Sqoop, mail # dev - Review Request: SQOOP-813 Sqoop2: LoaderExecutor might get into deadlock when exception is raised outside Loader itself

Review Request: SQOOP-813 Sqoop2: LoaderExecutor might get into deadlock when exception is raised outside Loader itself

Jarek Cecho 2013-01-06, 18:04

-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/8849/
-----------------------------------------------------------

Review request for Sqoop.
Description
-------

I've extended the try-catch block to protect entire execution of LoaderExecutor.
This addresses bug SQOOP-813.
https://issues.apache.org/jira/browse/SQOOP-813
Diffs
-----

execution/mapreduce/src/main/java/org/apache/sqoop/job/mr/SqoopOutputFormatLoadExecutor.java d158327

Diff: https://reviews.apache.org/r/8849/diff/
Testing
-------

Unit tests are passing and I've tested this on real cluster (in cracked code to trigger exception in framework code).
Thanks,

Jarek Cecho

Re: Review Request: SQOOP-813 Sqoop2: LoaderExecutor might get into deadlock when exception is raised outside Loader itself

Hari Shreedharan 2013-01-07, 04:17

-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/8849/#review15106
-----------------------------------------------------------

Ship it!

execution/mapreduce/src/main/java/org/apache/sqoop/job/mr/SqoopOutputFormatLoadExecutor.java
<https://reviews.apache.org/r/8849/#comment32696>

Really minor nit:
Since the default block from the switch-case can throw - even that exception is wrapped in another SqoopException - maybe we should throw that as is to reduce the one extra level in the stack trace.
- Hari Shreedharan
On Jan. 6, 2013, 6:04 p.m., Jarek Cecho wrote:
>
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/8849/
> -----------------------------------------------------------
>
> (Updated Jan. 6, 2013, 6:04 p.m.)
>
>
> Review request for Sqoop.
>
>
> Description
> -------
>
> I've extended the try-catch block to protect entire execution of LoaderExecutor.
>
>
> This addresses bug SQOOP-813.
> https://issues.apache.org/jira/browse/SQOOP-813
>
>
> Diffs
> -----
>
> execution/mapreduce/src/main/java/org/apache/sqoop/job/mr/SqoopOutputFormatLoadExecutor.java d158327
>
> Diff: https://reviews.apache.org/r/8849/diff/
>
>
> Testing
> -------
>
> Unit tests are passing and I've tested this on real cluster (in cracked code to trigger exception in framework code).
>
>
> Thanks,
>
> Jarek Cecho
>
>

Re: Review Request: SQOOP-813 Sqoop2: LoaderExecutor might get into deadlock when exception is raised outside Loader itself

Cheolsoo Park 2013-01-23, 00:42

-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/8849/#review15602
-----------------------------------------------------------

Ship it!
In fact, I was wrong. The patch is up-to-date. Thanks for clarifying it Jarcec!

- Cheolsoo Park
On Jan. 6, 2013, 6:04 p.m., Jarek Cecho wrote:
>
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/8849/
> -----------------------------------------------------------
>
> (Updated Jan. 6, 2013, 6:04 p.m.)
>
>
> Review request for Sqoop.
>
>
> Description
> -------
>
> I've extended the try-catch block to protect entire execution of LoaderExecutor.
>
>
> This addresses bug SQOOP-813.
> https://issues.apache.org/jira/browse/SQOOP-813
>
>
> Diffs
> -----
>
> execution/mapreduce/src/main/java/org/apache/sqoop/job/mr/SqoopOutputFormatLoadExecutor.java d158327
>
> Diff: https://reviews.apache.org/r/8849/diff/
>
>
> Testing
> -------
>
> Unit tests are passing and I've tested this on real cluster (in cracked code to trigger exception in framework code).
>
>
> Thanks,
>
> Jarek Cecho
>
>

Re: Review Request: SQOOP-813 Sqoop2: LoaderExecutor might get into deadlock when exception is raised outside Loader itself

Cheolsoo Park 2013-01-23, 00:21

-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/8849/#review15597
-----------------------------------------------------------
Jarcec, I noticed that the patch on the JIRA is different from this one. Would you mind updating the RB?

- Cheolsoo Park
On Jan. 6, 2013, 6:04 p.m., Jarek Cecho wrote:
>
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/8849/
> -----------------------------------------------------------
>
> (Updated Jan. 6, 2013, 6:04 p.m.)
>
>
> Review request for Sqoop.
>
>
> Description
> -------
>
> I've extended the try-catch block to protect entire execution of LoaderExecutor.
>
>
> This addresses bug SQOOP-813.
> https://issues.apache.org/jira/browse/SQOOP-813
>
>
> Diffs
> -----
>
> execution/mapreduce/src/main/java/org/apache/sqoop/job/mr/SqoopOutputFormatLoadExecutor.java d158327
>
> Diff: https://reviews.apache.org/r/8849/diff/
>
>
> Testing
> -------
>
> Unit tests are passing and I've tested this on real cluster (in cracked code to trigger exception in framework code).
>
>
> Thanks,
>
> Jarek Cecho
>
>