[jira] [Updated] (YARN-230) Make changes for RM restart phase 1

[ https://issues.apache.org/jira/browse/YARN-230?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Bikas Saha updated YARN-230: Attachment: (was: YARN-230.3.patch) > Make changes for RM restart phase 1 > -

[jira] [Commented] (YARN-244) Application Master Retries fail due to FileNotFoundException

[ https://issues.apache.org/jira/browse/YARN-244?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13507877#comment-13507877 ] Bikas Saha commented on YARN-244: - Would be good to open a jira for the AM crashing. Why is

[jira] [Moved] (YARN-255) Support secure AM launch for unmanaged AM's

[ https://issues.apache.org/jira/browse/YARN-255?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Bikas Saha moved MAPREDUCE-4472 to YARN-255: Affects Version/s: (was: 3.0.0) 3.0.0

[jira] [Commented] (YARN-253) Container launch may fail if no files were localized

[ https://issues.apache.org/jira/browse/YARN-253?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13507882#comment-13507882 ] Bikas Saha commented on YARN-253: - I had posted an identical solution (the code has been ref

[jira] [Commented] (YARN-252) Unmanaged AMs should not have to set the ContainerLaunchContext

[ https://issues.apache.org/jira/browse/YARN-252?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13507883#comment-13507883 ] Bikas Saha commented on YARN-252: - I had initially done this for MAPREDUCE-4438 but later ch

[jira] [Commented] (YARN-257) NM should gracefully handle a full local disk

[ https://issues.apache.org/jira/browse/YARN-257?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13510812#comment-13510812 ] Bikas Saha commented on YARN-257: - Before the complete change, would it help if the NM did n

[jira] [Commented] (YARN-252) Unmanaged AMs should not have to set the ContainerLaunchContext

[ https://issues.apache.org/jira/browse/YARN-252?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13510887#comment-13510887 ] Bikas Saha commented on YARN-252: - I think it would be good to add the assert in the other p

[jira] [Commented] (YARN-128) Resurrect RM Restart

[ https://issues.apache.org/jira/browse/YARN-128?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13511501#comment-13511501 ] Bikas Saha commented on YARN-128: - Yes we need to. This is because many things like failure

[jira] [Updated] (YARN-103) Add a yarn AM - RM client module

[ https://issues.apache.org/jira/browse/YARN-103?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Bikas Saha updated YARN-103: Attachment: YARN-103.6.patch Attaching patch that fixes warnings where they make sense. > Ad

[jira] [Updated] (YARN-103) Add a yarn AM - RM client module

[ https://issues.apache.org/jira/browse/YARN-103?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Bikas Saha updated YARN-103: Attachment: YARN-103.7.patch Attaching patch that fixes the test failure. > Add a yarn AM -

[jira] [Updated] (YARN-230) Make changes for RM restart phase 1

[ https://issues.apache.org/jira/browse/YARN-230?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Bikas Saha updated YARN-230: Attachment: YARN-230.5.patch > Make changes for RM restart phase 1 > ---

[jira] [Updated] (YARN-230) Make changes for RM restart phase 1

[ https://issues.apache.org/jira/browse/YARN-230?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Bikas Saha updated YARN-230: Attachment: (was: YARN-230.5.patch) > Make changes for RM restart phase 1 > -

[jira] [Updated] (YARN-230) Make changes for RM restart phase 1

[ https://issues.apache.org/jira/browse/YARN-230?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Bikas Saha updated YARN-230: Attachment: YARN-230.5.patch > Make changes for RM restart phase 1 > ---

[jira] [Commented] (YARN-230) Make changes for RM restart phase 1

[ https://issues.apache.org/jira/browse/YARN-230?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13530439#comment-13530439 ] Bikas Saha commented on YARN-230: - Thanks for the reviews guys! @Tom Sorry for the delayed

[jira] [Commented] (YARN-252) Unmanaged AMs should not have to set the AM's ContainerLaunchContext

[ https://issues.apache.org/jira/browse/YARN-252?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13531245#comment-13531245 ] Bikas Saha commented on YARN-252: - bq. I'm not sure where you are suggesting such a check is

