Re: Spark cluster multi tenancy
Would be interested to know the answer too. On Wed, Aug 26, 2015 at 11:45 AM, Sadhan Sood sadhan.s...@gmail.com wrote: Interestingly, if there is nothing running on dev spark-shell, it recovers successfully and regains the lost executors. Attaching the log for that. Notice, the Registering block manager .. statements in the very end after all executors were lost. On Wed, Aug 26, 2015 at 11:27 AM, Sadhan Sood sadhan.s...@gmail.com wrote: Attaching log for when the dev job gets stuck (once all its executors are lost due to preemption). This is a spark-shell job running in yarn-client mode. On Wed, Aug 26, 2015 at 10:45 AM, Sadhan Sood sadhan.s...@gmail.com wrote: Hi All, We've set up our spark cluster on aws running on yarn (running on hadoop 2.3) with fair scheduling and preemption turned on. The cluster is shared for prod and dev work where prod runs with a higher fair share and can preempt dev jobs if there are not enough resources available for it. It appears that dev jobs which get preempted often get unstable after losing some executors and the whole jobs gets stuck (without making any progress) or end up getting restarted (and hence losing all the work done). Has someone encountered this before ? Is the solution just to set spark.task.maxFailures to a really high value to recover from task failures in such scenarios? Are there other approaches that people have taken for spark multi tenancy that works better in such scenario? Thanks, Sadhan - To unsubscribe, e-mail: user-unsubscr...@spark.apache.org For additional commands, e-mail: user-h...@spark.apache.org
Re: Spark Sql behaves strangely with tables with a lot of partitions
@Michael: would listStatus calls read the actual parquet footers within the folders? On Mon, Aug 24, 2015 at 11:36 AM, Sereday, Scott scott.sere...@nielsen.com wrote: Can you please remove me from this distribution list? (Filling up my inbox too fast) *From:* Michael Armbrust [mailto:mich...@databricks.com] *Sent:* Monday, August 24, 2015 2:13 PM *To:* Philip Weaver philip.wea...@gmail.com *Cc:* Jerrick Hoang jerrickho...@gmail.com; Raghavendra Pandey raghavendra.pan...@gmail.com; User user@spark.apache.org; Cheng, Hao hao.ch...@intel.com *Subject:* Re: Spark Sql behaves strangely with tables with a lot of partitions I think we are mostly bottlenecked at this point by how fast we can make listStatus calls to discover the folders. That said, we are happy to accept suggestions or PRs to make this faster. Perhaps you can describe how your home grown partitioning works? On Sun, Aug 23, 2015 at 7:38 PM, Philip Weaver philip.wea...@gmail.com wrote: 1 minute to discover 1000s of partitions -- yes, that is what I have observed. And I would assert that is very slow. On Sun, Aug 23, 2015 at 7:16 PM, Michael Armbrust mich...@databricks.com wrote: We should not be actually scanning all of the data of all of the partitions, but we do need to at least list all of the available directories so that we can apply your predicates to the actual values that are present when we are deciding which files need to be read in a given spark job. While this is a somewhat expensive operation, we do it in parallel and we cache this information when you access the same relation more than once. Can you provide a little more detail about how exactly you are accessing the parquet data (are you using sqlContext.read or creating persistent tables in the metastore?), and how long it is taking? It would also be good to know how many partitions we are talking about and how much data is in each. Finally, I'd like to see the stacktrace where it is hanging to make sure my above assertions are correct. We have several tables internally that have 1000s of partitions and while it takes ~1 minute initially to discover the metadata, after that we are able to query the data interactively. On Sun, Aug 23, 2015 at 2:00 AM, Jerrick Hoang jerrickho...@gmail.com wrote: anybody has any suggestions? On Fri, Aug 21, 2015 at 3:14 PM, Jerrick Hoang jerrickho...@gmail.com wrote: Is there a workaround without updating Hadoop? Would really appreciate if someone can explain what spark is trying to do here and what is an easy way to turn this off. Thanks all! On Fri, Aug 21, 2015 at 11:09 AM, Raghavendra Pandey raghavendra.pan...@gmail.com wrote: Did you try with hadoop version 2.7.1 .. It is known that s3a works really well with parquet which is available in 2.7. They fixed lot of issues related to metadata reading there... On Aug 21, 2015 11:24 PM, Jerrick Hoang jerrickho...@gmail.com wrote: @Cheng, Hao : Physical plans show that it got stuck on scanning S3! (table is partitioned by date_prefix and hour) explain select count(*) from test_table where date_prefix='20150819' and hour='00'; TungstenAggregate(key=[], value=[(count(1),mode=Final,isDistinct=false)] TungstenExchange SinglePartition TungstenAggregate(key=[], value=[(count(1),mode=Partial,isDistinct=false)] Scan ParquetRelation[ .. about 1000 partition paths go here ] Why does spark have to scan all partitions when the query only concerns with 1 partitions? Doesn't it defeat the purpose of partitioning? Thanks! On Thu, Aug 20, 2015 at 4:12 PM, Philip Weaver philip.wea...