Hey,
I'm looking for the place where the actual map input is transfered to
the 'map worker'/container (in case it's assigned to a host that doesn't
have the data already).
Something similar to the copyFromHost(..) method in Fetcher.java
(org.apache.hadoop.mapreduce.task.reduce), where the
Ashutosh Jindal created MAPREDUCE-5938:
--
Summary: Shuffle port in nodemanager is binding to all IPs
Key: MAPREDUCE-5938
URL: https://issues.apache.org/jira/browse/MAPREDUCE-5938
Project: Hadoop
I¹d be fine with 5 days. Any shorter and you could have issues with folks
being just out for a weekend and miss the entire vote unless we explicitly
tried to stay away from ending the vote on the weekend.
Tom
On 6/21/14, 12:54 AM, Arun C Murthy a...@hortonworks.com wrote:
Folks,
I'd like
See https://builds.apache.org/job/Hadoop-Mapreduce-trunk/1810/
###
## LAST 60 LINES OF THE CONSOLE
###
[...truncated 36203 lines...]
Tests run: 4, Failures: 0, Errors:
someone's filed a JIRA on loops in Hedged Read, with tests ...
https://issues.apache.org/jira/browse/HDFS-6591
On 23 June 2014 08:58, Mit Desai mitde...@yahoo-inc.com.invalid wrote:
+1 (non-binding)
Tested on: Fedora17
-Successful build from src (including native)
-Verified Signature
This seems reasonable, +1.
In case there is inactivity, we already extend dates even today, we can do the
same going forward.
I don't see major disagreements, time to start a vote?
Thanks
+Vinod
On Jun 20, 2014, at 10:54 PM, Arun C Murthy a...@hortonworks.com wrote:
Folks,
I'd like to
+1 here as well, let's do a vote thread (for 7 days, maybe for the last
time!)
On Mon, Jun 23, 2014 at 11:46 AM, Vinod Kumar Vavilapalli
vino...@apache.org wrote:
This seems reasonable, +1.
In case there is inactivity, we already extend dates even today, we can do
the same going forward.
+1, sounds good!
On Mon, Jun 23, 2014 at 12:02 PM, Andrew Wang andrew.w...@cloudera.com
wrote:
+1 here as well, let's do a vote thread (for 7 days, maybe for the last
time!)
On Mon, Jun 23, 2014 at 11:46 AM, Vinod Kumar Vavilapalli
vino...@apache.org wrote:
This seems reasonable, +1.
Folks,
I'd appreciate some help here. Due to family reasons (all good ones), I'll be
away for a couple of weeks. Can someone else pick up the RM duties for
hadoop-2.5? Maybe Andrew since he's expressed interest in the past? I will pick
up the thread again for hadoop-2.6, but I don't want to
Kihwal Lee created MAPREDUCE-5939:
-
Summary: StartTime showing up as the epoch time in JHS UI after
upgrade
Key: MAPREDUCE-5939
URL: https://issues.apache.org/jira/browse/MAPREDUCE-5939
Project:
On Jun 21, 2014, at 8:01 AM, Andrew Wang andrew.w...@cloudera.com wrote:
This is why I'd like to keep my original proposal on the table: keep going
with branch-2 in the near term, while working towards a JDK8-based Hadoop 3
by April next year. It doesn't need to be a big bang release either.
Andrew, correct me if I'm misunderstanding, but the incompatible change
that would require a major version bump is dropping support for JDK6.
On Mon, Jun 23, 2014 at 1:53 PM, sanjay Radia san...@hortonworks.com
wrote:
On Jun 21, 2014, at 8:01 AM, Andrew Wang andrew.w...@cloudera.com wrote:
I can pick up the RM duties for 2.5. If I run into any HDFS doubts, I might
need some help from someone more familiar with HDFS.
On Mon, Jun 23, 2014 at 12:07 PM, Arun C Murthy a...@hortonworks.com wrote:
Folks,
I'd appreciate some help here. Due to family reasons (all good ones),
I'll be
It would be great if we can consider the shared cache (YARN-1492) as part
of 2.5. It obviously will depend on the reviews from folks, but the feature
itself should be reasonably solid.
