You are viewing a plain text version of this content. The canonical link for it is here.
Posted to user@tez.apache.org by Chris K Wensel <ch...@wensel.net> on 2014/04/04 22:38:57 UTC
standalone mode
sorry if i'm missing something obvious, but is there a "standalone" mode I can run tez on within my IDE. MR has this, and its quite useful.
current'y when I apply similar voodoo, I get a hang on TezClient#createApplication
to be clear, i'm not using mini dfs/tez cluster, but want to simply use LocalFileSystem and have the ability to step through my code.
logs look like this below.
2014-04-04 09:15:41,653 INFO [Thread-3] client.RMProxy (RMProxy.java:createRMProxy(56)) - Connecting to ResourceManager at /0.0.0.0:8032
2014-04-04 09:15:42,767 INFO [Thread-3] ipc.Client (Client.java:handleConnectionFailure(783)) - Retrying connect to server: 0.0.0.0/0.0.0.0:8032. Already tried 0 time(s); retry policy is RetryUpToMaximumCountWithFixedSleep(maxRetries=10, sleepTime=1 SECONDS)
2014-04-04 09:15:43,770 INFO [Thread-3] ipc.Client (Client.java:handleConnectionFailure(783)) - Retrying connect to server: 0.0.0.0/0.0.0.0:8032. Already tried 1 time(s); retry policy is RetryUpToMaximumCountWithFixedSleep(maxRetries=10, sleepTime=1 SECONDS)
2014-04-04 09:15:44,773 INFO [Thread-3] ipc.Client (Client.java:handleConnectionFailure(783)) - Retrying connect to server: 0.0.0.0/0.0.0.0:8032. Already tried 2 time(s); retry policy is RetryUpToMaximumCountWithFixedSleep(maxRetries=10, sleepTime=1 SECONDS)
2014-04-04 09:15:45,775 INFO [Thread-3] ipc.Client (Client.java:handleConnectionFailure(783)) - Retrying connect to server: 0.0.0.0/0.0.0.0:8032. Already tried 3 time(s); retry policy is RetryUpToMaximumCountWithFixedSleep(maxRetries=10, sleepTime=1 SECONDS)
2014-04-04 09:15:46,776 INFO [Thread-3] ipc.Client (Client.java:handleConnectionFailure(783)) - Retrying connect to server: 0.0.0.0/0.0.0.0:8032. Already tried 4 time(s); retry policy is RetryUpToMaximumCountWithFixedSleep(maxRetries=10, sleepTime=1 SECONDS)
2014-04-04 09:15:47,779 INFO [Thread-3] ipc.Client (Client.java:handleConnectionFailure(783)) - Retrying connect to server: 0.0.0.0/0.0.0.0:8032. Already tried 5 time(s); retry policy is RetryUpToMaximumCountWithFixedSleep(maxRetries=10, sleepTime=1 SECONDS)
ckw
--
Chris K Wensel
chris@concurrentinc.com
http://concurrentinc.com
RE: standalone mode
Posted by Bikas Saha <bi...@hortonworks.com>.
I agree and we are all facing the same issue in debuggability.
The MR local runner is more like a simulation of MR than an actual local
execution of the same (app master) code that runs in the cluster. This is
possible to do in MR since we can essentially simulate the M stage followed
by the R stage. In addition to that the MR local mode has some interesting
hacks to do the intermediate data transfer (which normally happens via the
shuffle service in the cluster).
For Tez the above won't work because we don't have a pre-determined DAG to
run. So we have to run the same app master code locally. We also want to do
a better job of abstracting out the intermediate data transfer since that
part is pluggable in Tez (as opposed to being more or less built into MR).
For these reasons, the local mode effort is large and taking time. I know
this does not solve our problems but hopefully clarifies why its isn't
there yet.
One thing we could potentially do is see if TezMiniCluster can run the Tez
AM in unmanaged mode. This would still needs to run a pseudo-distributed
cluster (YARN+HDFS) and would be slow. But the Tez AM would run inline in
the test thread and thus could be debuggable within the IDE.
