Re: Picking up Giraph 1.1.0 release

2014-06-09 Thread Roman Shaposhnik
Hi!

good news everyone! We've got just one JIRA left
that is assigned to be fixed in 1.1:
https://issues.apache.org/jira/browse/GIRAPH-747
Everything else is either resolved or just tracks
the release activities.

Based on this, I have create a release-1.1 branch
and will proceed with Bigtop-based testing for the
next week. Based on the result of system testing
hopefully we would be able to release in a week or so.

For the next couple of weeks please be extra
careful marking JIRAs with 1.1 to get things
into the release.

I plan to announce our first RC mid-week. It would
be really nice if folks can jump on it and provide
some additional testing.

Finally, if somebody can look at GIRAPH-747 that
also would be very much appreciated!

Thanks,
Roman.


[jira] [Commented] (GIRAPH-886) Partition Balancing does not work

2014-06-09 Thread Lukas Nalezenec (JIRA)

[ 
https://issues.apache.org/jira/browse/GIRAPH-886?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14021757#comment-14021757
 ] 

Lukas Nalezenec commented on GIRAPH-886:


Hi Pavan, 
Thanks for your review. I am really busy this month. I wanted to refactor the 
first patch but it i havent finished it yet. I will make some simplified 
version.

 Partition Balancing does not work
 -

 Key: GIRAPH-886
 URL: https://issues.apache.org/jira/browse/GIRAPH-886
 Project: Giraph
  Issue Type: Bug
  Components: graph
Affects Versions: 1.1.0
Reporter: Lukas Nalezenec
  Labels: patch
 Attachments: GIRAPH-886.patch


 PartitionBalancer returns PartitionOwners in list in arbitrary order but both 
 Range and Hash partitioners are assuming that list is sorted by partitionId.



--
This message was sent by Atlassian JIRA
(v6.2#6252)


[jira] [Created] (GIRAPH-914) giraph-core compiled with cdh4.1.2 profile produces 50 MB fat jar.

2014-06-09 Thread Lukas Nalezenec (JIRA)
Lukas Nalezenec created GIRAPH-914:
--

 Summary: giraph-core compiled with cdh4.1.2 profile produces 50 MB 
fat jar.
 Key: GIRAPH-914
 URL: https://issues.apache.org/jira/browse/GIRAPH-914
 Project: Giraph
  Issue Type: Bug
  Components: build
Affects Versions: 1.1.0
Reporter: Lukas Nalezenec
Priority: Critical


When you compile giraph-core with cdh4.1.2 it has got 50 MB. Its too much. It 
has got lot of useless dependencies. I think that we can save a lot setting 
scope to provided. 
We could also make jython dependencies optional - jython dependencies has got 8 
MB.




--
This message was sent by Atlassian JIRA
(v6.2#6252)


[jira] [Updated] (GIRAPH-914) giraph-core compiled with cdh4.1.2 profile produces 50 MB fat jar.

2014-06-09 Thread Lukas Nalezenec (JIRA)

 [ 
https://issues.apache.org/jira/browse/GIRAPH-914?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Lukas Nalezenec updated GIRAPH-914:
---

Attachment: effective.pom.xml
dependency.tree

 giraph-core compiled with cdh4.1.2 profile produces 50 MB fat jar.
 --

 Key: GIRAPH-914
 URL: https://issues.apache.org/jira/browse/GIRAPH-914
 Project: Giraph
  Issue Type: Bug
  Components: build
Affects Versions: 1.1.0
Reporter: Lukas Nalezenec
Priority: Critical
 Attachments: dependency.tree, effective.pom.xml


 When you compile giraph-core with cdh4.1.2 it has got 50 MB. Its too much. It 
 has got lot of useless dependencies. I think that we can save a lot setting 
 scope to provided. 
 We could also make jython dependencies optional - jython dependencies has got 
 8 MB.



--
This message was sent by Atlassian JIRA
(v6.2#6252)


[jira] [Updated] (GIRAPH-914) giraph-core compiled with cdh4.1.2 profile produces 50 MB fat jar.

