[GitHub] [incubator-druid] clintropolis opened a new pull request #8342: use Number instead of long for response context

[GitBox <gi...@apache.org>]

clintropolis opened a new pull request #8342: use Number instead of long for response context
URL: https://github.com/apache/incubator-druid/pull/8342
 
 
   ### Description
   This PR fixes an issue with scan queries introduced by #8157, which I believe is caused by variation in JSON serde of the response context, and switches `NUM_SCANNED_ROWS` and `CPU_CONSUMED_NANOS` to cast to `Number` and use `Number.longValue` instead of casting to `long` primitive, to avoid errors of the form:
   
   ```
   2019-08-19T22:56:36,180 WARN [HttpClient-Netty-Worker-3] org.apache.druid.java.util.http.client.NettyHttpClient - [POST http://localhost:8084/druid/v2/] Exception thrown while processing message, closing channel.
   java.lang.ClassCastException: java.lang.Integer cannot be cast to java.lang.Long
       at org.apache.druid.query.context.ResponseContext$Key.lambda$static$3(ResponseContext.java:153) ~[classes/:?]
       at java.util.concurrent.ConcurrentHashMap.merge(ConcurrentHashMap.java:1990) ~[?:1.8.0_192]
       at org.apache.druid.query.context.ResponseContext.add(ResponseContext.java:308) ~[classes/:?]
       at org.apache.druid.query.context.ResponseContext.lambda$merge$2(ResponseContext.java:319) ~[classes/:?]
       at java.util.HashMap.forEach(HashMap.java:1289) ~[?:1.8.0_192]
       at org.apache.druid.query.context.ResponseContext.merge(ResponseContext.java:317) ~[classes/:?]
       at org.apache.druid.client.DirectDruidClient$1.handleResponse(DirectDruidClient.java:233) ~[classes/:?]
       at org.apache.druid.java.util.http.client.NettyHttpClient$1.messageReceived(NettyHttpClient.java:224) [classes/:?]
       at org.jboss.netty.channel.SimpleChannelUpstreamHandler.handleUpstream(SimpleChannelUpstreamHandler.java:70) [netty-3.10.6.Final.jar:?]
       at org.jboss.netty.channel.DefaultChannelPipeline.sendUpstream(DefaultChannelPipeline.java:564) [netty-3.10.6.Final.jar:?]
       at org.jboss.netty.channel.DefaultChannelPipeline$DefaultChannelHandlerContext.sendUpstream(DefaultChannelPipeline.java:791) [netty-3.10.6.Final.jar:?]
   ```
   
   which result eventually in the broker failing to return a result:
   ```
   org.apache.druid.query.QueryInterruptedException: query[6cc5653a-c9b8-49e3-ad9c-94c14a4a07af] url[http://localhost:8084/druid/v2/] timed out or max bytes limit reached.
   ```
   
   This issue is repeatable by having a broker and 2 historical servers with the example wikipedia batch dataset, and issuing a `SELECT * FROM "wikipedia"`.
   
   After this fix the query successfully returns results.
   
   <hr>
   
   This PR has:
   - [x] been self-reviewed.
   - [x] been tested in a test Druid cluster running on my laptop.
   

----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
users@infra.apache.org


With regards,
Apache Git Services

---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscribe@druid.apache.org
For additional commands, e-mail: commits-help@druid.apache.org