Bikas
*From:* Chris K Wensel [mailto:chris@wensel.net]
*Sent:* Friday, April 04, 2014 2:23 PM
*To:* user@tez.incubator.apache.org
*Subject:* Re: standalone mode
thanks. I have the MiniTezCluster working, but its not as friendly.
On Apr 4, 2014, at 1:48 PM, Bikas Saha <bi...@hortonworks.com> wrote:
Chris,
That is work in progress tracked at
https://issues.apache.org/jira/browse/TEZ-684. Folks from Yahoo are driving
this.
In the interim, Tez/Hive/Pig are using MiniTezCluster to write unit tests.
This runs a pseudo-distributed cluster within your unit test.
Bikas
*From:* Chris K Wensel [mailto:chris@wensel.net]
*Sent:* Friday, April 04, 2014 1:39 PM
*To:* user@tez.incubator.apache.org
*Subject:* standalone mode
sorry if i'm missing something obvious, but is there a "standalone" mode I
can run tez on within my IDE. MR has this, and its quite useful.
current'y when I apply similar voodoo, I get a hang on
TezClient#createApplication
to be clear, i'm not using mini dfs/tez cluster, but want to simply use
LocalFileSystem and have the ability to step through my code.
logs look like this below.
2014-04-04 09:15:41,653 INFO [Thread-3] client.RMProxy
(RMProxy.java:createRMProxy(56)) - Connecting to ResourceManager at /
0.0.0.0:8032
2014-04-04 09:15:42,767 INFO [Thread-3] ipc.Client
(Client.java:handleConnectionFailure(783)) - Retrying connect to server:
0.0.0.0/0.0.0.0:8032. Already tried 0 time(s); retry policy is
RetryUpToMaximumCountWithFixedSleep(maxRetries=10, sleepTime=1 SECONDS)
2014-04-04 09:15:43,770 INFO [Thread-3] ipc.Client
(Client.java:handleConnectionFailure(783)) - Retrying connect to server:
0.0.0.0/0.0.0.0:8032. Already tried 1 time(s); retry policy is
RetryUpToMaximumCountWithFixedSleep(maxRetries=10, sleepTime=1 SECONDS)
2014-04-04 09:15:44,773 INFO [Thread-3] ipc.Client
(Client.java:handleConnectionFailure(783)) - Retrying connect to server:
0.0.0.0/0.0.0.0:8032. Already tried 2 time(s); retry policy is
RetryUpToMaximumCountWithFixedSleep(maxRetries=10, sleepTime=1 SECONDS)
2014-04-04 09:15:45,775 INFO [Thread-3] ipc.Client
(Client.java:handleConnectionFailure(783)) - Retrying connect to server:
0.0.0.0/0.0.0.0:8032. Already tried 3 time(s); retry policy is
RetryUpToMaximumCountWithFixedSleep(maxRetries=10, sleepTime=1 SECONDS)
2014-04-04 09:15:46,776 INFO [Thread-3] ipc.Client
(Client.java:handleConnectionFailure(783)) - Retrying connect to server:
0.0.0.0/0.0.0.0:8032. Already tried 4 time(s); retry policy is
RetryUpToMaximumCountWithFixedSleep(maxRetries=10, sleepTime=1 SECONDS)
2014-04-04 09:15:47,779 INFO [Thread-3] ipc.Client
(Client.java:handleConnectionFailure(783)) - Retrying connect to server:
0.0.0.0/0.0.0.0:8032. Already tried 5 time(s); retry policy is
RetryUpToMaximumCountWithFixedSleep(maxRetries=10, sleepTime=1 SECONDS)
ckw
--
Chris K Wensel
chris@concurrentinc.com
http://concurrentinc.com
CONFIDENTIALITY NOTICE
NOTICE: This message is intended for the use of the individual or entity to
which it is addressed and may contain information that is confidential,
privileged and exempt from disclosure under applicable law. If the reader
of this message is not the intended recipient, you are hereby notified that
any printing, copying, dissemination, distribution, disclosure or
forwarding of this communication is strictly prohibited. If you have
received this communication in error, please contact the sender immediately
and delete it from your system. Thank You.