[jira] [Commented] (YARN-230) Make changes for RM restart phase 1

[ https://issues.apache.org/jira/browse/YARN-230?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13531255#comment-13531255 ] Bikas Saha commented on YARN-230: - I think the NullRMStateStore is a coding convenience as o

[jira] [Commented] (YARN-252) Unmanaged AMs should not have to set the AM's ContainerLaunchContext

[ https://issues.apache.org/jira/browse/YARN-252?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13532483#comment-13532483 ] Bikas Saha commented on YARN-252: - Patch looks good to me. Thats interesting. My initial im

[jira] [Commented] (YARN-230) Make changes for RM restart phase 1

[ https://issues.apache.org/jira/browse/YARN-230?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13532499#comment-13532499 ] Bikas Saha commented on YARN-230: - Yes. Will do when I refresh the patch on YARN-231 after Y

[jira] [Commented] (YARN-273) Add an unmanaged AM client for in-process AMs

[ https://issues.apache.org/jira/browse/YARN-273?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13532505#comment-13532505 ] Bikas Saha commented on YARN-273: - IMO, preferable to add this to the existing client. Will

[jira] [Commented] (YARN-230) Make changes for RM restart phase 1

[ https://issues.apache.org/jira/browse/YARN-230?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13532597#comment-13532597 ] Bikas Saha commented on YARN-230: - A note on testing. Among other test changes, there is a f

[jira] [Commented] (YARN-230) Make changes for RM restart phase 1

[ https://issues.apache.org/jira/browse/YARN-230?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13534098#comment-13534098 ] Bikas Saha commented on YARN-230: - Thanks guys! > Make changes for RM resta

[jira] [Commented] (YARN-223) Change processTree interface to work better with native code

[ https://issues.apache.org/jira/browse/YARN-223?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13534156#comment-13534156 ] Bikas Saha commented on YARN-223: - +1 for the code and approach. The patch changes some publ

[jira] [Commented] (YARN-103) Add a yarn AM - RM client module

[ https://issues.apache.org/jira/browse/YARN-103?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13535183#comment-13535183 ] Bikas Saha commented on YARN-103: - I went with the interface + impl approach because that se

[jira] [Commented] (YARN-103) Add a yarn AM - RM client module

[ https://issues.apache.org/jira/browse/YARN-103?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13536147#comment-13536147 ] Bikas Saha commented on YARN-103: - I think the patch follows the example you mention about i

[jira] [Commented] (YARN-103) Add a yarn AM - RM client module

[ https://issues.apache.org/jira/browse/YARN-103?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13536193#comment-13536193 ] Bikas Saha commented on YARN-103: - That would mean creating an empty concrete client that ex

[jira] [Created] (YARN-277) Use AMRMClient in DistributedShell to exemplify the approach

Bikas Saha created YARN-277: --- Summary: Use AMRMClient in DistributedShell to exemplify the approach Key: YARN-277 URL: https://issues.apache.org/jira/browse/YARN-277 Project: Hadoop YARN Issue Typ

[jira] [Commented] (YARN-286) Add a YARN ApplicationClassLoader

[ https://issues.apache.org/jira/browse/YARN-286?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13538350#comment-13538350 ] Bikas Saha commented on YARN-286: - Patch does not show any consumers for this functionality?

[jira] [Commented] (YARN-103) Add a yarn AM - RM client module

[ https://issues.apache.org/jira/browse/YARN-103?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13538364#comment-13538364 ] Bikas Saha commented on YARN-103: - Looks like there are different views on interface usage.

