Re: Worker profiling on 1.2.1

Thanks Julien for providing the patch! Will take a look. 2018년 3월 27일 (화) 오후 10:46, Julien Nioche 님이 작성: > Done, see https://github.com/apache/storm/pull/2607 > > On 27 March 2018 at 11:17, Julien Nioche > wrote: > >> Hi Jungtaek >>

Re: Worker profiling on 1.2.1

Done, see https://github.com/apache/storm/pull/2607 On 27 March 2018 at 11:17, Julien Nioche wrote: > Hi Jungtaek > > I had to move to a different server and can't reproduce the same issue > anymore. I can now see the 'start' option on the UI. > > I had to modify

Long command line

Hi, when Storm runs, it sets up the command line with full path of all jars to be included in the classpath. How can I simplify the command line by specifying directories and wildcard where the jars reside? In doing that, does Storm care about the order of the jars in cp?

Re: Worker profiling on 1.2.1

Hi Jungtaek I had to move to a different server and can't reproduce the same issue anymore. I can now see the 'start' option on the UI. I had to modify the *flight.bash* script as $JAVA_HOME is undefined, this yields */usr/share/apache-storm-1.2.1/bin/flight.bash: line 62: /bin/jstack: No such

答复: Storm kafka Spout Stuck When Kafka leader is Down

When I use Storm-Kafka-Client 1.2.1, kafka-client 0.10.2.1, kafak-server 1.0.0. leader down, kafka spout went well. But When I use Storm-kafka-client 1.2.1,kafka-client1.0.0 (the same as kafka sever version),kafka –server 1.0.0. Kafka spout stuck when kafka leader down. Is

Re: Storm hung issue

Also I have implemented automatic backpressure mechanism available in latest storm version. Optimized slow processing bolt code to process faster. On Tue, Mar 27, 2018, 2:30 PM Ajeesh wrote: > Hi, I have checked the thread dump, I couldn't find any blocked threads. I >

Re: Storm hung issue

Hi, I have checked the thread dump, I couldn't find any blocked threads. I solved the issue by upgrading Kafka, Storm and Zookeeper applications to latest version. Also I was batching few tuples for insertion into database and so I was not emitting these tuples until insertion, modified logic to

Re: New supervisor is not able to join cluster (storm 1.0.2)

Hi, try to make the telnet command to connect with nimbus machine, it may solve the problem. On Fri, Mar 23, 2018, 4:09 PM pwozniak wrote: > Hi, > > I have storm cluster that basically works fine. Now I have to add two > more supervisors to this cluster and when I'm

Re: Storm kafka Spout Stuck When Kafka leader is Down

Hi, Use the storm-kafka-client version same as Kafka Server version On Tue, Mar 27, 2018, 2:04 PM Han Jing wrote: > Hi All， > >I’m using Storm-Kafka-Client 1.2.1 to read from Kafka sever(1.0.0, > 1.0.1).When Kafka topic leader progress is down, Storm Kafka

Storm kafka Spout Stuck When Kafka leader is Down

Hi All， I’m using Storm-Kafka-Client 1.2.1 to read from Kafka sever(1.0.0, 1.0.1).When Kafka topic leader progress is down, Storm Kafka Spout is stuck, there is no responseon UI website ,even kakfa topic leader is alter to another broker, It’s still stuck, until restart the kafka server

Re: Worker profiling on 1.2.1

Julien, could you check whether 'profilingAndDebuggingCapable' and 'profileActionEnabled' are true at the result of component API? You can check it via opening developer tool in browser, and open component page, open network tab, and find the request which path is like