Thanks,
Sangjin
On Mon, Jun 23, 2014 at 2:24 PM, Arun C Murthy a...@hortonworks.com wrote:
Thanks Karthik!
Thanks for volunteering Karthik, I can help out with anything on the HDFS
side if you need it.
Best,
Andrew
On Mon, Jun 23, 2014 at 2:30 PM, Sangjin Lee sjl...@gmail.com wrote:
It would be great if we can consider the shared cache (YARN-1492) as part
of 2.5. It obviously will depend on the
While YARN-1492 would greatly benefit performance, I think it might be a
little too ambitious at this point to include it in 2.5 without delaying
it.
As I mentioned before and from my offline conversation with Arun, we might
want to release 2.6 end of July and that might be a better vehicle for
Just found the roadmap includes a wrong jira link for RM Webservices
YARN-1731, it should be YARN-1695
https://issues.apache.org/jira/browse/YARN-1695.
Jian
On Mon, Jun 23, 2014 at 2:51 PM, Karthik Kambatla ka...@cloudera.com
wrote:
While YARN-1492 would greatly benefit performance, I think
Thanks everyone. I'll start a vote tmrw if there are no objections.
Arun
On Mon, Jun 23, 2014 at 12:06 PM, Jitendra Pandey jiten...@hortonworks.com
wrote:
+1, sounds good!
On Mon, Jun 23, 2014 at 12:02 PM, Andrew Wang andrew.w...@cloudera.com
wrote:
+1 here as well, let's do a vote
IMHO, I'd like to +1 to include YARN-1492 and work-preserving restart
in 2.6. I believe they are good improvement, but they are
work-in-progress and it takes a bit time to be done.
Thanks,
- Tsuyoshi
On Tue, Jun 24, 2014 at 7:13 AM, Jian He j...@hortonworks.com wrote:
Just found the roadmap
I use hadoop 2.2.0 under Redhat Linux,I run hadoop mapreduce examples like
follows:
[yarn@localhost sbin]$ hadoop jar
/opt/yarn/hadoop-2.2.0/share/hadoop/mapreduce/hadoop-mapreduce-examples-2.2.0.jar
pi
-Dmapreduce.clientfactory.class.name=org.apache.hadoop.mapred.YarnClientFactory
-libjars
Extended attributes (HDFS-2006) is already merged to branch-2,
so I'll update the Roadmap wiki to change it from 2.6 to 2.5.
Please correct me if I am wrong.
Thanks,
Akira
(2014/06/23 16:28), Tsuyoshi OZAWA wrote:
IMHO, I'd like to +1 to include YARN-1492 and work-preserving restart
in 2.6. I
+1 (non-binding)
built it from source and ram some MR examples successfully.
On Mon, Jun 23, 2014 at 11:59 PM, Mit Desai mitde...@yahoo-inc.com.invalid
wrote:
+1 (non-binding)
Tested on: Fedora17
-Successful build from src
-Verified Signature
-Deployed source to my single node cluster
Tx Jian, updated the wiki.
+Vinod
On Jun 23, 2014, at 3:13 PM, Jian He j...@hortonworks.com wrote:
Just found the roadmap includes a wrong jira link for RM Webservices
YARN-1731, it should be YARN-1695
https://issues.apache.org/jira/browse/YARN-1695.
Jian
On Mon, Jun 23, 2014 at 2:51
I started reviewing YARN-1492 and synced up offline with Sangjin and Chris who
are leading this. I proposed that we should get the reviews and commits going
as they go and then mark the feature alpha, beta or stable depending on how far
we are ahead with the feature and how it aligns with the
Hey all,
This one started as an innocuous thread of enabling JDK7 on trunk and now it
seems like (haven't still finished reading the entire thing, and I started a
while ago) it has become a full blown proposal on 2.x, 3.x and 4.x releases.
Some of us haven't been tracking this (at least me and
25 matches
Mail list logo