[jira] [Updated] (YARN-231) Add persistent store implementation for RMStateStore

[ https://issues.apache.org/jira/browse/YARN-231?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Bikas Saha updated YARN-231: Attachment: YARN-231.2.patch Attaching patch after rebasing on YARN-230. Addressing review comments and addi

[jira] [Commented] (YARN-231) Add persistent store implementation for RMStateStore

[ https://issues.apache.org/jira/browse/YARN-231?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13538442#comment-13538442 ] Bikas Saha commented on YARN-231: - {noformat} CodeWarning IS Inconsistent synchroni

[jira] [Commented] (YARN-289) Fair scheduler doesn't satisfy smaller request when bigger one is outstanding

[ https://issues.apache.org/jira/browse/YARN-289?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13539339#comment-13539339 ] Bikas Saha commented on YARN-289: - Could it be that 2048 request got "reserved" onto that no

[jira] [Commented] (YARN-2) Enhance CS to schedule accounting for both memory and cpu cores

[ https://issues.apache.org/jira/browse/YARN-2?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13539455#comment-13539455 ] Bikas Saha commented on YARN-2: --- Could you please elaborate on your proposal of virtual cores a

[jira] [Commented] (YARN-103) Add a yarn AM - RM client module

[ https://issues.apache.org/jira/browse/YARN-103?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13539462#comment-13539462 ] Bikas Saha commented on YARN-103: - bq. Why not use the existing ResourceRequest... I am tryi

[jira] [Updated] (YARN-103) Add a yarn AM - RM client module

[ https://issues.apache.org/jira/browse/YARN-103?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Bikas Saha updated YARN-103: Attachment: YARN-103.8.patch Attaching patch with review comments fixed. If this looks ok then I will break

[jira] [Commented] (YARN-2) Enhance CS to schedule accounting for both memory and cpu cores

[ https://issues.apache.org/jira/browse/YARN-2?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13540107#comment-13540107 ] Bikas Saha commented on YARN-2: --- I am not against shares if there are technical reasons going fo

[jira] [Updated] (YARN-103) Add a yarn AM - RM client module

[ https://issues.apache.org/jira/browse/YARN-103?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Bikas Saha updated YARN-103: Attachment: YARN-103.9.patch Clarified Sids question offline. Added a comment for that code that explains th

[jira] [Commented] (YARN-275) Make NodeManagers to NOT blindly heartbeat irrespective of whether previous heartbeat is processed or not.

[ https://issues.apache.org/jira/browse/YARN-275?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13540621#comment-13540621 ] Bikas Saha commented on YARN-275: - I briefly looked at the patch. The general approach seems

[jira] [Commented] (YARN-291) Dynamic resource configuration on NM

[ https://issues.apache.org/jira/browse/YARN-291?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13542259#comment-13542259 ] Bikas Saha commented on YARN-291: - Junping, it would be helpful if you attach the patches al

[jira] [Commented] (YARN-103) Add a yarn AM - RM client module

[ https://issues.apache.org/jira/browse/YARN-103?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13543303#comment-13543303 ] Bikas Saha commented on YARN-103: - Thanks guys! Appreciate the feedback. Will post a smarter

[jira] [Updated] (YARN-277) Use AMRMClient in DistributedShell to exemplify the approach

[ https://issues.apache.org/jira/browse/YARN-277?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Bikas Saha updated YARN-277: Attachment: YARN-277.1.patch Attaching patch that uses AMRMClient in DistShell > Use AMRMCli

[jira] [Commented] (YARN-277) Use AMRMClient in DistributedShell to exemplify the approach

[ https://issues.apache.org/jira/browse/YARN-277?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13543457#comment-13543457 ] Bikas Saha commented on YARN-277: - This is a refactoring of code in DistShell and so the exi

[jira] [Commented] (YARN-309) Make RM provide heartbeat interval to NM

[ https://issues.apache.org/jira/browse/YARN-309?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13543513#comment-13543513 ] Bikas Saha commented on YARN-309: - Thanks for the early patch. Overall approach looks ok. So

[jira] [Commented] (YARN-308) Improve documentation about what "asks" means in AMRMProtocol

[ https://issues.apache.org/jira/browse/YARN-308?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13544064#comment-13544064 ] Bikas Saha commented on YARN-308: - You would want to look at YARN-103 and related jiras that

[jira] [Commented] (YARN-314) Allow resource requests of different sizes at the same priority and locality

[ https://issues.apache.org/jira/browse/YARN-314?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13544068#comment-13544068 ] Bikas Saha commented on YARN-314: - To be clear, I think there is no issue in requesting reso

[jira] [Commented] (YARN-308) Improve documentation about what "asks" means in AMRMProtocol