@gmail.com wrote: I hadn't heard of spark.sql.sources.partitionDiscovery.enabled before, and I couldn't find much information about it online. What does it mean exactly to disable it? Are there any negative consequences to disabling it? On Wed, Aug 19, 2015 at 10:53 PM, Cheng, Hao hao.ch...@intel.com wrote: Can you make some more profiling? I am wondering if the driver is busy with scanning the HDFS / S3. Like jstack pid of driver process And also, it’s will be great if you can paste the physical plan for the simple query. *From:* Jerrick Hoang [mailto:jerrickho...@gmail.com] *Sent:* Thursday, August 20, 2015 1:46 PM *To:* Cheng, Hao *Cc:* Philip Weaver; user *Subject:* Re: Spark Sql behaves strangely with tables with a lot of partitions I cloned from TOT after 1.5.0 cut off. I noticed there were a couple of CLs trying to speed up spark sql with tables with a huge number of partitions, I've made sure that those CLs are included but it's still very slow On Wed, Aug 19, 2015 at 10:43 PM, Cheng, Hao hao.ch...@intel.com wrote: Yes, you can try set the spark.sql.sources.partitionDiscovery.enabled to false. BTW, which version are you using? Hao *From:* Jerrick Hoang [mailto:jerrickho...@gmail.com] *Sent:* Thursday, August 20, 2015 12:16 PM *To:* Philip
Re: Spark Sql behaves strangely with tables with a lot of partitions
anybody has any suggestions? On Fri, Aug 21, 2015 at 3:14 PM, Jerrick Hoang jerrickho...@gmail.com wrote: Is there a workaround without updating Hadoop? Would really appreciate if someone can explain what spark is trying to do here and what is an easy way to turn this off. Thanks all! On Fri, Aug 21, 2015 at 11:09 AM, Raghavendra Pandey raghavendra.pan...@gmail.com wrote: Did you try with hadoop version 2.7.1 .. It is known that s3a works really well with parquet which is available in 2.7. They fixed lot of issues related to metadata reading there... On Aug 21, 2015 11:24 PM, Jerrick Hoang jerrickho...@gmail.com wrote: @Cheng, Hao : Physical plans show that it got stuck on scanning S3! (table is partitioned by date_prefix and hour) explain select count(*) from test_table where date_prefix='20150819' and hour='00'; TungstenAggregate(key=[], value=[(count(1),mode=Final,isDistinct=false)] TungstenExchange SinglePartition TungstenAggregate(key=[], value=[(count(1),mode=Partial,isDistinct=false)] Scan ParquetRelation[ .. about 1000 partition paths go here ] Why does spark have to scan all partitions when the query only concerns with 1 partitions? Doesn't it defeat the purpose of partitioning? Thanks! On Thu, Aug 20, 2015 at 4:12 PM, Philip Weaver philip.wea...@gmail.com wrote: I hadn't heard of spark.sql.sources.partitionDiscovery.enabled before, and I couldn't find much information about it online. What does it mean exactly to disable it? Are there any negative consequences to disabling it? On Wed, Aug 19, 2015 at 10:53 PM, Cheng, Hao hao.ch...@intel.com wrote: Can you make some more profiling? I am wondering if the driver is busy with scanning the HDFS / S3. Like jstack pid of driver process And also, it’s will be great if you can paste the physical plan for the simple query. *From:* Jerrick Hoang [mailto:jerrickho...@gmail.com] *Sent:* Thursday, August 20, 2015 1:46 PM *To:* Cheng, Hao *Cc:* Philip Weaver; user *Subject:* Re: Spark Sql behaves strangely with tables with a lot of partitions I cloned from TOT after 1.5.0 cut off. I noticed there were a couple of CLs trying to speed up spark sql with tables with a huge number of partitions, I've made sure that those CLs are included but it's still very slow On Wed, Aug 19, 2015 at 10:43 PM, Cheng, Hao hao.ch...@intel.com wrote: Yes, you can try set the spark.sql.sources.partitionDiscovery.enabled to false. BTW, which version are you using? Hao *From:* Jerrick Hoang [mailto:jerrickho...@gmail.com] *Sent:* Thursday, August 20, 2015 12:16 PM *To:* Philip Weaver *Cc:* user *Subject:* Re: Spark Sql behaves strangely with tables with a lot of partitions I guess the question is why does spark have to do partition discovery with all partitions when the query only needs to look at one partition? Is there a conf flag to turn this off? On Wed, Aug 19, 2015 at 9:02 PM, Philip Weaver philip.wea...@gmail.com wrote: I've had the same problem. It turns out that Spark (specifically parquet) is very slow at partition discovery. It got better in 1.5 (not yet released), but was still unacceptably slow. Sadly, we ended up reading parquet files manually in Python (via C++) and had to abandon Spark SQL because of this problem. On Wed, Aug 19, 2015 at 7:51 PM, Jerrick Hoang jerrickho...@gmail.com wrote: Hi all, I did a simple experiment with Spark SQL. I created a partitioned parquet table with only one partition (date=20140701). A simple `select count(*) from table where date=20140701` would run very fast (0.1 seconds). However, as I added more partitions the query takes longer and longer. When I added about 10,000 partitions, the query took way too long. I feel like querying for a single partition should not be affected by having more partitions. Is this a known behaviour? What does spark try to do here? Thanks, Jerrick