2014-06-09 Thread Lukas Nalezenec (JIRA)

 [ 
https://issues.apache.org/jira/browse/GIRAPH-914?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Lukas Nalezenec updated GIRAPH-914:
---

Priority: Major  (was: Critical)

 giraph-core compiled with cdh4.1.2 profile produces 50 MB fat jar.
 --

 Key: GIRAPH-914
 URL: https://issues.apache.org/jira/browse/GIRAPH-914
 Project: Giraph
  Issue Type: Bug
  Components: build
Affects Versions: 1.1.0
Reporter: Lukas Nalezenec
 Attachments: dependency.tree, effective.pom.xml


 When you compile giraph-core with cdh4.1.2 it has got 50 MB. Its too much. It 
 has got lot of useless dependencies. I think that we can save a lot setting 
 scope to provided. 
 We could also make jython dependencies optional - jython dependencies has got 
 8 MB.



--
This message was sent by Atlassian JIRA
(v6.2#6252)


Re: Picking up Giraph 1.1.0 release

2014-06-09 Thread Claudio Martella
Actually, I was going to commit 874. Shall I commit it to release-1.1?


On Mon, Jun 9, 2014 at 8:30 AM, Roman Shaposhnik ro...@shaposhnik.org
wrote:

 Hi!

 good news everyone! We've got just one JIRA left
 that is assigned to be fixed in 1.1:
 https://issues.apache.org/jira/browse/GIRAPH-747
 Everything else is either resolved or just tracks
 the release activities.

 Based on this, I have create a release-1.1 branch
 and will proceed with Bigtop-based testing for the
 next week. Based on the result of system testing
 hopefully we would be able to release in a week or so.

 For the next couple of weeks please be extra
 careful marking JIRAs with 1.1 to get things
 into the release.

 I plan to announce our first RC mid-week. It would
 be really nice if folks can jump on it and provide
 some additional testing.

 Finally, if somebody can look at GIRAPH-747 that
 also would be very much appreciated!

 Thanks,
 Roman.




-- 
   Claudio Martella


Re: Picking up Giraph 1.1.0 release

2014-06-09 Thread Roman Shaposhnik
On Mon, Jun 9, 2014 at 2:14 PM, Claudio Martella
claudio.marte...@gmail.com wrote:
 Actually, I was going to commit 874. Shall I commit it to release-1.1?

Just keep committing to trunk (the usual way), but every time
you think the JIRA should also go into 1.1 -- please ping me.

Thanks,
Roman.


Re: Review Request 21987: Detect crashes of Netty threads

2014-06-09 Thread Sergey Edunov


 On May 29, 2014, 6:40 p.m., Pavan Kumar Athivarapu wrote:
  findbugs-exclude.xml, line 46
  https://reviews.apache.org/r/21987/diff/1/?file=597788#file597788line46
 
  why are u removing these stuff,
  wouldn't it be fine to just add GraphTaskManager to allow exit

I just removed unused code from GiraphYarnTask and hence it doesn't have 
System.exit() anymore and that made this check obsolete.


 On May 29, 2014, 6:40 p.m., Pavan Kumar Athivarapu wrote:
  giraph-core/src/main/java/org/apache/giraph/comm/netty/handler/ExceptionHandler.java,
   line 29
  https://reviews.apache.org/r/21987/diff/1/?file=597792#file597792line29
 
  this catches exceptions only for inbound conenctions, do u want to do 
  this for outbound as well.
  
  in that case would it not be better to just piggy back on the counters 
   override exceptioncaught in those methods?
  
  I am fine with creating new classes or just overriding the method in 
  counters
 
 Pavan Kumar Athivarapu wrote:
 one more thing (for exception handling) - inbound handler should be at 
 the end and outbound handler should be at the beginning of the pipeline 
 so u catch exceptions at the end of processing input - 1
 u catch exceptions at the end of sending output - 2

I felt like it's nice to have a separate class for exception handling as it 
outlines it's intention and makes it clear how and when to use it.


- Sergey


---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/21987/#review44270
---


On June 5, 2014, 10:26 p.m., Sergey Edunov wrote:
 
 ---
 This is an automatically generated e-mail. To reply, visit:
 https://reviews.apache.org/r/21987/
 ---
 
 (Updated June 5, 2014, 10:26 p.m.)
 
 
 Review request for giraph.
 
