Robert Metzger created FLINK-2601:
-------------------------------------
Summary: IOManagerAsync may produce NPE during shutdown
Key: FLINK-2601
URL: https://issues.apache.org/jira/browse/FLINK-2601
Project: Flink
Issue Type: Bug
Components: Tests
Affects Versions: 0.10
Reporter: Robert Metzger
Priority: Minor
While analyzing a failed YARN test, I detected that it failed because it found
the following exception in the logs:
taskmanager-stderr:
{code}
Exception in thread "I/O manager shutdown hook" java.lang.NullPointerException
at
org.apache.flink.runtime.io.disk.iomanager.IOManagerAsync.shutdown(IOManagerAsync.java:144)
at
org.apache.flink.runtime.io.disk.iomanager.IOManager$1.run(IOManager.java:103)
{code}
taskmanager.log
{code}
18:45:00,812 INFO org.apache.flink.runtime.taskmanager.TaskManager
- Starting TaskManager actor
18:45:00,819 INFO org.apache.flink.runtime.io.network.netty.NettyConfig
- NettyConfig [server address:
testing-worker-linux-docker-56ee9bbf-3203-linux-2.prod.travis-ci.org/172.17.9.129,
server port: 38689, memory segment size (bytes): 32768, transport type: NIO,
number of server threads: 0 (use Netty's default), number of client threads: 0
(use Netty's default), server connect backlog: 0 (use Netty's default), client
connect timeout (sec): 120, send/receive buffer size (bytes): 0 (use Netty's
default)]
18:45:00,822 INFO org.apache.flink.runtime.taskmanager.TaskManager
- Messages between TaskManager and JobManager have a max timeout of 100000
milliseconds
18:45:00,825 INFO org.apache.flink.runtime.taskmanager.TaskManager
- Temporary file directory
'/home/travis/build/rmetzger/flink/flink-yarn-tests/target/flink-yarn-tests-fifo/flink-yarn-tests-fifo-localDir-nm-1_0/usercache/travis/appcache/application_1441046584836_0007':
total 15 GB, usable 7 GB (46.67% usable)
18:45:00,929 INFO org.apache.flink.runtime.io.network.buffer.NetworkBufferPool
- Allocated 64 MB for network buffer pool (number of memory segments: 2048,
bytes per segment: 32768).
18:45:01,186 INFO org.apache.flink.runtime.taskmanager.TaskManager
- Using 0.7 of the currently free heap space for Flink managed memory (236 MB).
18:45:01,755 INFO org.apache.flink.runtime.io.disk.iomanager.IOManager
- I/O manager uses directory
/home/travis/build/rmetzger/flink/flink-yarn-tests/target/flink-yarn-tests-fifo/flink-yarn-tests-fifo-localDir-nm-1_0/usercache/travis/appcache/application_1441046584836_0007/flink-io-1befed3c-89c5-4b5e-9043-1b92c4c047d4
for spill files.
18:45:01,831 ERROR org.apache.flink.yarn.appMaster.YarnTaskManagerRunner
- RECEIVED SIGNAL 15: SIGTERM
18:45:01,833 ERROR org.apache.flink.runtime.io.disk.iomanager.IOManagerAsync
- Error while shutting down IO Manager reader thread.
java.lang.NullPointerException
at
org.apache.flink.runtime.io.disk.iomanager.IOManagerAsync.shutdown(IOManagerAsync.java:133)
at
org.apache.flink.runtime.io.disk.iomanager.IOManager$1.run(IOManager.java:103)
18:45:01,841 INFO org.apache.flink.runtime.io.disk.iomanager.IOManager
- I/O manager removed spill file directory
/home/travis/build/rmetzger/flink/flink-yarn-tests/target/flink-yarn-tests-fifo/flink-yarn-tests-fifo-localDir-nm-1_0/usercache/travis/appcache/application_1441046584836_0007/flink-io-1befed3c-89c5-4b5e-9043-1b92c4c047d4
{code}
Looks like the TM is shutting down while still starting up. Hardening this
should be easy.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