Re: Spark Sql behaves strangely with tables with a lot of partitions
@Cheng, Hao : Physical plans show that it got stuck on scanning S3! (table is partitioned by date_prefix and hour) explain select count(*) from test_table where date_prefix='20150819' and hour='00'; TungstenAggregate(key=[], value=[(count(1),mode=Final,isDistinct=false)] TungstenExchange SinglePartition TungstenAggregate(key=[], value=[(count(1),mode=Partial,isDistinct=false)] Scan ParquetRelation[ .. about 1000 partition paths go here ] Why does spark have to scan all partitions when the query only concerns with 1 partitions? Doesn't it defeat the purpose of partitioning? Thanks! On Thu, Aug 20, 2015 at 4:12 PM, Philip Weaver philip.wea...@gmail.com wrote: I hadn't heard of spark.sql.sources.partitionDiscovery.enabled before, and I couldn't find much information about it online. What does it mean exactly to disable it? Are there any negative consequences to disabling it? On Wed, Aug 19, 2015 at 10:53 PM, Cheng, Hao hao.ch...@intel.com wrote: Can you make some more profiling? I am wondering if the driver is busy with scanning the HDFS / S3. Like jstack pid of driver process And also, it’s will be great if you can paste the physical plan for the simple query. *From:* Jerrick Hoang [mailto:jerrickho...@gmail.com] *Sent:* Thursday, August 20, 2015 1:46 PM *To:* Cheng, Hao *Cc:* Philip Weaver; user *Subject:* Re: Spark Sql behaves strangely with tables with a lot of partitions I cloned from TOT after 1.5.0 cut off. I noticed there were a couple of CLs trying to speed up spark sql with tables with a huge number of partitions, I've made sure that those CLs are included but it's still very slow On Wed, Aug 19, 2015 at 10:43 PM, Cheng, Hao hao.ch...@intel.com wrote: Yes, you can try set the spark.sql.sources.partitionDiscovery.enabled to false. BTW, which version are you using? Hao *From:* Jerrick Hoang [mailto:jerrickho...@gmail.com] *Sent:* Thursday, August 20, 2015 12:16 PM *To:* Philip Weaver *Cc:* user *Subject:* Re: Spark Sql behaves strangely with tables with a lot of partitions I guess the question is why does spark have to do partition discovery with all partitions when the query only needs to look at one partition? Is there a conf flag to turn this off? On Wed, Aug 19, 2015 at 9:02 PM, Philip Weaver philip.wea...@gmail.com wrote: I've had the same problem. It turns out that Spark (specifically parquet) is very slow at partition discovery. It got better in 1.5 (not yet released), but was still unacceptably slow. Sadly, we ended up reading parquet files manually in Python (via C++) and had to abandon Spark SQL because of this problem. On Wed, Aug 19, 2015 at 7:51 PM, Jerrick Hoang jerrickho...@gmail.com wrote: Hi all, I did a simple experiment with Spark SQL. I created a partitioned parquet table with only one partition (date=20140701). A simple `select count(*) from table where date=20140701` would run very fast (0.1 seconds). However, as I added more partitions the query takes longer and longer. When I added about 10,000 partitions, the query took way too long. I feel like querying for a single partition should not be affected by having more partitions. Is this a known behaviour? What does spark try to do here? Thanks, Jerrick
Re: Spark Sql behaves strangely with tables with a lot of partitions
Is there a workaround without updating Hadoop? Would really appreciate if someone can explain what spark is trying to do here and what is an easy way to turn this off. Thanks all! On Fri, Aug 21, 2015 at 11:09 AM, Raghavendra Pandey raghavendra.pan...@gmail.com wrote: Did you try with hadoop version 2.7.1 .. It is known that s3a works really well with parquet which is available in 2.7. They fixed lot of issues related to metadata reading there... On Aug 21, 2015 11:24 PM, Jerrick Hoang jerrickho...@gmail.com wrote: @Cheng, Hao : Physical plans show that it got stuck on scanning S3! (table is partitioned by date_prefix and hour) explain select count(*) from test_table where date_prefix='20150819' and hour='00'; TungstenAggregate(key=[], value=[(count(1),mode=Final,isDistinct=false)] TungstenExchange SinglePartition TungstenAggregate(key=[], value=[(count(1),mode=Partial,isDistinct=false)] Scan ParquetRelation[ .. about 1000 partition paths go here ] Why does spark have to scan all partitions when the query only concerns with 1 partitions? Doesn't it defeat the purpose of partitioning? Thanks! On Thu, Aug 20, 2015 at 4:12 PM, Philip Weaver philip.wea...@gmail.com wrote: I hadn't heard of spark.sql.sources.partitionDiscovery.enabled before, and I couldn't find much information about it online. What does it mean exactly to disable it? Are there any negative consequences to disabling it? On Wed, Aug 19, 2015 at 10:53 PM, Cheng, Hao hao.ch...