flink-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Vera Coberley <veracober...@gmail.com>
Subject Flink 1.2.0 - Flink Job restart config not using Flink Cluster restart config.
Date Wed, 28 Jun 2017 14:22:33 GMT
Hi all,

We are running Flink 1.2.0. Our flink-conf.yaml is configured to use a
default restart-strategy of fixed-delay, with 3 attempts:

restart-strategy: fixed-delay
restart-strategy.fixed-delay.attempts: 3

These settings are echoed by the GlobalConfiguration (see first set of log
statements). However, the job is submitted with a maxNumberRestartAttempts
of Max INT instead of 3 (see second set of log statements)

The job is enabled for checkpointing, and it does not have any job-specific
restart strategy defined:
StreamExecutionEnvironment env = StreamExecutionEnvironment.
getExecutionEnvironment();
env.enableCheckpointing(params.getLong("checkpoint.interval", 3000L));

I assumed the default restart configuration would carry over to the job. Am
I mistaken in my assumption, do I have a configuration error, or is this a
bug?

-- Vera

2017-06-27 19:52:11.288 [main] INFO
org.apache.flink.configuration.GlobalConfiguration  - Loading
configuration property: restart-strategy, fixed-delay
2017-06-27 19:52:11.288 [main] INFO
org.apache.flink.configuration.GlobalConfiguration  - Loading
configuration property: restart-strategy.fixed-delay.attempts, 3

2017-06-27 19:52:17.642 [flink-akka.actor.default-dispatcher-16] INFO
org.apache.flink.yarn.YarnJobManager  - Submitting job XYZ
2017-06-27 19:52:17.652 [flink-akka.actor.default-dispatcher-16] INFO
org.apache.flink.yarn.YarnJobManager  - Using restart strategy
FixedDelayRestartStrategy(maxNumberRestartAttempts=2147483647,
delayBetweenRestartAttempts=10000) for XYZ

Mime
View raw message