you can use 3772 as long as it is not already in use. you can try netstat -ant | grep 3772 before you start the topo and should be no references using it. After you submit the topo should see a LISTENer if it starts correctly.
But also I think storm.yaml supports only single drpc port so doubt you can submit 2 topos trying diff drpc ports. I would think all active topologies have to use common port On Wed, Jul 12, 2017 at 10:00 AM, sam mohel <[email protected]> wrote: > yes i tried another topology using drpc and worked except this i'm trying > it now . although it worked in local mode , but same configuration of > distributed for first topology not work with it . i will try 127.0.0.1 , i > used 3772 with first topology and worked can i use it again with the second > or changed it to 3774 ? > > Really appreciate your time > > On Wed, Jul 12, 2017 at 3:49 PM, J.R. Pauley <[email protected]> wrote: > >> you really have 192.168.x.x literally? I would think that would never >> work. Have you tried 127.0.0.1 to see if it makes any diff? Have you >> verified the port is open and listening? >> >> On Wed, Jul 12, 2017 at 9:16 AM, sam mohel <[email protected]> wrote: >> >>> i set it in the code >>> Config conf = new Config(); >>> List<String> dprcServers = new ArrayList<String>(); >>> dprcServers.add("192.168.x.x"); >>> conf.put(Config.DRPC_SERVERS, dprcServers); >>> conf.put(Config.DRPC_PORT, 3772); >>> >>> LocalDRPC drpc = null; >>> StormSubmitter.submitTopology(args[0], conf, buildTopology(drpc)); >>> DRPCClient client=new DRPCClient("192.168.x.x",3772); >>> >>> >>> On Wed, Jul 12, 2017 at 2:31 PM, J.R. Pauley <[email protected]> wrote: >>> >>>> How come no drpc.servers config shown in storm.yaml? >>>> >>>> On Wed, Jul 12, 2017 at 8:22 AM, sam mohel <[email protected]> wrote: >>>> >>>>> thanks for replying Nazar . but how can i solve my problem " >>>>> DRPCExecutionException(msg:Request timed out)" ? >>>>> Should i increase or decrease time of drpc server or client ? >>>>> >>>>> >>>>> >>>>> On Wed, Jul 12, 2017 at 11:16 AM, Nazar Kushpir < >>>>> [email protected]> wrote: >>>>> >>>>>> Sam, >>>>>> I don't see any gc-related info in your errors, but there is a >>>>>> "java.io.IOException: >>>>>> Connection reset by peer" message, that can be related to some >>>>>> network problems. >>>>>> As for GC and more RAM: storm has an option >>>>>> "topology.max.spout.pending" which allows you to limit the number of >>>>>> tuples >>>>>> in topology at any point of time (by default it has no limit) - it helped >>>>>> me to overcome "out of memory" errors, so it may help you too, since it >>>>>> isn't set in your "storm.yaml" file. >>>>>> >>>>>> On Tue, Jul 11, 2017 at 10:20 PM, sam mohel <[email protected]> >>>>>> wrote: >>>>>> >>>>>>> How can I figure the problem ? Is there any other place I can post >>>>>>> my problem ? >>>>>>> >>>>>>> On Tuesday, July 11, 2017, sam mohel <[email protected]> wrote: >>>>>>> > Is there any help please? >>>>>>> > >>>>>>> > On Tuesday, July 11, 2017, sam mohel <[email protected]> wrote: >>>>>>> >> thanks for replying and for this clarification . >>>>>>> >> here my full error and i hope some can help >>>>>>> >> I'm using apache-stom-0.9.6 with jdk 1.7 and zookeeper-3.4.6 . i >>>>>>> submitted my trident topology but got this in terminal >>>>>>> >> [main] INFO