@intel.com wrote: Can you make some more profiling? I am wondering if the driver is busy with scanning the HDFS / S3. Like jstack pid of driver process And also, it’s will be great if you can paste the physical plan for the simple query. *From:* Jerrick Hoang [mailto:jerrickho...@gmail.com] *Sent:* Thursday, August 20, 2015 1:46 PM *To:* Cheng, Hao *Cc:* Philip Weaver; user *Subject:* Re: Spark Sql behaves strangely with tables with a lot of partitions I cloned from TOT after 1.5.0 cut off. I noticed there were a couple of CLs trying to speed up spark sql with tables with a huge number of partitions, I've made sure that those CLs are included but it's still very slow On Wed, Aug 19, 2015 at 10:43 PM, Cheng, Hao hao.ch...@intel.com wrote: Yes, you can try set the spark.sql.sources.partitionDiscovery.enabled to false. BTW, which version are you using? Hao *From:* Jerrick Hoang [mailto:jerrickho...@gmail.com] *Sent:* Thursday, August 20, 2015 12:16 PM *To:* Philip Weaver *Cc:* user *Subject:* Re: Spark Sql behaves strangely with tables with a lot of partitions I guess the question is why does spark have to do partition discovery with all partitions when the query only needs to look at one partition? Is there a conf flag to turn this off? On Wed, Aug 19, 2015 at 9:02 PM, Philip Weaver philip.wea...@gmail.com wrote: I've had the same problem. It turns out that Spark (specifically parquet) is very slow at partition discovery. It got better in 1.5 (not yet released), but was still unacceptably slow. Sadly, we ended up reading parquet files manually in Python (via C++) and had to abandon Spark SQL because of this problem. On Wed, Aug 19, 2015 at 7:51 PM, Jerrick Hoang jerrickho...@gmail.com wrote: Hi all, I did a simple experiment with Spark SQL. I created a partitioned parquet table with only one partition (date=20140701). A simple `select count(*) from table where date=20140701` would run very fast (0.1 seconds). However, as I added more partitions the query takes longer and longer. When I added about 10,000 partitions, the query took way too long. I feel like querying for a single partition should not be affected by having more partitions. Is this a known behaviour? What does spark try to do here? Thanks, Jerrick
Re: Spark Sql behaves strangely with tables with a lot of partitions
I cloned from TOT after 1.5.0 cut off. I noticed there were a couple of CLs trying to speed up spark sql with tables with a huge number of partitions, I've made sure that those CLs are included but it's still very slow On Wed, Aug 19, 2015 at 10:43 PM, Cheng, Hao hao.ch...@intel.com wrote: Yes, you can try set the spark.sql.sources.partitionDiscovery.enabled to false. BTW, which version are you using? Hao *From:* Jerrick Hoang [mailto:jerrickho...@gmail.com] *Sent:* Thursday, August 20, 2015 12:16 PM *To:* Philip Weaver *Cc:* user *Subject:* Re: Spark Sql behaves strangely with tables with a lot of partitions I guess the question is why does spark have to do partition discovery with all partitions when the query only needs to look at one partition? Is there a conf flag to turn this off? On Wed, Aug 19, 2015 at 9:02 PM, Philip Weaver philip.wea...@gmail.com wrote: I've had the same problem. It turns out that Spark (specifically parquet) is very slow at partition discovery. It got better in 1.5 (not yet released), but was still unacceptably slow. Sadly, we ended up reading parquet files manually in Python (via C++) and had to abandon Spark SQL because of this problem. On Wed, Aug 19, 2015 at 7:51 PM, Jerrick Hoang jerrickho...@gmail.com wrote: Hi all, I did a simple experiment with Spark SQL. I created a partitioned parquet table with only one partition (date=20140701). A simple `select count(*) from table where date=20140701` would run very fast (0.1 seconds). However, as I added more partitions the query takes longer and longer. When I added about 10,000 partitions, the query took way too long. I feel like querying for a single partition should not be affected by having more partitions. Is this a known behaviour? What does spark try to do here? Thanks, Jerrick
Re: Spark Sql behaves strangely with tables with a lot of partitions
I guess the question is why does spark have to do partition discovery with all partitions when the query only needs to look at one partition? Is there a conf flag to turn this off? On Wed, Aug 19, 2015 at 9:02 PM, Philip Weaver philip.wea...@gmail.com wrote: I've had the same problem. It turns out that Spark (specifically parquet) is very slow at partition discovery. It got better in 1.5 (not yet released), but was still unacceptably slow. Sadly, we ended up reading parquet files manually in Python (via C++) and had to abandon Spark SQL because of this problem. On Wed, Aug 19, 2015 at 7:51 PM, Jerrick Hoang jerrickho...@gmail.com wrote: Hi all, I did a simple experiment with Spark SQL. I created a partitioned parquet table with only one partition (date=20140701). A simple `select count(*) from table where date=20140701` would run very fast (0.1 seconds). However, as I added more partitions the query takes longer and longer. When I added about 10,000 partitions, the query took way too long. I feel like querying for a single partition should not be affected by having more partitions. Is this a known behaviour? What does spark try to do here? Thanks, Jerrick