--
Chris K Wensel
chris@concurrentinc.com
http://concurrentinc.com
--
CONFIDENTIALITY NOTICE
NOTICE: This message is intended for the use of the individual or entity to
which it is addressed and may contain information that is confidential,
privileged and exempt from disclosure under applicable law. If the reader
of this message is not the intended recipient, you are hereby notified that
any printing, copying, dissemination, distribution, disclosure or
forwarding of this communication is strictly prohibited. If you have
received this communication in error, please contact the sender immediately
and delete it from your system. Thank You.
Re: standalone mode
Posted by Chris K Wensel <ch...@wensel.net>.
thanks. I have the MiniTezCluster working, but its not as friendly.
On Apr 4, 2014, at 1:48 PM, Bikas Saha <bi...@hortonworks.com> wrote:
> Chris,
>
> That is work in progress tracked at https://issues.apache.org/jira/browse/TEZ-684. Folks from Yahoo are driving this.
>
> In the interim, Tez/Hive/Pig are using MiniTezCluster to write unit tests. This runs a pseudo-distributed cluster within your unit test.
>
> Bikas
>
> From: Chris K Wensel [mailto:chris@wensel.net]
> Sent: Friday, April 04, 2014 1:39 PM
> To: user@tez.incubator.apache.org
> Subject: standalone mode
>
> sorry if i'm missing something obvious, but is there a "standalone" mode I can run tez on within my IDE. MR has this, and its quite useful.
>
> current'y when I apply similar voodoo, I get a hang on TezClient#createApplication
>
> to be clear, i'm not using mini dfs/tez cluster, but want to simply use LocalFileSystem and have the ability to step through my code.
>
> logs look like this below.
>
>
> 2014-04-04 09:15:41,653 INFO [Thread-3] client.RMProxy (RMProxy.java:createRMProxy(56)) - Connecting to ResourceManager at /0.0.0.0:8032
> 2014-04-04 09:15:42,767 INFO [Thread-3] ipc.Client (Client.java:handleConnectionFailure(783)) - Retrying connect to server: 0.0.0.0/0.0.0.0:8032. Already tried 0 time(s); retry policy is RetryUpToMaximumCountWithFixedSleep(maxRetries=10, sleepTime=1 SECONDS)
> 2014-04-04 09:15:43,770 INFO [Thread-3] ipc.Client (Client.java:handleConnectionFailure(783)) - Retrying connect to server: 0.0.0.0/0.0.0.0:8032. Already tried 1 time(s); retry policy is RetryUpToMaximumCountWithFixedSleep(maxRetries=10, sleepTime=1 SECONDS)
> 2014-04-04 09:15:44,773 INFO [Thread-3] ipc.Client (Client.java:handleConnectionFailure(783)) - Retrying connect to server: 0.0.0.0/0.0.0.0:8032. Already tried 2 time(s); retry policy is RetryUpToMaximumCountWithFixedSleep(maxRetries=10, sleepTime=1 SECONDS)
> 2014-04-04 09:15:45,775 INFO [Thread-3] ipc.Client (Client.java:handleConnectionFailure(783)) - Retrying connect to server: 0.0.0.0/0.0.0.0:8032. Already tried 3 time(s); retry policy is RetryUpToMaximumCountWithFixedSleep(maxRetries=10, sleepTime=1 SECONDS)
> 2014-04-04 09:15:46,776 INFO [Thread-3] ipc.Client (Client.java:handleConnectionFailure(783)) - Retrying connect to server: 0.0.0.0/0.0.0.0:8032. Already tried 4 time(s); retry policy is RetryUpToMaximumCountWithFixedSleep(maxRetries=10, sleepTime=1 SECONDS)
> 2014-04-04 09:15:47,779 INFO [Thread-3] ipc.Client (Client.java:handleConnectionFailure(783)) - Retrying connect to server: 0.0.0.0/0.0.0.0:8032. Already tried 5 time(s); retry policy is RetryUpToMaximumCountWithFixedSleep(maxRetries=10, sleepTime=1 SECONDS)
>
> ckw
>
> --
> Chris K Wensel
> chris@concurrentinc.com
> http://concurrentinc.com
>
>
> CONFIDENTIALITY NOTICE
> NOTICE: This message is intended for the use of the individual or entity to which it is addressed and may contain information that is confidential, privileged and exempt from disclosure under applicable law. If the reader of this message is not the intended recipient, you are hereby notified that any printing, copying, dissemination, distribution, disclosure or forwarding of this communication is strictly prohibited. If you have received this communication in error, please contact the sender immediately and delete it from your system. Thank You.