[ https://issues.apache.org/jira/browse/YARN-308?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13544319#comment-13544319 ] Bikas Saha commented on YARN-308: - Sorry for not being clear. ResourceRequests are host loca

[jira] [Commented] (YARN-309) Make RM provide heartbeat interval to NM

[ https://issues.apache.org/jira/browse/YARN-309?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13547576#comment-13547576 ] Bikas Saha commented on YARN-309: - I see the config being really a minimum value and not the

[jira] [Updated] (YARN-231) Add persistent store implementation for RMStateStore

[ https://issues.apache.org/jira/browse/YARN-231?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Bikas Saha updated YARN-231: Attachment: YARN-231.3.FS.patch Attaching a patch containing only FS store. Will upload one for ZK once this

[jira] [Commented] (YARN-231) Add persistent store implementation for RMStateStore

[ https://issues.apache.org/jira/browse/YARN-231?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13559824#comment-13559824 ] Bikas Saha commented on YARN-231: - Hitesh, createConnection() is private and called from wit

[jira] [Commented] (YARN-231) Add persistent store implementation for RMStateStore

[ https://issues.apache.org/jira/browse/YARN-231?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13560042#comment-13560042 ] Bikas Saha commented on YARN-231: - There is a default value in yarn-defaults.xml {noformat}

[jira] [Commented] (YARN-316) YARN container launch may exceed maximum Windows command line length due to long classpath

[ https://issues.apache.org/jira/browse/YARN-316?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13561269#comment-13561269 ] Bikas Saha commented on YARN-316: - If its not too much work, could you try using http://com

[jira] [Commented] (YARN-316) YARN container launch may exceed maximum Windows command line length due to long classpath

[ https://issues.apache.org/jira/browse/YARN-316?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13561891#comment-13561891 ] Bikas Saha commented on YARN-316: - I am in favor of keeping a windows specific path for clas

[jira] [Updated] (YARN-353) Add Zookeeper-based store implementation for RMStateStore

[ https://issues.apache.org/jira/browse/YARN-353?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Bikas Saha updated YARN-353: Attachment: YARN-353.1.patch Attaching patch from YARN-231 after rebasing with FS changes. Some of the discu

[jira] [Commented] (YARN-316) YARN container launch may exceed maximum Windows command line length due to long classpath

[ https://issues.apache.org/jira/browse/YARN-316?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13562956#comment-13562956 ] Bikas Saha commented on YARN-316: - We could mention the jira number here {code} +// TODO

[jira] [Commented] (YARN-316) YARN container launch may exceed maximum Windows command line length due to long classpath

[ https://issues.apache.org/jira/browse/YARN-316?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13565824#comment-13565824 ] Bikas Saha commented on YARN-316: - +1. This ends up checking for existence of wildcard jars

[jira] [Commented] (YARN-275) Make NodeManagers to NOT blindly heartbeat irrespective of whether previous heartbeat is processed or not.

[ https://issues.apache.org/jira/browse/YARN-275?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13566289#comment-13566289 ] Bikas Saha commented on YARN-275: - I think there is some merit in making the change to have

[jira] [Commented] (YARN-275) Make NodeManagers to NOT blindly heartbeat irrespective of whether previous heartbeat is processed or not.

[ https://issues.apache.org/jira/browse/YARN-275?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13567364#comment-13567364 ] Bikas Saha commented on YARN-275: - To be clear, I did not mean we should implement dynamic h

[jira] [Commented] (YARN-275) Make NodeManagers to NOT blindly heartbeat irrespective of whether previous heartbeat is processed or not.