Spark Sql behaves strangely with tables with a lot of partitions
Hi all, I did a simple experiment with Spark SQL. I created a partitioned parquet table with only one partition (date=20140701). A simple `select count(*) from table where date=20140701` would run very fast (0.1 seconds). However, as I added more partitions the query takes longer and longer. When I added about 10,000 partitions, the query took way too long. I feel like querying for a single partition should not be affected by having more partitions. Is this a known behaviour? What does spark try to do here? Thanks, Jerrick
Re: Spark failed while trying to read parquet files
Yes! I was being dumb, should have caught that earlier, thank you Cheng Lian On Fri, Aug 7, 2015 at 4:25 PM, Cheng Lian lian.cs@gmail.com wrote: It doesn't seem to be Parquet 1.7.0 since the package name isn't under org.apache.parquet (1.7.0 is the first official Apache release of Parquet). The version you were using is probably Parquet 1.6.0rc3 according to the line number information: https://github.com/apache/parquet-mr/blob/parquet-1.6.0rc3/parquet-hadoop/src/main/java/parquet/format/converter/ParquetMetadataConverter.java#L249 And you're hitting PARQUET-136, which has been fixed in (the real) Parquet 1.7.0 https://issues.apache.org/jira/browse/PARQUET-136 Cheng On 8/8/15 6:20 AM, Jerrick Hoang wrote: Hi all, I have a partitioned parquet table (very small table with only 2 partitions). The version of spark is 1.4.1, parquet version is 1.7.0. I applied this patch to spark [SPARK-7743] so I assume that spark can read parquet files normally, however, I'm getting this when trying to do a simple `select count(*) from table`, ```org.apache.spark.SparkException: Job aborted due to stage failure: Task 29 in stage 44.0 failed 15 times, most recent failure: Lost task 29.14 in stage 44.0: java.lang.NullPointerException at parquet.format.converter.ParquetMetadataConverter.fromParquetStatistics(ParquetMetadataConverter.java:249) at parquet.format.converter.ParquetMetadataConverter.fromParquetMetadata(ParquetMetadataConverter.java:543) at parquet.format.converter.ParquetMetadataConverter.readParquetMetadata(ParquetMetadataConverter.java:520) at parquet.hadoop.ParquetFileReader.readFooter(ParquetFileReader.java:426) at parquet.hadoop.ParquetFileReader.readFooter(ParquetFileReader.java:381) at parquet.hadoop.ParquetRecordReader.initializeInternalReader(ParquetRecordReader.java:155) at parquet.hadoop.ParquetRecordReader.initialize(ParquetRecordReader.java:138) at org.apache.spark.sql.sources.SqlNewHadoopRDD$$anon$1.init(SqlNewHadoopRDD.scala:153) at org.apache.spark.sql.sources.SqlNewHadoopRDD.compute(SqlNewHadoopRDD.scala:124) at org.apache.spark.sql.sources.SqlNewHadoopRDD.compute(SqlNewHadoopRDD.scala:66) at org.apache.spark.rdd.RDD.computeOrReadCheckpoint(RDD.scala:277) at org.apache.spark.rdd.RDD.iterator(RDD.scala:244) at org.apache.spark.rdd.MapPartitionsRDD.compute(MapPartitionsRDD.scala:35) at org.apache.spark.rdd.RDD.computeOrReadCheckpoint(RDD.scala:277) at org.apache.spark.rdd.RDD.iterator(RDD.scala:244) at org.apache.spark.rdd.UnionRDD.compute(UnionRDD.scala:87) at org.apache.spark.rdd.RDD.computeOrReadCheckpoint(RDD.scala:277) at org.apache.spark.rdd.RDD.iterator(RDD.scala:244) at org.apache.spark.rdd.MapPartitionsRDD.compute(MapPartitionsRDD.scala:35) at org.apache.spark.rdd.RDD.computeOrReadCheckpoint(RDD.scala:277) at org.apache.spark.rdd.RDD.iterator(RDD.scala:244) at org.apache.spark.rdd.MapPartitionsRDD.compute(MapPartitionsRDD.scala:35) at org.apache.spark.rdd.RDD.computeOrReadCheckpoint(RDD.scala:277) at org.apache.spark.rdd.RDD.iterator(RDD.scala:244) at org.apache.spark.scheduler.ShuffleMapTask.runTask(ShuffleMapTask.scala:70) at org.apache.spark.scheduler.ShuffleMapTask.runTask(ShuffleMapTask.scala:41) at org.apache.spark.scheduler.Task.run(Task.scala:70) at org.apache.spark.executor.Executor$TaskRunner.run(Executor.scala:213) at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1142) at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:617) at java.lang.Thread.run(Thread.java:745)``` Has anybody seen this before? Thanks
Spark failed while trying to read parquet files