--
Chris K Wensel
chris@concurrentinc.com
http://concurrentinc.com
RE: standalone mode
Posted by Bikas Saha <bi...@hortonworks.com>.
Chris,
That is work in progress tracked at
https://issues.apache.org/jira/browse/TEZ-684. Folks from Yahoo are driving
this.
In the interim, Tez/Hive/Pig are using MiniTezCluster to write unit tests.
This runs a pseudo-distributed cluster within your unit test.
Bikas
*From:* Chris K Wensel [mailto:chris@wensel.net]
*Sent:* Friday, April 04, 2014 1:39 PM
*To:* user@tez.incubator.apache.org
*Subject:* standalone mode
sorry if i'm missing something obvious, but is there a "standalone" mode I
can run tez on within my IDE. MR has this, and its quite useful.
current'y when I apply similar voodoo, I get a hang on
TezClient#createApplication
to be clear, i'm not using mini dfs/tez cluster, but want to simply use
LocalFileSystem and have the ability to step through my code.
logs look like this below.
2014-04-04 09:15:41,653 INFO [Thread-3] client.RMProxy
(RMProxy.java:createRMProxy(56)) - Connecting to ResourceManager at /
0.0.0.0:8032
2014-04-04 09:15:42,767 INFO [Thread-3] ipc.Client
(Client.java:handleConnectionFailure(783)) - Retrying connect to server:
0.0.0.0/0.0.0.0:8032. Already tried 0 time(s); retry policy is
RetryUpToMaximumCountWithFixedSleep(maxRetries=10, sleepTime=1 SECONDS)
2014-04-04 09:15:43,770 INFO [Thread-3] ipc.Client
(Client.java:handleConnectionFailure(783)) - Retrying connect to server:
0.0.0.0/0.0.0.0:8032. Already tried 1 time(s); retry policy is
RetryUpToMaximumCountWithFixedSleep(maxRetries=10, sleepTime=1 SECONDS)
2014-04-04 09:15:44,773 INFO [Thread-3] ipc.Client
(Client.java:handleConnectionFailure(783)) - Retrying connect to server:
0.0.0.0/0.0.0.0:8032. Already tried 2 time(s); retry policy is
RetryUpToMaximumCountWithFixedSleep(maxRetries=10, sleepTime=1 SECONDS)
2014-04-04 09:15:45,775 INFO [Thread-3] ipc.Client
(Client.java:handleConnectionFailure(783)) - Retrying connect to server:
0.0.0.0/0.0.0.0:8032. Already tried 3 time(s); retry policy is
RetryUpToMaximumCountWithFixedSleep(maxRetries=10, sleepTime=1 SECONDS)
2014-04-04 09:15:46,776 INFO [Thread-3] ipc.Client
(Client.java:handleConnectionFailure(783)) - Retrying connect to server:
0.0.0.0/0.0.0.0:8032. Already tried 4 time(s); retry policy is
RetryUpToMaximumCountWithFixedSleep(maxRetries=10, sleepTime=1 SECONDS)
2014-04-04 09:15:47,779 INFO [Thread-3] ipc.Client
(Client.java:handleConnectionFailure(783)) - Retrying connect to server:
0.0.0.0/0.0.0.0:8032. Already tried 5 time(s); retry policy is
RetryUpToMaximumCountWithFixedSleep(maxRetries=10, sleepTime=1 SECONDS)
ckw
--
Chris K Wensel
chris@concurrentinc.com
http://concurrentinc.com
--
CONFIDENTIALITY NOTICE
NOTICE: This message is intended for the use of the individual or entity to
which it is addressed and may contain information that is confidential,
privileged and exempt from disclosure under applicable law. If the reader
of this message is not the intended recipient, you are hereby notified that
any printing, copying, dissemination, distribution, disclosure or
forwarding of this communication is strictly prohibited. If you have
received this communication in error, please contact the sender immediately
and delete it from your system. Thank You.