 
 Repository: giraph-git
 
 
 Description
 ---
 
 When some of the request processing threads fails, the worker gets stuck but 
 the job doesn't fail and it has to be killed manually. We should detect netty 
 thread crashes and fail the job automatically.
 
 
 Diffs
 -
 
   findbugs-exclude.xml e0466f7 
   giraph-core/src/main/java/org/apache/giraph/comm/netty/NettyClient.java 
 ae40c3b 
   
 giraph-core/src/main/java/org/apache/giraph/comm/netty/NettyMasterClient.java 
 c982209 
   
 giraph-core/src/main/java/org/apache/giraph/comm/netty/NettyMasterServer.java 
 cb36c3e 
   giraph-core/src/main/java/org/apache/giraph/comm/netty/NettyServer.java 
 14d4ea8 
   
 giraph-core/src/main/java/org/apache/giraph/comm/netty/NettyWorkerClient.java 
 7541418 
   
 giraph-core/src/main/java/org/apache/giraph/comm/netty/NettyWorkerServer.java 
 adb96cb 
   
 giraph-core/src/main/java/org/apache/giraph/comm/netty/handler/ExceptionHandler.java
  PRE-CREATION 
   
 giraph-core/src/main/java/org/apache/giraph/comm/netty/handler/RequestServerHandler.java
  601cd2f 
   giraph-core/src/main/java/org/apache/giraph/graph/GraphMapper.java c86a024 
   giraph-core/src/main/java/org/apache/giraph/graph/GraphTaskManager.java 
 ad5fc91 
   giraph-core/src/main/java/org/apache/giraph/master/BspServiceMaster.java 
 90dc9f3 
   giraph-core/src/main/java/org/apache/giraph/worker/BspServiceWorker.java 
 aff7084 
   giraph-core/src/main/java/org/apache/giraph/yarn/GiraphYarnTask.java 
 f4719cc 
   giraph-core/src/test/java/org/apache/giraph/comm/ConnectionTest.java 
 e771e36 
   giraph-core/src/test/java/org/apache/giraph/comm/MockExceptionHandler.java 
 PRE-CREATION 
   giraph-core/src/test/java/org/apache/giraph/comm/RequestFailureTest.java 
 236bc88 
   giraph-core/src/test/java/org/apache/giraph/comm/RequestTest.java fcdfa5c 
   giraph-core/src/test/java/org/apache/giraph/comm/SaslConnectionTest.java 
 c026cf8 
 
 Diff: https://reviews.apache.org/r/21987/diff/
 
 
 Testing
 ---
 
 Run some production jobs with this change. 
 Also introduced random bugs in deserialization logic and confirmed that job 
 fails. 
 
 
 Thanks,
 
 Sergey Edunov
 




Re: Review Request 21987: Detect crashes of Netty threads

2014-06-09 Thread Pavan Kumar Athivarapu


 On May 29, 2014, 6:40 p.m., Pavan Kumar Athivarapu wrote:
  giraph-core/src/main/java/org/apache/giraph/comm/netty/handler/ExceptionHandler.java,
   line 29
  https://reviews.apache.org/r/21987/diff/1/?file=597792#file597792line29
 
  this catches exceptions only for inbound conenctions, do u want to do 
  this for outbound as well.
  
  in that case would it not be better to just piggy back on the counters 
   override exceptioncaught in those methods?
  
  I am fine with creating new classes or just overriding the method in 
  counters
 
 Pavan Kumar Athivarapu wrote:
 one more thing (for exception handling) - inbound handler should be at 
 the end and outbound handler should be at the beginning of the pipeline 
 so u catch exceptions at the end of processing input - 1
 u catch exceptions at the end of sending output - 2
 
 Sergey Edunov wrote:
 I felt like it's nice to have a separate class for exception handling as 
 it outlines it's intention and makes it clear how and when to use it.

even the new diff does not handle exceptions for outbound
also please name is the handler to something like serverexceptionHandler, 
etc. instead of just exception


- Pavan Kumar


---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/21987/#review44270
---


On June 5, 2014, 10:26 p.m., Sergey Edunov wrote:
 
 ---
 This is an automatically generated e-mail. To reply, visit:
 https://reviews.apache.org/r/21987/
 ---
 
 (Updated June 5, 2014, 10:26 p.m.)
 
 
 Review request for giraph.
 