Hi all, I have a partitioned parquet table (very small table with only 2 partitions). The version of spark is 1.4.1, parquet version is 1.7.0. I applied this patch to spark [SPARK-7743] so I assume that spark can read parquet files normally, however, I'm getting this when trying to do a simple `select count(*) from table`, ```org.apache.spark.SparkException: Job aborted due to stage failure: Task 29 in stage 44.0 failed 15 times, most recent failure: Lost task 29.14 in stage 44.0: java.lang.NullPointerException at parquet.format.converter.ParquetMetadataConverter.fromParquetStatistics(ParquetMetadataConverter.java:249) at parquet.format.converter.ParquetMetadataConverter.fromParquetMetadata(ParquetMetadataConverter.java:543) at parquet.format.converter.ParquetMetadataConverter.readParquetMetadata(ParquetMetadataConverter.java:520) at parquet.hadoop.ParquetFileReader.readFooter(ParquetFileReader.java:426) at parquet.hadoop.ParquetFileReader.readFooter(ParquetFileReader.java:381) at parquet.hadoop.ParquetRecordReader.initializeInternalReader(ParquetRecordReader.java:155) at parquet.hadoop.ParquetRecordReader.initialize(ParquetRecordReader.java:138) at org.apache.spark.sql.sources.SqlNewHadoopRDD$$anon$1.init(SqlNewHadoopRDD.scala:153) at org.apache.spark.sql.sources.SqlNewHadoopRDD.compute(SqlNewHadoopRDD.scala:124) at org.apache.spark.sql.sources.SqlNewHadoopRDD.compute(SqlNewHadoopRDD.scala:66) at org.apache.spark.rdd.RDD.computeOrReadCheckpoint(RDD.scala:277) at org.apache.spark.rdd.RDD.iterator(RDD.scala:244) at org.apache.spark.rdd.MapPartitionsRDD.compute(MapPartitionsRDD.scala:35) at org.apache.spark.rdd.RDD.computeOrReadCheckpoint(RDD.scala:277) at org.apache.spark.rdd.RDD.iterator(RDD.scala:244) at org.apache.spark.rdd.UnionRDD.compute(UnionRDD.scala:87) at org.apache.spark.rdd.RDD.computeOrReadCheckpoint(RDD.scala:277) at org.apache.spark.rdd.RDD.iterator(RDD.scala:244) at org.apache.spark.rdd.MapPartitionsRDD.compute(MapPartitionsRDD.scala:35) at org.apache.spark.rdd.RDD.computeOrReadCheckpoint(RDD.scala:277) at org.apache.spark.rdd.RDD.iterator(RDD.scala:244) at org.apache.spark.rdd.MapPartitionsRDD.compute(MapPartitionsRDD.scala:35) at org.apache.spark.rdd.RDD.computeOrReadCheckpoint(RDD.scala:277) at org.apache.spark.rdd.RDD.iterator(RDD.scala:244) at org.apache.spark.scheduler.ShuffleMapTask.runTask(ShuffleMapTask.scala:70) at org.apache.spark.scheduler.ShuffleMapTask.runTask(ShuffleMapTask.scala:41) at org.apache.spark.scheduler.Task.run(Task.scala:70) at org.apache.spark.executor.Executor$TaskRunner.run(Executor.scala:213) at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1142) at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:617) at java.lang.Thread.run(Thread.java:745)``` Has anybody seen this before? Thanks
Re: Spark is much slower than direct access MySQL
how big is the dataset? how complicated is the query? On Sun, Jul 26, 2015 at 12:47 AM Louis Hust louis.h...@gmail.com wrote: Hi, all, I am using spark DataFrame to fetch small table from MySQL, and i found it cost so much than directly access MySQL Using JDBC. Time cost for Spark is about 2033ms, and direct access at about 16ms. Code can be found at: https://github.com/louishust/sparkDemo/blob/master/src/main/java/DirectQueryTest.java So If my configuration for spark is wrong? How to optimise Spark to achieve the similar performance like direct access? Any idea will be appreciated!
Re: Spark-hive parquet schema evolution
Hi Lian, Sorry I'm new to Spark so I did not express myself very clearly. I'm concerned about the situation when let's say I have a Parquet table some partitions and I add a new column A to parquet schema and write some data with the new schema to a new partition in the table. If i'm not mistaken, if I do a sqlContext.read.parquet(table_path).printSchema() it will print the correct schema with new column A. But if I do a 'describe table' from SparkSQLCLI I won't see the new column being added. I understand that this is because Hive doesn't support schema evolution. So what is the best way to support CLI queries in this situation? Do I need to manually alter the table everytime the underlying schema changes? Thanks On Tue, Jul 21, 2015 at 4:37 PM, Cheng Lian lian.cs@gmail.com wrote: Hey Jerrick, What do you mean by schema evolution with Hive metastore tables? Hive doesn't take schema evolution into account. Could you please give a concrete use case? Are you trying to write Parquet data with extra columns into an existing metastore Parquet table? Cheng On 7/21/15 1:04 AM, Jerrick Hoang wrote: I'm new to Spark, any ideas would be much appreciated! Thanks On Sat, Jul 18, 2015 at 11:11 AM, Jerrick Hoang jerrickho...@gmail.com wrote: Hi all, I'm aware of the support for schema evolution via DataFrame API. Just wondering what would be the best way to go about dealing with schema evolution with Hive metastore tables. So, say I create a table via SparkSQL CLI, how would I deal with Parquet schema evolution? Thanks, J
Re: Spark-hive parquet schema evolution
I'm new to Spark, any ideas would be much appreciated! Thanks On Sat, Jul 18, 2015 at 11:11 AM, Jerrick Hoang jerrickho...@gmail.com wrote: Hi all, I'm aware of the support for schema evolution via DataFrame API. Just wondering what would be the best way to go about dealing with schema evolution with Hive metastore tables. So, say I create a table via SparkSQL CLI, how would I deal with Parquet schema evolution? Thanks, J
Spark-hive parquet schema evolution
Hi all, I'm aware of the support for schema evolution via DataFrame API. Just wondering what would be the best way to go about dealing with schema evolution with Hive metastore tables. So, say I create a table via SparkSQL CLI, how would I deal with Parquet schema evolution? Thanks, J
