[GitHub] spark issue #22114: [SPARK-24938][Core] Prevent Netty from using onheap memo...

2018-08-20 Thread vanzin
Github user vanzin commented on the issue:

https://github.com/apache/spark/pull/22114
  
The tests are failing because they're now using too much direct memory:

```
18/08/17 09:57:35.396 rpc-server-4647-3 WARN TransportChannelHandler: 
Exception in connection from /192.168.10.22:36322
io.netty.util.internal.OutOfDirectMemoryError: failed to allocate 16777216 
byte(s) of direct memory (used: 2852126720, max: 2863661056)
```

That's the first test in jenkins; the others seem to fail because of 
improper cleanup in that first test. But it seems that either we can't just use 
direct buffers here, or maybe the test might need some tweaking.


---

-
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org



[GitHub] spark issue #22114: [SPARK-24938][Core] Prevent Netty from using onheap memo...

2018-08-17 Thread AmplabJenkins
Github user AmplabJenkins commented on the issue:

https://github.com/apache/spark/pull/22114
  
Test FAILed.
Refer to this link for build results (access rights to CI server needed): 
https://amplab.cs.berkeley.edu/jenkins//job/SparkPullRequestBuilder/94900/
Test FAILed.


---

-
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org



[GitHub] spark issue #22114: [SPARK-24938][Core] Prevent Netty from using onheap memo...

2018-08-17 Thread AmplabJenkins
Github user AmplabJenkins commented on the issue:

https://github.com/apache/spark/pull/22114
  
Merged build finished. Test FAILed.


---

-
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org



[GitHub] spark issue #22114: [SPARK-24938][Core] Prevent Netty from using onheap memo...

2018-08-17 Thread SparkQA
Github user SparkQA commented on the issue:

https://github.com/apache/spark/pull/22114
  
**[Test build #94900 has 
finished](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/94900/testReport)**
 for PR 22114 at commit 
[`c2f9ed1`](https://github.com/apache/spark/commit/c2f9ed10776842ffe0746fcc89b157675fa6c455).
 * This patch **fails Spark unit tests**.
 * This patch merges cleanly.
 * This patch adds no public classes.


---

-
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org



[GitHub] spark issue #22114: [SPARK-24938][Core] Prevent Netty from using onheap memo...

2018-08-17 Thread vanzin
Github user vanzin commented on the issue:

https://github.com/apache/spark/pull/22114
  
FYI I wrote a small micro bench to pound on the pool in both the old way 
(mixed on heap and off heap allocations) and new way (only off heap) and it 
actually seems faster if all allocations use the `buffer()` call. So I guess 
we're safe here.


---

-
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org



[GitHub] spark issue #22114: [SPARK-24938][Core] Prevent Netty from using onheap memo...

2018-08-17 Thread SparkQA
Github user SparkQA commented on the issue:

https://github.com/apache/spark/pull/22114
  
**[Test build #94900 has 
started](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/94900/testReport)**
 for PR 22114 at commit 
[`c2f9ed1`](https://github.com/apache/spark/commit/c2f9ed10776842ffe0746fcc89b157675fa6c455).


---

-
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org



[GitHub] spark issue #22114: [SPARK-24938][Core] Prevent Netty from using onheap memo...

2018-08-17 Thread vanzin
Github user vanzin commented on the issue:

https://github.com/apache/spark/pull/22114
  
retest this please


---

-
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org



[GitHub] spark issue #22114: [SPARK-24938][Core] Prevent Netty from using onheap memo...

2018-08-16 Thread NiharS
Github user NiharS commented on the issue:

https://github.com/apache/spark/pull/22114
  
They pass on my machine :(


---

-
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org



[GitHub] spark issue #22114: [SPARK-24938][Core] Prevent Netty from using onheap memo...

2018-08-16 Thread AmplabJenkins
Github user AmplabJenkins commented on the issue:

https://github.com/apache/spark/pull/22114
  
Merged build finished. Test FAILed.


---

-
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org



[GitHub] spark issue #22114: [SPARK-24938][Core] Prevent Netty from using onheap memo...