[ https://issues.apache.org/jira/browse/YARN-275?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13569455#comment-13569455 ] Bikas Saha commented on YARN-275: - Right. For the first step, the ResourceTrackerService (th

[jira] [Commented] (YARN-359) NodeManager container-related tests fail on branch-trunk-win

[ https://issues.apache.org/jira/browse/YARN-359?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13573116#comment-13573116 ] Bikas Saha commented on YARN-359: - The main reason for moving these to Shell was to reduce t

[jira] [Updated] (YARN-149) ZK-based High Availability (HA) for ResourceManager (RM)

[ https://issues.apache.org/jira/browse/YARN-149?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Bikas Saha updated YARN-149: Assignee: Bikas Saha > ZK-based High Availability (HA) for ResourceManager (RM) > ---

[jira] [Commented] (YARN-209) Capacity scheduler can leave application in pending state

[ https://issues.apache.org/jira/browse/YARN-209?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13575473#comment-13575473 ] Bikas Saha commented on YARN-209: - Right. Thats what the test in the attached patch does. Bu

[jira] [Commented] (YARN-359) NodeManager container-related tests fail on branch-trunk-win

[ https://issues.apache.org/jira/browse/YARN-359?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13575623#comment-13575623 ] Bikas Saha commented on YARN-359: - bq. potential argument for moving them back to Shell is

[jira] [Created] (YARN-392) Make it possible to schedule to specific nodes without dropping locality

Bikas Saha created YARN-392: --- Summary: Make it possible to schedule to specific nodes without dropping locality Key: YARN-392 URL: https://issues.apache.org/jira/browse/YARN-392 Project: Hadoop YARN

[jira] [Created] (YARN-393) * is overloaded for locality as well as resource demand

Bikas Saha created YARN-393: --- Summary: * is overloaded for locality as well as resource demand Key: YARN-393 URL: https://issues.apache.org/jira/browse/YARN-393 Project: Hadoop YARN Issue Type: Sub

[jira] [Updated] (YARN-393) In RM * is overloaded for locality as well as resource demand

[ https://issues.apache.org/jira/browse/YARN-393?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Bikas Saha updated YARN-393: Summary: In RM * is overloaded for locality as well as resource demand (was: * is overloaded for locality as

[jira] [Commented] (YARN-392) Make it possible to schedule to specific nodes without dropping locality

[ https://issues.apache.org/jira/browse/YARN-392?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13576299#comment-13576299 ] Bikas Saha commented on YARN-392: - Since you have already gone ahead and assigned it to your

[jira] [Created] (YARN-394) RM should be able to return requests that it cannot fulfill

Bikas Saha created YARN-394: --- Summary: RM should be able to return requests that it cannot fulfill Key: YARN-394 URL: https://issues.apache.org/jira/browse/YARN-394 Project: Hadoop YARN Issue Type

[jira] [Updated] (YARN-386) [Umbrella] YARN API cleanup and improvements

[ https://issues.apache.org/jira/browse/YARN-386?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Bikas Saha updated YARN-386: Summary: [Umbrella] YARN API cleanup and improvements (was: [Umbrella] YARN API cleanup) > [Umbrella] Y

[jira] [Created] (YARN-395) RM should have a way to disable scheduling to a set of nodes

Bikas Saha created YARN-395: --- Summary: RM should have a way to disable scheduling to a set of nodes Key: YARN-395 URL: https://issues.apache.org/jira/browse/YARN-395 Project: Hadoop YARN Issue Typ

[jira] [Commented] (YARN-395) RM should have a way to disable scheduling to a set of nodes

[ https://issues.apache.org/jira/browse/YARN-395?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13576302#comment-13576302 ] Bikas Saha commented on YARN-395: - This might as simple as specifying a resource request wit

[jira] [Created] (YARN-396) Rationalize AllocateResponse in RM scheduler API

Bikas Saha created YARN-396: --- Summary: Rationalize AllocateResponse in RM scheduler API Key: YARN-396 URL: https://issues.apache.org/jira/browse/YARN-396 Project: Hadoop YARN Issue Type: Sub-task

[jira] [Updated] (YARN-394) RM should be able to return requests that it cannot fulfill

[ https://issues.apache.org/jira/browse/YARN-394?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Bikas Saha updated YARN-394: Assignee: Bikas Saha > RM should be able to return requests that it cannot fulfill >

[jira] [Updated] (YARN-308) Improve documentation about what "asks" means in AMRMProtocol

[ https://issues.apache.org/jira/browse/YARN-308?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Bikas Saha updated YARN-308: Issue Type: Sub-task (was: Bug) Parent: YARN-386 > Improve documentation about what "asks" means