Re: Getting not implemented by the TFS FileSystem implementation
So, this has to do with the fact that 1.4 has a new way to interact with HiveMetastore, still investigating. Would really appreciate if anybody has any insights :) On Tue, Jul 14, 2015 at 4:28 PM, Jerrick Hoang jerrickho...@gmail.com wrote: Hi all, I'm upgrading from spark1.3 to spark1.4 and when trying to run spark-sql CLI. It gave an ```ava.lang.UnsupportedOperationException: Not implemented by the TFS FileSystem implementation``` exception. I did not get this error with 1.3 and I don't use any TFS FileSystem. Full stack trace is ```Exception in thread main java.lang.RuntimeException: java.lang.UnsupportedOperationException: Not implemented by the TFS FileSystem implementation at org.apache.hadoop.hive.ql.session.SessionState.start(SessionState.java:346) at org.apache.spark.sql.hive.client.ClientWrapper.init(ClientWrapper.scala:105) at sun.reflect.NativeConstructorAccessorImpl.newInstance0(Native Method) at sun.reflect.NativeConstructorAccessorImpl.newInstance(NativeConstructorAccessorImpl.java:62) at sun.reflect.DelegatingConstructorAccessorImpl.newInstance(DelegatingConstructorAccessorImpl.java:45) at java.lang.reflect.Constructor.newInstance(Constructor.java:408) at org.apache.spark.sql.hive.client.IsolatedClientLoader.liftedTree1$1(IsolatedClientLoader.scala:170) at org.apache.spark.sql.hive.client.IsolatedClientLoader.init(IsolatedClientLoader.scala:166) at org.apache.spark.sql.hive.HiveContext.metadataHive$lzycompute(HiveContext.scala:212) at org.apache.spark.sql.hive.HiveContext.metadataHive(HiveContext.scala:175) at org.apache.spark.sql.hive.HiveContext.setConf(HiveContext.scala:358) at org.apache.spark.sql.SQLContext$$anonfun$4.apply(SQLContext.scala:205) at org.apache.spark.sql.SQLContext$$anonfun$4.apply(SQLContext.scala:204) at scala.collection.Iterator$class.foreach(Iterator.scala:727) at scala.collection.AbstractIterator.foreach(Iterator.scala:1157) at scala.collection.IterableLike$class.foreach(IterableLike.scala:72) at scala.collection.AbstractIterable.foreach(Iterable.scala:54) at org.apache.spark.sql.SQLContext.init(SQLContext.scala:204) at org.apache.spark.sql.hive.HiveContext.init(HiveContext.scala:71) at org.apache.spark.sql.hive.thriftserver.SparkSQLEnv$.init(SparkSQLEnv.scala:53) at org.apache.spark.sql.hive.thriftserver.SparkSQLCLIDriver.init(SparkSQLCLIDriver.scala:248) at org.apache.spark.sql.hive.thriftserver.SparkSQLCLIDriver$.main(SparkSQLCLIDriver.scala:136) at org.apache.spark.sql.hive.thriftserver.SparkSQLCLIDriver.main(SparkSQLCLIDriver.scala) at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method) at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:62) at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43) at java.lang.reflect.Method.invoke(Method.java:483) at org.apache.spark.deploy.SparkSubmit$.org$apache$spark$deploy$SparkSubmit$$runMain(SparkSubmit.scala:664) at org.apache.spark.deploy.SparkSubmit$.doRunMain$1(SparkSubmit.scala:169) at org.apache.spark.deploy.SparkSubmit$.submit(SparkSubmit.scala:192) at org.apache.spark.deploy.SparkSubmit$.main(SparkSubmit.scala:111) at org.apache.spark.deploy.SparkSubmit.main(SparkSubmit.scala) Caused by: java.lang.UnsupportedOperationException: Not implemented by the TFS FileSystem implementation at org.apache.hadoop.fs.FileSystem.getScheme(FileSystem.java:214) at org.apache.hadoop.fs.FileSystem.loadFileSystems(FileSystem.java:2365) at org.apache.hadoop.fs.FileSystem.getFileSystemClass(FileSystem.java:2375) at org.apache.hadoop.fs.FileSystem.createFileSystem(FileSystem.java:2392) at org.apache.hadoop.fs.FileSystem.access$200(FileSystem.java:89) at org.apache.hadoop.fs.FileSystem$Cache.getInternal(FileSystem.java:2431) at org.apache.hadoop.fs.FileSystem$Cache.get(FileSystem.java:2413) at org.apache.hadoop.fs.FileSystem.get(FileSystem.java:368) at org.apache.hadoop.fs.FileSystem.get(FileSystem.java:167) at org.apache.hadoop.hive.ql.session.SessionState.start(SessionState.java:342) ... 31 more``` Thanks all
Getting not implemented by the TFS FileSystem implementation
Hi all, I'm upgrading from spark1.3 to spark1.4 and when trying to run spark-sql CLI. It gave an ```ava.lang.UnsupportedOperationException: Not implemented by the TFS FileSystem implementation``` exception. I did not get this error with 1.3 and I don't use any TFS FileSystem. Full stack trace is ```Exception in thread main java.lang.RuntimeException: java.lang.UnsupportedOperationException: Not implemented by the TFS FileSystem implementation at org.apache.hadoop.hive.ql.session.SessionState.start(SessionState.java:346) at org.apache.spark.sql.hive.client.ClientWrapper.init(ClientWrapper.scala:105) at sun.reflect.NativeConstructorAccessorImpl.newInstance0(Native Method) at sun.reflect.NativeConstructorAccessorImpl.newInstance(NativeConstructorAccessorImpl.java:62) at sun.reflect.DelegatingConstructorAccessorImpl.newInstance(DelegatingConstructorAccessorImpl.java:45) at java.lang.reflect.Constructor.newInstance(Constructor.java:408) at org.apache.spark.sql.hive.client.IsolatedClientLoader.liftedTree1$1(IsolatedClientLoader.scala:170) at