backtype.storm.StormSubmitter - Finished submitting >>>>>>> topology: top >>>>>>> >> Exception in thread "main" DRPCExecutionException(msg:Request >>>>>>> timed out) >>>>>>> >> at backtype.storm.generated.DistributedRPC$execute_result.read( >>>>>>> DistributedRPC.java:904) >>>>>>> >> at org.apache.thrift7.TServiceClient.receiveBase(TServiceClient >>>>>>> .java:78) >>>>>>> >> at backtype.storm.generated.DistributedRPC$Client.recv_execute( >>>>>>> DistributedRPC.java:92) >>>>>>> >> at backtype.storm.generated.DistributedRPC$Client.execute(Distr >>>>>>> ibutedRPC.java:78) >>>>>>> >> at backtype.storm.utils.DRPCClient.execute(DRPCClient.java:71) >>>>>>> >> at trident.mytopology.main(mytopology.java:319) >>>>>>> >> in drpc.log file >>>>>>> >> [INFO] Starting Distributed RPC servers... >>>>>>> >> 2017-07-11T08:10:44.139+0200 o.a.t.s.TNonblockingServer [WARN] >>>>>>> Got an IOException in internalRead! >>>>>>> >> java.io.IOException: Connection reset by peer >>>>>>> >> at sun.nio.ch.FileDispatcherImpl.read0(Native Method) >>>>>>> ~[na:1.7.0_121] >>>>>>> >> at sun.nio.ch.SocketDispatcher.read(SocketDispatcher.java:39) >>>>>>> ~[na:1.7.0_121] >>>>>>> >> at sun.nio.ch.IOUtil.readIntoNativeBuffer(IOUtil.java:223) >>>>>>> ~[na:1.7.0_121] >>>>>>> >> at sun.nio.ch.IOUtil.read(IOUtil.java:197) ~[na:1.7.0_121] >>>>>>> >> at sun.nio.ch.SocketChannelImpl.read(SocketChannelImpl.java:384) >>>>>>> ~[na:1.7.0_121] >>>>>>> >> at >>>>>>> >> org.apache.thrift7.transport.TNonblockingSocket.read(TNonblockingSocket.java:141) >>>>>>> ~[storm-core-0.9.6.jar:0.9.6] >>>>>>> >> at org.apache.thrift7.server.TNonblockingServer$FrameBuffer.int >>>>>>> ernalRead(TNonblockingServer.java:669) [storm-core-0.9.6.jar:0.9.6] >>>>>>> >> at >>>>>>> >> org.apache.thrift7.server.TNonblockingServer$FrameBuffer.read(TNonblockingServer.java:458) >>>>>>> [storm-core-0.9.6.jar:0.9.6] >>>>>>> >> at org.apache.thrift7.server.TNonblockingServer$SelectThread.ha >>>>>>> ndleRead(TNonblockingServer.java:359) [storm-core-0.9.6.jar:0.9.6] >>>>>>> >> at org.apache.thrift7.server.TNonblockingServer$SelectThread.se >>>>>>> lect(TNonblockingServer.java:304) [storm-core-0.9.6.jar:0.9.6] >>>>>>> >> at >>>>>>> >> org.apache.thrift7.server.TNonblockingServer$SelectThread.run(TNonblockingServer.java:243) >>>>>>> [storm-core-0.9.6.jar:0.9.6] >>>>>>> >> 2017-07-11T08:19:39.090+0200 b.s.d.drpc [WARN] Timeout DRPC >>>>>>> request id: 1 start at 1499753374 >>>>>>> >> nimbus.log file didn't contain any errors or warning >>>>>>> >> supervisor.log file contain still hasn't started >>>>>>> >> storm.yaml >>>>>>> >> storm.zookeeper.servers: >>>>>>> >> - "192.168.x.x" >>>>>>> >> >>>>>>> >> nimbus.host : "192.168.x.x" >>>>>>> >> storm.local.dir: "/var/storm" >>>>>>> >> >>>>>>> >> supervisor.childopts: "-Xmx1024m -XX:MaxPermSize=512m" >>>>>>> >> worker.childopts: "-Xmx2048m -XX:MaxPermSize=512m" >>>>>>> >> nimbus.childopts: "-Xmx2048m -XX:MaxPermSize=512m" >>>>>>> >> ui.port: 8080 >>>>>>> >> storm.zookeeper.session.timeout: 40000 >>>>>>> >> storm.zookeeper.connection.timeout: 30000 >>>>>>> >> nimbus.task.timeout.secs: 600 >>>>>>> >> >>>>>>> >> >>>>>>> >> On Tue, Jul 4, 2017 at 7:42 AM, Navin Ipe < >>>>>>> [email protected]> wrote: >>>>>>> >>> >>>>>>> >>> :-) There are no demands here, dear Sam. Just requests for >>>>>>> information so that we can help you