-718 Project: Hadoop HDFS Issue Type: Improvement Components: name-node Affects Versions: 0.22.0 Environment: Any Reporter: Andrew Ryan Assignee: Andrew Ryan Priority: Minor Attachments: HDFS-718.patch-2.txt, HDFS-718.patch.txt Currently, any time the NameNode is not running, an HDFS filesystem will accept the 'format' command, and will duly format itself. There are those of us who have multi-PB HDFS filesystems who are really quite uncomfortable with this behavior. There is Y/N confirmation in the format command, but if the formatter genuinely believes themselves to be doing the right thing, the filesystem will be formatted. This patch adds a configuration parameter to the namenode, dfs.namenode.support.allowformat, which defaults to true, the current behavior: always allow formatting if the NameNode is down or some other process is not holding the namenode lock. But if dfs.namenode.support.allowformat is set to false, the NameNode will not allow itself to be formatted until this config parameter is changed to true. The general idea is that for production HDFS filesystems, the user would format the HDFS once, then set dfs.namenode.support.allowformat to false for all time. The attached patch was generated against trunk and +1's on my test machine. We have a 0.20 version that we are using in our cluster as well. -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.