org.apache.spark.sql.hive.client.IsolatedClientLoader.init(IsolatedClientLoader.scala:166) at org.apache.spark.sql.hive.HiveContext.metadataHive$lzycompute(HiveContext.scala:212) at org.apache.spark.sql.hive.HiveContext.metadataHive(HiveContext.scala:175) at org.apache.spark.sql.hive.HiveContext.setConf(HiveContext.scala:358) at org.apache.spark.sql.SQLContext$$anonfun$4.apply(SQLContext.scala:205) at org.apache.spark.sql.SQLContext$$anonfun$4.apply(SQLContext.scala:204) at scala.collection.Iterator$class.foreach(Iterator.scala:727) at scala.collection.AbstractIterator.foreach(Iterator.scala:1157) at scala.collection.IterableLike$class.foreach(IterableLike.scala:72) at scala.collection.AbstractIterable.foreach(Iterable.scala:54) at org.apache.spark.sql.SQLContext.init(SQLContext.scala:204) at org.apache.spark.sql.hive.HiveContext.init(HiveContext.scala:71) at org.apache.spark.sql.hive.thriftserver.SparkSQLEnv$.init(SparkSQLEnv.scala:53) at org.apache.spark.sql.hive.thriftserver.SparkSQLCLIDriver.init(SparkSQLCLIDriver.scala:248) at org.apache.spark.sql.hive.thriftserver.SparkSQLCLIDriver$.main(SparkSQLCLIDriver.scala:136) at org.apache.spark.sql.hive.thriftserver.SparkSQLCLIDriver.main(SparkSQLCLIDriver.scala) at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method) at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:62) at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43) at java.lang.reflect.Method.invoke(Method.java:483) at org.apache.spark.deploy.SparkSubmit$.org$apache$spark$deploy$SparkSubmit$$runMain(SparkSubmit.scala:664) at org.apache.spark.deploy.SparkSubmit$.doRunMain$1(SparkSubmit.scala:169) at org.apache.spark.deploy.SparkSubmit$.submit(SparkSubmit.scala:192) at org.apache.spark.deploy.SparkSubmit$.main(SparkSubmit.scala:111) at org.apache.spark.deploy.SparkSubmit.main(SparkSubmit.scala) Caused by: java.lang.UnsupportedOperationException: Not implemented by the TFS FileSystem implementation at org.apache.hadoop.fs.FileSystem.getScheme(FileSystem.java:214) at org.apache.hadoop.fs.FileSystem.loadFileSystems(FileSystem.java:2365) at org.apache.hadoop.fs.FileSystem.getFileSystemClass(FileSystem.java:2375) at org.apache.hadoop.fs.FileSystem.createFileSystem(FileSystem.java:2392) at org.apache.hadoop.fs.FileSystem.access$200(FileSystem.java:89) at org.apache.hadoop.fs.FileSystem$Cache.getInternal(FileSystem.java:2431) at org.apache.hadoop.fs.FileSystem$Cache.get(FileSystem.java:2413) at org.apache.hadoop.fs.FileSystem.get(FileSystem.java:368) at org.apache.hadoop.fs.FileSystem.get(FileSystem.java:167) at org.apache.hadoop.hive.ql.session.SessionState.start(SessionState.java:342) ... 31 more``` Thanks all
Re: Basic Spark SQL question
Well for adhoc queries you can use the CLI On Mon, Jul 13, 2015 at 5:34 PM, Ron Gonzalez zlgonza...@yahoo.com.invalid wrote: Hi, I have a question for Spark SQL. Is there a way to be able to use Spark SQL on YARN without having to submit a job? Bottom line here is I want to be able to reduce the latency of running queries as a job. I know that the spark sql default submission is like a job, but was wondering if it's possible to run queries like one would with a regular db like MySQL or Oracle. Thanks, Ron - To unsubscribe, e-mail: user-unsubscr...@spark.apache.org For additional commands, e-mail: user-h...@spark.apache.org
hive-site.xml spark1.3
Hi all, I'm having conf/hive-site.xml pointing to my Hive metastore but sparksql CLI doesn't pick it up. (copying the same conf/ files to spark1.4 and 1.2 works fine). Just wondering if someone has seen this before, Thanks
Re: SparkSQL 'describe table' tries to look at all records
Sorry all for not being clear. I'm using spark 1.4 and the table is a hive table, and the table is partitioned. On Sun, Jul 12, 2015 at 6:36 PM, Yin Huai yh...@databricks.com wrote: Jerrick, Let me ask a few clarification questions. What is the version of Spark? Is the table a hive table? What is the format of the table? Is the table partitioned? Thanks, Yin On Sun, Jul 12, 2015 at 6:01 PM, ayan guha guha.a...@gmail.com wrote: Describe computes statistics, so it will try to query the table. The one you are looking for is df.printSchema() On Mon, Jul 13, 2015 at 10:03 AM, Jerrick Hoang jerrickho...@gmail.com wrote: Hi all, I'm new to Spark and this question may be trivial or has already been answered, but when I do a 'describe table' from SparkSQL CLI it seems to try looking at all records at the table (which takes a really long time for big table) instead of just giving me the metadata of the table. Would appreciate if someone can give me some pointers, thanks! -- Best Regards, Ayan Guha
SparkSQL 'describe table' tries to look at all records
Hi all, I'm new to Spark and this question may be trivial or has already been answered, but when I do a 'describe table' from SparkSQL CLI it seems to try looking at all records at the table (which takes a really long time for big table) instead of just giving me the metadata of the table. Would appreciate if someone can give me some pointers, thanks!