better. >>>>>>> >>> I haven't had to deal with DRPC, so couldn't help you with that, >>>>>>> which is why I had also mentioned that the others on this forum will be >>>>>>> able to help you if you provided more info. >>>>>>> >>> This is a general procedure in every forum and in real life too. >>>>>>> The help you get is directly proportional with the amount of quality >>>>>>> info >>>>>>> you provide. >>>>>>> >>> >>>>>>> >>> To overcome a GC problem you either have to minimise the amount >>>>>>> of memory your program uses (consider restructuring your datastructures >>>>>>> or >>>>>>> using primitive datatypes) or you increase the amount of RAM and alter >>>>>>> Storm's configuration to be able to recognize the newly available >>>>>>> increase >>>>>>> in RAM. This is, if it is indeed a memory allocation problem. >>>>>>> >>> >>>>>>> >>> On Tue, Jul 4, 2017 at 2:08 AM, sam mohel <[email protected]> >>>>>>> wrote: >>>>>>> >>>> >>>>>>> >>>> I wrote before another post and you demand from me to write my >>>>>>> configurations and some details and I wrote it but you didn't help !! So >>>>>>> thanks for that because other people helped and replied . Now this post >>>>>>> in >>>>>>> general how can I overcome GC problem ? What are the things that I >>>>>>> should >>>>>>> concentrate on it ? Again my question is on general >>>>>>> >>>> >>>>>>> >>>> On Monday, July 3, 2017, Navin Ipe < >>>>>>> [email protected]> wrote: >>>>>>> >>>> > Think about this Sam. If some stranger wrote what you wrote, >>>>>>> and you tried understanding their problem looking at only what they >>>>>>> wrote, >>>>>>> would you be able to figure out anything at all? >>>>>>> >>>> > You haven't provided enough information for us to help you. >>>>>>> >>>> > What is the exact error you are encountering? Paste it here. >>>>>>> >>>> > How much memory have you allocated for storm (in the >>>>>>> settings)? >>>>>>> >>>> > How much of memory is available for storm to use (other apps >>>>>>> also use memory)? >>>>>>> >>>> > Are you trying to submit two topologies, each of which need >>>>>>> 6GB RAM on an 8GB laptop? >>>>>>> >>>> > Have you considered using online servers? >>>>>>> >>>> > >>>>>>> >>>> > Plenty of other things like this. Tell us exactly what the >>>>>>> problem is, what you tried to solve it, what you googled to resolve it >>>>>>> before asking us. This is how forums work. If you want people to help >>>>>>> you, >>>>>>> give them the right info and respect their time. >>>>>>> >>>> > >>>>>>> >>>> > >>>>>>> >>>> > On Sat, Jul 1, 2017 at 4:40 AM, sam mohel < >>>>>>> [email protected]> wrote: >>>>>>> >>>> >> >>>>>>> >>>> >> I hope can find any help and many thanks for that >>>>>>> >>>> >> I have problem with GC "garbage collector " and this is the >>>>>>> second time i face this problem as my laptop with RAM 6 GB and it didn't >>>>>>> work with the my topology so i increased my RAM to be 8 GB to overcome >>>>>>> this and fixed . Now i want to submit another one but RAM is not enough >>>>>>> , >>>>>>> How can i overcome this problem ? Can i fix it instead of increasing >>>>>>> size >>>>>>> of RAM ? >>>>>>> >>>> > >>>>>>> >>>> > >>>>>>> >>>> > -- >>>>>>> >>>> > Regards, >>>>>>> >>>> > Navin >>>>>>> >>> >>>>>>> >>> >>>>>>> >>> -- >>>>>>> >>> Regards, >>>>>>> >>> Navin >>>>>>> >> >>>>>>> >>>>>> >>>>>> >>>>> >>>> >>> >> >