2018-08-16 Thread AmplabJenkins
Github user AmplabJenkins commented on the issue:

https://github.com/apache/spark/pull/22114
  
Test FAILed.
Refer to this link for build results (access rights to CI server needed): 
https://amplab.cs.berkeley.edu/jenkins//job/SparkPullRequestBuilder/94867/
Test FAILed.


---

-
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org



[GitHub] spark issue #22114: [SPARK-24938][Core] Prevent Netty from using onheap memo...

2018-08-16 Thread SparkQA
Github user SparkQA commented on the issue:

https://github.com/apache/spark/pull/22114
  
**[Test build #94867 has 
finished](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/94867/testReport)**
 for PR 22114 at commit 
[`c2f9ed1`](https://github.com/apache/spark/commit/c2f9ed10776842ffe0746fcc89b157675fa6c455).
 * This patch **fails Spark unit tests**.
 * This patch merges cleanly.
 * This patch adds no public classes.


---

-
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org



[GitHub] spark issue #22114: [SPARK-24938][Core] Prevent Netty from using onheap memo...

2018-08-16 Thread SparkQA
Github user SparkQA commented on the issue:

https://github.com/apache/spark/pull/22114
  
**[Test build #94867 has 
started](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/94867/testReport)**
 for PR 22114 at commit 
[`c2f9ed1`](https://github.com/apache/spark/commit/c2f9ed10776842ffe0746fcc89b157675fa6c455).


---

-
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org



[GitHub] spark issue #22114: [SPARK-24938][Core] Prevent Netty from using onheap memo...

2018-08-16 Thread vanzin
Github user vanzin commented on the issue:

https://github.com/apache/spark/pull/22114
  
ok to test


---

-
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org



[GitHub] spark issue #22114: [SPARK-24938][Core] Prevent Netty from using onheap memo...

2018-08-16 Thread NiharS
Github user NiharS commented on the issue:

https://github.com/apache/spark/pull/22114
  
Tried with a significantly larger input, both with and without the change. 
They ran in just about the same time.


---

-
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org



[GitHub] spark issue #22114: [SPARK-24938][Core] Prevent Netty from using onheap memo...

2018-08-15 Thread NiharS
Github user NiharS commented on the issue:

https://github.com/apache/spark/pull/22114
  
Working on a better test now! With my small program there was definitely no 
noticeable change in runtime. The memory monitor was running too, though, so I 
did have some external factors. I'll try with a significantly larger input and 
see if there's anything that stands out


---

-
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org



[GitHub] spark issue #22114: [SPARK-24938][Core] Prevent Netty from using onheap memo...

2018-08-15 Thread AmplabJenkins
Github user AmplabJenkins commented on the issue:

https://github.com/apache/spark/pull/22114
  
Can one of the admins verify this patch?


---

-
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org



[GitHub] spark issue #22114: [SPARK-24938][Core] Prevent Netty from using onheap memo...

2018-08-15 Thread vanzin
Github user vanzin commented on the issue:

https://github.com/apache/spark/pull/22114
  
Did you get a chance to measure whether this has any negative effect of 
performance? Since it's using a buffer pool I wouldn't expect any, but you 
never know...


---

-
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org



[GitHub] spark issue #22114: [SPARK-24938][Core] Prevent Netty from using onheap memo...

2018-08-15 Thread vanzin
Github user vanzin commented on the issue:

https://github.com/apache/spark/pull/22114
  
ok to test


---

-
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org



[GitHub] spark issue #22114: [SPARK-24938][Core] Prevent Netty from using onheap memo...

2018-08-15 Thread AmplabJenkins
Github user AmplabJenkins commented on the issue:

https://github.com/apache/spark/pull/22114
  
Can one of the admins verify this patch?


---

-
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org



[GitHub] spark issue #22114: [SPARK-24938][Core] Prevent Netty from using onheap memo...

2018-08-15 Thread AmplabJenkins
Github user AmplabJenkins commented on the issue:

https://github.com/apache/spark/pull/22114
  
Can one of the admins verify this patch?


---

-
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org