[
https://issues.apache.org/jira/browse/FLINK-14311?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17029661#comment-17029661
]
Gary Yao commented on FLINK-14311:
----------------------------------
I think the function that waits for the number of restarts will wait
indefinitely if {{fullRestarts}} becomes greater than
{{expected_num_restarts}}, which can happen if the the TMs startup time is slow.
{noformat}
function wait_for_restart_to_complete {
local base_num_restarts=$1
local jobid=$2
local current_num_restarts=${base_num_restarts}
local expected_num_restarts=$((current_num_restarts + 1))
echo "Waiting for restart to happen"
while ! [[ ${current_num_restarts} -eq ${expected_num_restarts} ]]; do
sleep 5
current_num_restarts=$(get_job_metric ${jobid} "fullRestarts")
if [[ -z ${current_num_restarts} ]]; then
current_num_restarts=${base_num_restarts}
fi
done
}
{noformat}
> Streaming File Sink end-to-end test failed on Travis
> ----------------------------------------------------
>
> Key: FLINK-14311
> URL: https://issues.apache.org/jira/browse/FLINK-14311
> Project: Flink
> Issue Type: Bug
> Components: Connectors / FileSystem, Tests
> Affects Versions: 1.11.0
> Reporter: Till Rohrmann
> Assignee: PengFei Li
> Priority: Critical
> Labels: test-stability
>
> The {{Streaming File Sink end-to-end test}} fails on Travis because it does
> not produce output for 10 minutes.
> https://api.travis-ci.org/v3/job/591992274/log.txt
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