 
 Repository: giraph-git
 
 
 Description
 ---
 
 When some of the request processing threads fails, the worker gets stuck but 
 the job doesn't fail and it has to be killed manually. We should detect netty 
 thread crashes and fail the job automatically.
 
 
 Diffs
 -
 
   findbugs-exclude.xml e0466f7 
   giraph-core/src/main/java/org/apache/giraph/comm/netty/NettyClient.java 
 ae40c3b 
   
 giraph-core/src/main/java/org/apache/giraph/comm/netty/NettyMasterClient.java 
 c982209 
   
 giraph-core/src/main/java/org/apache/giraph/comm/netty/NettyMasterServer.java 
 cb36c3e 
   giraph-core/src/main/java/org/apache/giraph/comm/netty/NettyServer.java 
 14d4ea8 
   
 giraph-core/src/main/java/org/apache/giraph/comm/netty/NettyWorkerClient.java 
 7541418 
   
 giraph-core/src/main/java/org/apache/giraph/comm/netty/NettyWorkerServer.java 
 adb96cb 
   
 giraph-core/src/main/java/org/apache/giraph/comm/netty/handler/ExceptionHandler.java
  PRE-CREATION 
   
 giraph-core/src/main/java/org/apache/giraph/comm/netty/handler/RequestServerHandler.java
  601cd2f 
   giraph-core/src/main/java/org/apache/giraph/graph/GraphMapper.java c86a024 
   giraph-core/src/main/java/org/apache/giraph/graph/GraphTaskManager.java 
 ad5fc91 
   giraph-core/src/main/java/org/apache/giraph/master/BspServiceMaster.java 
 90dc9f3 
   giraph-core/src/main/java/org/apache/giraph/worker/BspServiceWorker.java 
 aff7084 
   giraph-core/src/main/java/org/apache/giraph/yarn/GiraphYarnTask.java 
 f4719cc 
   giraph-core/src/test/java/org/apache/giraph/comm/ConnectionTest.java 
 e771e36 
   giraph-core/src/test/java/org/apache/giraph/comm/MockExceptionHandler.java 
 PRE-CREATION 
   giraph-core/src/test/java/org/apache/giraph/comm/RequestFailureTest.java 
 236bc88 
   giraph-core/src/test/java/org/apache/giraph/comm/RequestTest.java fcdfa5c 
   giraph-core/src/test/java/org/apache/giraph/comm/SaslConnectionTest.java 
 c026cf8 
 
 Diff: https://reviews.apache.org/r/21987/diff/
 
 
 Testing
 ---
 
 Run some production jobs with this change. 
 Also introduced random bugs in deserialization logic and confirmed that job 
 fails. 
 
 
 Thanks,
 
 Sergey Edunov
 




[jira] [Commented] (GIRAPH-842) option to dump histogram of memory usage when heap is low on memory

2014-06-09 Thread Sergey Edunov (JIRA)

[ 
https://issues.apache.org/jira/browse/GIRAPH-842?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14025989#comment-14025989
 ] 

Sergey Edunov commented on GIRAPH-842:
--

Hey Pavan, this is great feature, here are some comments on implementation:

1) Please assign your thread some meaningful name, it can be helpful in 
debugging stack traces.
2) Stop flag has to be volatile, otherwise your thread may never see the change
3) I would let the thread die if you get InterruptedException.
4) runtime.freeMemory() might give you false alarms if you run job with 
different values of Xmx and Xms. This is because freeMemory only counts free 
bytes currently allocated to JVM. If your Xmx setting is bigger than Xms you 
can go low on freeMemory before JVM allocates another chunk. To get more 
accurate results you can do (maxMemory - totalMemory + freeMemory) 

 option to dump histogram of memory usage when heap is low on memory
 ---

 Key: GIRAPH-842
 URL: https://issues.apache.org/jira/browse/GIRAPH-842
 Project: Giraph
  Issue Type: Bug
Reporter: Pavan Kumar
Assignee: Pavan Kumar
Priority: Minor
 Attachments: GIRAPH-842.patch, master-stderr, worker-stderr


 Currently we are left in blind for jobs that OOM, it would be helpful if we 
 can do a jmap -histo dump when heap has very little free space left.



--
This message was sent by Atlassian JIRA
(v6.2#6252)