[jira] [Updated] (YARN-195) AM should have a way of telling RM to not allocate containers on some nodes

[ https://issues.apache.org/jira/browse/YARN-195?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Bikas Saha updated YARN-195: Issue Type: Sub-task (was: Bug) Parent: YARN-397 > AM should have a way of telling RM to not all

[jira] [Updated] (YARN-195) AM should have a way of telling RM to not allocate containers on some nodes

[ https://issues.apache.org/jira/browse/YARN-195?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Bikas Saha updated YARN-195: Assignee: Hitesh Shah > AM should have a way of telling RM to not allocate containers on some nodes > ---

[jira] [Reopened] (YARN-195) AM should have a way of telling RM to not allocate containers on some nodes

[ https://issues.apache.org/jira/browse/YARN-195?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Bikas Saha reopened YARN-195: - > AM should have a way of telling RM to not allocate containers on some nodes > ---

[jira] [Commented] (YARN-195) AM should have a way of telling RM to not allocate containers on some nodes

[ https://issues.apache.org/jira/browse/YARN-195?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13577802#comment-13577802 ] Bikas Saha commented on YARN-195: - Re-opening because this jira describes the problem to be

[jira] [Commented] (YARN-392) Make it possible to schedule to specific nodes without dropping locality

[ https://issues.apache.org/jira/browse/YARN-392?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13583713#comment-13583713 ] Bikas Saha commented on YARN-392: - >From what I understand this seems to be tangentially goi

[jira] [Created] (YARN-418) Libraries for YARN application developers

Bikas Saha created YARN-418: --- Summary: Libraries for YARN application developers Key: YARN-418 URL: https://issues.apache.org/jira/browse/YARN-418 Project: Hadoop YARN Issue Type: Improvement

[jira] [Updated] (YARN-29) Add a yarn-client module

[ https://issues.apache.org/jira/browse/YARN-29?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Bikas Saha updated YARN-29: --- Issue Type: Sub-task (was: Bug) Parent: YARN-418 > Add a yarn-client module > -

[jira] [Updated] (YARN-103) Add a yarn AM - RM client module

[ https://issues.apache.org/jira/browse/YARN-103?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Bikas Saha updated YARN-103: Issue Type: Sub-task (was: Improvement) Parent: YARN-418 > Add a yarn AM - RM client module > --

[jira] [Moved] (YARN-420) Enable the RM to work with AM's that are not managed by it

[ https://issues.apache.org/jira/browse/YARN-420?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Bikas Saha moved MAPREDUCE-4427 to YARN-420: Fix Version/s: (was: 2.0.2-alpha) 2.0.2-alpha

[jira] [Updated] (YARN-128) RM Restart

[ https://issues.apache.org/jira/browse/YARN-128?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Bikas Saha updated YARN-128: Summary: RM Restart (was: Resurrect RM Restart ) > RM Restart > --- > > Key: Y

[jira] [Moved] (YARN-419) Add client side for Unmanaged-AMs (MAPREDUCE-4427)

[ https://issues.apache.org/jira/browse/YARN-419?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Bikas Saha moved MAPREDUCE-4438 to YARN-419: Target Version/s: (was: ) Affects Version/s: (was: 3.0.0)

[jira] [Updated] (YARN-419) Add client side for Unmanaged-AMs

[ https://issues.apache.org/jira/browse/YARN-419?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Bikas Saha updated YARN-419: Summary: Add client side for Unmanaged-AMs (was: Add client side for Unmanaged-AMs (MAPREDUCE-4427)) >

[jira] [Created] (YARN-421) Create asynchronous AM RM Client

Bikas Saha created YARN-421: --- Summary: Create asynchronous AM RM Client Key: YARN-421 URL: https://issues.apache.org/jira/browse/YARN-421 Project: Hadoop YARN Issue Type: Sub-task Repor

[jira] [Created] (YARN-422) Add AM-NM client library

Bikas Saha created YARN-422: --- Summary: Add AM-NM client library Key: YARN-422 URL: https://issues.apache.org/jira/browse/YARN-422 Project: Hadoop YARN Issue Type: Sub-task Reporter: Bik

[jira] [Created] (YARN-423) Create a simple prioritized task assignment module

Bikas Saha created YARN-423: --- Summary: Create a simple prioritized task assignment module Key: YARN-423 URL: https://issues.apache.org/jira/browse/YARN-423 Project: Hadoop YARN Issue Type: Sub-task

[jira] [Commented] (YARN-392) Make it possible to schedule to specific nodes without dropping locality

[ https://issues.apache.org/jira/browse/YARN-392?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13585671#comment-13585671 ] Bikas Saha commented on YARN-392: - That would be a different JIRA, dont you think?

