[jira] Commented: (HADOOP-181) task trackers should not restart for having a late heartbeat

[ http://issues.apache.org/jira/browse/HADOOP-181?page=comments#action_12427115 ] Doug Cutting commented on HADOOP-181: - > improved detection of tasktracker death is a separate issue It's certainly related. This issue deals with fixing thin

[jira] Created: (HADOOP-441) SequenceFile should support 'custom compressors'

SequenceFile should support 'custom compressors' Key: HADOOP-441 URL: http://issues.apache.org/jira/browse/HADOOP-441 Project: Hadoop Issue Type: New Feature Components: io

[jira] Created: (HADOOP-440) In streaming, error messages issued in stream mappers or reducers don't go anywhere

In streaming, error messages issued in stream mappers or reducers don't go anywhere --- Key: HADOOP-440 URL: http://issues.apache.org/jira/browse/HADOOP-440 Project: Hado

[jira] Created: (HADOOP-439) Streaming does not work for text data if the records don't fit in a short UTF8 [2^16/3 characters]

Streaming does not work for text data if the records don't fit in a short UTF8 [2^16/3 characters] -- Key: HADOOP-439 URL: http://issues.apache.org/jira/browse/HADOOP-

[jira] Created: (HADOOP-438) DFS pathname limitation.

DFS pathname limitation. Key: HADOOP-438 URL: http://issues.apache.org/jira/browse/HADOOP-438 Project: Hadoop Issue Type: Bug Components: dfs Affects Versions: 0.5.0, 0.4.0, 0.3.2, 0.3.1, 0.3.0, 0.2.1,

[jira] Created: (HADOOP-437) support gzip input files

support gzip input files Key: HADOOP-437 URL: http://issues.apache.org/jira/browse/HADOOP-437 Project: Hadoop Issue Type: Improvement Components: contrib/streaming Reporter: Michel Tourn

Re: [jira] Commented: (HADOOP-433) Better access to the RecordReader

Advancing the reader sounds like "a bad idea". But an exotic reader might have all kinds of context it could publish. maybe current line number, rowID, SQL statement used... Who knows. There could be lots of stuff. It would be nice to have an interface that lets you get to any methods

[jira] Commented: (HADOOP-181) task trackers should not restart for having a late heartbeat

[ http://issues.apache.org/jira/browse/HADOOP-181?page=comments#action_12427052 ] eric baldeschwieler commented on HADOOP-181: Would this have a substantial impact on the large sort benchmark tests our team runs? > task trackers sho

[jira] Commented: (HADOOP-431) default behaviour of dfsShell -rm should resemble 'rm -i', not 'rm -rf'

[ http://issues.apache.org/jira/browse/HADOOP-431?page=comments#action_12427048 ] eric baldeschwieler commented on HADOOP-431: I'd support a -r command to remove a directory. A lot of user commands may not be on reduce directories,

Re: [jira] Commented: (HADOOP-433) Better access to the RecordReader

just fyi, here are the reasons we tend to encounter: - regex'd metadata from a filename (for example, date, data source, etc... anything that exists on a per-file basis and not embedded in every record) - error message ("illegal widget name", "file XXX"), usually in an output record which is late

Re: [jira] Commented: (HADOOP-433) Better access to the RecordReader

On Aug 9, 2006, at 12:21 PM, Eric Baldeschwieler wrote: Why not provide a pointer to the real record reader? Seems like a valid OO way to get access to all kinds of things. Those attributes were put in to the JobConf so that Hadoop could re- run an isolated task, so they had to be serializ

[jira] Commented: (HADOOP-181) task trackers should not restart for having a late heartbeat

[ http://issues.apache.org/jira/browse/HADOOP-181?page=comments#action_12427033 ] Sameer Paranjpye commented on HADOOP-181: - I feel that improved detection of tasktracker death is a separate issue, which needs addressing. At the same tim

[jira] Updated: (HADOOP-373) Some calls to mkdirs do not check return value

[ http://issues.apache.org/jira/browse/HADOOP-373?page=all ] Wendy Chien updated HADOOP-373: --- Attachment: hadoop-373.patch > Some calls to mkdirs do not check return value > -- > > Key: HADOOP-373

