Jorge Machado created SPARK-26324: ------------------------------------- Summary: Spark submit does not work with messos over ssl Key: SPARK-26324 URL: https://issues.apache.org/jira/browse/SPARK-26324 Project: Spark Issue Type: Bug Components: Spark Submit Affects Versions: 2.4.0 Reporter: Jorge Machado
Hi guys, I was trying to run the examples on a mesos cluster that uses https. I tried with rest endpoint: {code:java} ./spark-submit --class org.apache.spark.examples.SparkPi --master mesos://<mesos_master_with_https>:5050 --conf spark.master.rest.enabled=true --deploy-mode cluster --supervise --executor-memory 10G --total-executor-cores 100 ../examples/jars/spark-examples_2.11-2.4.0.jar 1000 {code} The error that I get on the host where I started the spark-submit is: {code:java} 2018-12-10 15:08:39 WARN NativeCodeLoader:62 - Unable to load native-hadoop library for your platform... using builtin-java classes where applicable 2018-12-10 15:08:39 INFO RestSubmissionClient:54 - Submitting a request to launch an application in mesos://<mesos_master_with_https>:5050. 2018-12-10 15:08:39 WARN RestSubmissionClient:66 - Unable to connect to server mesos://<mesos_master_with_https>:5050. Exception in thread "main" org.apache.spark.deploy.rest.SubmitRestConnectionException: Unable to connect to server at org.apache.spark.deploy.rest.RestSubmissionClient$$anonfun$createSubmission$3.apply(RestSubmissionClient.scala:104) at org.apache.spark.deploy.rest.RestSubmissionClient$$anonfun$createSubmission$3.apply(RestSubmissionClient.scala:86) at scala.collection.TraversableLike$WithFilter$$anonfun$foreach$1.apply(TraversableLike.scala:733) at scala.collection.IndexedSeqOptimized$class.foreach(IndexedSeqOptimized.scala:33) at scala.collection.mutable.ArrayOps$ofRef.foreach(ArrayOps.scala:186) at scala.collection.TraversableLike$WithFilter.foreach(TraversableLike.scala:732) at org.apache.spark.deploy.rest.RestSubmissionClient.createSubmission(RestSubmissionClient.scala:86) at org.apache.spark.deploy.rest.RestSubmissionClientApp.run(RestSubmissionClient.scala:443) at org.apache.spark.deploy.rest.RestSubmissionClientApp.start(RestSubmissionClient.scala:455) at org.apache.spark.deploy.SparkSubmit.org$apache$spark$deploy$SparkSubmit$$runMain(SparkSubmit.scala:849) at org.apache.spark.deploy.SparkSubmit.doRunMain$1(SparkSubmit.scala:167) at org.apache.spark.deploy.SparkSubmit.submit(SparkSubmit.scala:195) at org.apache.spark.deploy.SparkSubmit.doSubmit(SparkSubmit.scala:86) at org.apache.spark.deploy.SparkSubmit$$anon$2.doSubmit(SparkSubmit.scala:924) at org.apache.spark.deploy.SparkSubmit$.main(SparkSubmit.scala:933) at org.apache.spark.deploy.SparkSubmit.main(SparkSubmit.scala) Caused by: org.apache.spark.deploy.rest.SubmitRestConnectionException: Unable to connect to server at org.apache.spark.deploy.rest.RestSubmissionClient.readResponse(RestSubmissionClient.scala:281) at org.apache.spark.deploy.rest.RestSubmissionClient.org$apache$spark$deploy$rest$RestSubmissionClient$$postJson(RestSubmissionClient.scala:225) at org.apache.spark.deploy.rest.RestSubmissionClient$$anonfun$createSubmission$3.apply(RestSubmissionClient.scala:90) ... 15 more Caused by: java.net.SocketException: Connection reset {code} I'm pretty sure this is because of the hardcoded http:// here: {code:java} RestSubmissionClient.scala /** Return the base URL for communicating with the server, including the protocol version. */ private def getBaseUrl(master: String): String = { var masterUrl = master supportedMasterPrefixes.foreach { prefix => if (master.startsWith(prefix)) { masterUrl = master.stripPrefix(prefix) } } masterUrl = masterUrl.stripSuffix("/") s"http://$masterUrl/$PROTOCOL_VERSION/submissions" <--- hardcoded http } {code} Then I tried without the _--deploy-mode cluster_ and I get: {code:java} ./spark-submit --class org.apache.spark.examples.SparkPi --master mesos://<server_using_https>:5050 --supervise --executor-memory 10G --total-executor-cores 100 ../examples/jars/spark-examples_2.11-2.4.0.jar 1000 {code} On the spark console I get: {code:java} 2018-12-10 15:01:05 INFO SparkUI:54 - Bound SparkUI to 0.0.0.0, and started at http://_host:4040 2018-12-10 15:01:05 INFO SparkContext:54 - Added JAR file:/home/<user>/spark-2.4.0-bin-hadoop2.7/bin/../examples/jars/spark-examples_2.11-2.4.0.jar at spark://_host:35719/jars/spark-examples_2.11-2.4.0.jar with timestamp 1544450465799 I1210 15:01:05.963078 37943 sched.cpp:232] Version: 1.3.2 I1210 15:01:05.966814 37911 sched.cpp:336] New master detected at master@53.54.195.251:5050 I1210 15:01:05.967010 37911 sched.cpp:352] No credentials provided. Attempting to register without authentication E1210 15:01:05.967347 37942 process.cpp:2455] Failed to shutdown socket with fd 307, address 53.54.195.251:45206: Transport endpoint is not connected E1210 15:01:05.968212 37942 process.cpp:2369] Failed to shutdown socket with fd 307, address 53.54.195.251:45212: Transport endpoint is not connected E1210 15:01:05.969405 37942 process.cpp:2455] Failed to shutdown socket with fd 307, address 53.54.195.251:45222: Transport endpoint is not connected{code} On Mesos I get: {code:java} E1210 15:01:06.665076 2633 process.cpp:956] Failed to accept socket: Failed accept: connection error: error:1407609C:SSL routines:SSL23_GET_CLIENT_HELLO:http request {code} I could not found any documentation on how to connect both. Do I need to setup some acls on java_opts for the ssl ? -- This message was sent by Atlassian JIRA (v7.6.3#76005) --------------------------------------------------------------------- To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org For additional commands, e-mail: issues-h...@spark.apache.org