| clear query|facets|time |
Search criteria: .
Results from 121 to 130 from
491 (0.348s).
|
|
|
Loading phrases to help you
refine your search...
|
|
[HDFS-862] Potential NN deadlock in processDistributedUpgradeCommand - HDFS - [issue]
|
|
...Haven't seen this in practice, but the lock order is inconsistent. processReport locks FSNamesystem, then calls UpgradeManager.startUpgrade, getUpgradeState, and getUpgradeStatus (each of wh...
|
|
|
http://issues.apache.org/jira/browse/HDFS-862
Author: Todd Lipcon,
2012-09-11, 06:57
|
|
|
[HDFS-3915] QJM: Failover fails with auth error in secure cluster - HDFS - [issue]
|
|
...When testing failover in a secure cluster with QJM, we ran into the following error:java.io.IOException: Exception trying to open authenticated connection to http://xxxxx:8480/getJournal?jid...
|
|
|
http://issues.apache.org/jira/browse/HDFS-3915
Author: Todd Lipcon,
2012-09-11, 05:53
|
|
|
[HDFS-3914] QJM: acceptRecovery should abort current segment - HDFS - [issue]
|
|
...Found this bug with randomized testing. The following sequence causes a problem: JN writing segment starting at txid 1, and successfully wrote txid 1, but no more JN becomes partitioned from...
|
|
|
http://issues.apache.org/jira/browse/HDFS-3914
Author: Todd Lipcon,
2012-09-11, 00:21
|
|
|
[HDFS-3899] QJM: Writer-side metrics - HDFS - [issue]
|
|
...We already have some metrics on the server side (JournalNode) but it's useful to also gather metrics from the client side (NameNode). This is important in order to monitor that the client is...
|
|
|
http://issues.apache.org/jira/browse/HDFS-3899
Author: Todd Lipcon,
2012-09-10, 23:40
|
|
|
[HDFS-3901] QJM: send 'heartbeat' messages to JNs even when they are out-of-sync - HDFS - [issue]
|
|
...Currently, if one of the JNs has fallen out of sync with the writer (eg because it went down), it will be marked as such until the next log roll. This causes the writer to no longer send any...
|
|
|
http://issues.apache.org/jira/browse/HDFS-3901
Author: Todd Lipcon,
2012-09-10, 23:32
|
|
|
[HDFS-3900] QJM: avoid validating log segments on log rolls - HDFS - [issue]
|
|
...Currently, we are paranoid and validate every log segment when it is finalized. For the a log segment that has been written entirely by one writer, with no recovery in between, this is overl...
|
|
|
http://issues.apache.org/jira/browse/HDFS-3900
Author: Todd Lipcon,
2012-09-10, 19:54
|
|
|
[HDFS-3885] QJM: optimize log sync when JN is lagging behind - HDFS - [issue]
|
|
...This is a potential optimization that we can add to the JournalNode: when one of the nodes is lagging behind the others (eg because its local disk is slower or there was a network blip), it ...
|
|
|
http://issues.apache.org/jira/browse/HDFS-3885
Author: Todd Lipcon,
2012-09-10, 19:51
|
|
|
[HDFS-3898] QJM: enable TCP_NODELAY for IPC - HDFS - [issue]
|
|
...Currently, if the size of the edits batches is larger than the MTU, it can result in 40ms delays due to interaction between nagle's algorithm and delayed ack. Enabling TCP_NODELAY on the soc...
|
|
|
http://issues.apache.org/jira/browse/HDFS-3898
Author: Todd Lipcon,
2012-09-10, 19:47
|
|
|
[HDFS-3438] BootstrapStandby should not require a rollEdits on active node - HDFS - [issue]
|
|
...Currently, BootstrapStandby uses a rollEdits() call on the active NN in order to determine the most recent checkpoint and transaction ID. However, this means that you cannot bootstrap a stan...
|
|
|
http://issues.apache.org/jira/browse/HDFS-3438
Author: Todd Lipcon,
2012-09-07, 22:09
|
|
|
[HDFS-3548] NamenodeFsck.copyBlock fails to create a Block Reader - HDFS - [issue]
|
|
...NamenodeFsck.copyBlock creates a Socket using new Socket(), and thus that socket doesn't have an associated Channel. Then, it fails to create a BlockReader since RemoteBlockReader2 needs a s...
|
|
|
http://issues.apache.org/jira/browse/HDFS-3548
Author: Todd Lipcon,
2012-09-07, 22:09
|
|
|
|