[jira] Updated: (HADOOP-373) Some calls to mkdirs do not check return value

[ http://issues.apache.org/jira/browse/HADOOP-373?page=all ] Wendy Chien updated HADOOP-373: --- Status: Patch Available (was: Open) I've changed calls which did not check the return value. > Some calls to mkdirs do not check return value >

[jira] Commented: (HADOOP-64) DataNode should be capable of managing multiple volumes

[ http://issues.apache.org/jira/browse/HADOOP-64?page=comments#action_12427026 ] Konstantin Shvachko commented on HADOOP-64: --- This proposition looks good to me. The only thing that seems excessive is the dynamic data structures for mai

Re: [jira] Commented: (HADOOP-433) Better access to the RecordReader

Why not provide a pointer to the real record reader? Seems like a valid OO way to get access to all kinds of things. On Aug 8, 2006, at 3:48 PM, Owen O'Malley (JIRA) wrote: [ http://issues.apache.org/jira/browse/HADOOP-433? page=comments#action_12426763 ] Owen O'Malley commented on HAD

[jira] Created: (HADOOP-436) Concluding that the Map task failed may not be always right in getMapOutput.jsp

Concluding that the Map task failed may not be always right in getMapOutput.jsp --- Key: HADOOP-436 URL: http://issues.apache.org/jira/browse/HADOOP-436 Project: Hadoop

[jira] Updated: (HADOOP-434) Use Hadoop scripts to run smallJobsBenchmark to avoid classpath issues.

[ http://issues.apache.org/jira/browse/HADOOP-434?page=all ] Sanjay Dahiya updated HADOOP-434: - Attachment: patch_benchmark.txt sorry about that. The attached patch works for latest trunk. > Use Hadoop scripts to run smallJobsBenchmark to avoid classp

[jira] Updated: (HADOOP-324) "IOException: No space left on device" is handled incorrectly

[ http://issues.apache.org/jira/browse/HADOOP-324?page=all ] Doug Cutting updated HADOOP-324: Status: Resolved (was: Patch Available) Fix Version/s: 0.6.0 Resolution: Fixed I just committed this. Thanks! > "IOException: No space left

[jira] Updated: (HADOOP-434) Use Hadoop scripts to run smallJobsBenchmark to avoid classpath issues.

[ http://issues.apache.org/jira/browse/HADOOP-434?page=all ] Doug Cutting updated HADOOP-434: Status: Open (was: Patch Available) Fix Version/s: 0.6.0 I can't get this patch to apply to the current trunk. > Use Hadoop scripts to run smallJob

[jira] Created: (HADOOP-435) Encapsulating startup scripts and jars in a single Jar file.

Encapsulating startup scripts and jars in a single Jar file. Key: HADOOP-435 URL: http://issues.apache.org/jira/browse/HADOOP-435 Project: Hadoop Issue Type: New Feature Affect

[jira] Updated: (HADOOP-263) task status should include timestamps for when a job transitions

[ http://issues.apache.org/jira/browse/HADOOP-263?page=all ] Sanjay Dahiya updated HADOOP-263: - Attachment: patch.txt A new patch, now shows start, finish times for tasks and shuffle/sort finished time for reduce tasks. few points, which should probabl

[jira] Commented: (HADOOP-433) Better access to the RecordReader

[ http://issues.apache.org/jira/browse/HADOOP-433?page=comments#action_12426928 ] Benjamin Reed commented on HADOOP-433: -- If I understand correctly, you are suggesting that I instantiate another RecordReader using the information in JobConf

[jira] Updated: (HADOOP-400) the job tracker re-runs failed tasks on the same node

[ http://issues.apache.org/jira/browse/HADOOP-400?page=all ] Doug Cutting updated HADOOP-400: Status: Resolved (was: Patch Available) Resolution: Fixed I just committed this. Thanks, Owen! > the job tracker re-runs failed tasks on the same nod

[jira] Commented: (HADOOP-181) task trackers should not restart for having a late heartbeat

[ http://issues.apache.org/jira/browse/HADOOP-181?page=comments#action_12426909 ] Devaraj Das commented on HADOOP-181: I am open to a discussion on this but I thought that while we are discussing the possible solutions to the TaskTracker hea