[jira] [Commented] (YARN-392) Make it possible to schedule to specific nodes without dropping locality

[ https://issues.apache.org/jira/browse/YARN-392?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13585677#comment-13585677 ] Bikas Saha commented on YARN-392: - @Sandy It would be per priority like I mentioned in my ex

[jira] [Commented] (YARN-417) Add a poller that allows the AM to receive notifications when it is assigned containers

[ https://issues.apache.org/jira/browse/YARN-417?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13586051#comment-13586051 ] Bikas Saha commented on YARN-417: - Sandy, it will be beneficial if you can mention if the pa

[jira] [Commented] (YARN-417) Add a poller that allows the AM to receive notifications when it is assigned containers

[ https://issues.apache.org/jira/browse/YARN-417?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13586092#comment-13586092 ] Bikas Saha commented on YARN-417: - My first comment would be to see if we can provide this f

[jira] [Commented] (YARN-424) Allow resource requests to provide hints to the scheduler

[ https://issues.apache.org/jira/browse/YARN-424?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13586560#comment-13586560 ] Bikas Saha commented on YARN-424: - I agree that being able to extend the API without breakin

[jira] [Commented] (YARN-417) Add a poller that allows the AM to receive notifications when it is assigned containers

[ https://issues.apache.org/jira/browse/YARN-417?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13587312#comment-13587312 ] Bikas Saha commented on YARN-417: - Yes. add/release should be used instead of allocate becau

[jira] [Updated] (YARN-417) Add a poller that allows the AM to receive notifications when it is assigned containers

[ https://issues.apache.org/jira/browse/YARN-417?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Bikas Saha updated YARN-417: Attachment: AMRMClientAsync.java Attaching an interface proposal > Add a poller that allows

[jira] [Created] (YARN-430) Add HDFS based store for RM

Bikas Saha created YARN-430: --- Summary: Add HDFS based store for RM Key: YARN-430 URL: https://issues.apache.org/jira/browse/YARN-430 Project: Hadoop YARN Issue Type: Sub-task Reporter:

[jira] [Commented] (YARN-365) Each NM heartbeat should not generate an event for the Scheduler

[ https://issues.apache.org/jira/browse/YARN-365?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13588818#comment-13588818 ] Bikas Saha commented on YARN-365: - Do we need to worry about there being overlap between the

[jira] [Created] (YARN-433) When RM is catching up with node updates then it should not expire acquired containers

Bikas Saha created YARN-433: --- Summary: When RM is catching up with node updates then it should not expire acquired containers Key: YARN-433 URL: https://issues.apache.org/jira/browse/YARN-433 Project: Hadoo

[jira] [Commented] (YARN-417) Add a poller that allows the AM to receive notifications when it is assigned containers

[ https://issues.apache.org/jira/browse/YARN-417?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13589296#comment-13589296 ] Bikas Saha commented on YARN-417: - I think if ContainerExitCodes needs to be added then it s

[jira] [Updated] (YARN-149) ResourceManager (RM) High-Availability (HA)

[ https://issues.apache.org/jira/browse/YARN-149?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Bikas Saha updated YARN-149: Summary: ResourceManager (RM) High-Availability (HA) (was: ZK-based High Availability (HA) for ResourceManag

[jira] [Updated] (YARN-443) allow OS scheduling priority of NM to be different than the containers it launches

[ https://issues.apache.org/jira/browse/YARN-443?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Bikas Saha updated YARN-443: Summary: allow OS scheduling priority of NM to be different than the containers it launches (was: allow OS s

<    2   3   4   5   6   7   8   9   10   11   >