[
https://issues.apache.org/jira/browse/HBASE-19901?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
stack updated HBASE-19901:
--------------------------
Assignee: stack
Status: Patch Available (was: Open)
See if this works:
diff --git a/dev-support/hbase-personality.sh b/dev-support/hbase-personality.sh
index e24930d51f..80d25866a3 100755
--- a/dev-support/hbase-personality.sh
+++ b/dev-support/hbase-personality.sh
@@ -64,6 +64,11 @@ function personality_globals
# Override the maven options
MAVEN_OPTS="${MAVEN_OPTS:-"-Xmx3100M"}"
+
+ # Yetus 0.7.0 enforces limits.
+ # Default proclimit is 1000. Up it.
+ # We seem to use close to 3k.
+ PROCLIMIT=5000
}
## @description Parse extra arguments required by personalities, if any.
diff --git a/dev-support/hbase_nightly_yetus.sh
b/dev-support/hbase_nightly_yetus.sh
index 4e67354073..3e6fc7c97d 100755
--- a/dev-support/hbase_nightly_yetus.sh
+++ b/dev-support/hbase_nightly_yetus.sh
@@ -67,6 +67,7 @@
YETUS_ARGS=("--whitespace-tabs-ignore-list=${WHITESPACE_IGNORE_LIST}" "${YETUS_A
YETUS_ARGS=("--sentinel" "${YETUS_ARGS[@]}")
YETUS_ARGS=("--branch=${BRANCH_NAME}" "${YETUS_ARGS[@]}")
YETUS_ARGS=("--tests-filter=${TESTS_FILTER}" "${YETUS_ARGS[@]}")
+YETUS_ARGS=("--proclimit=${PROCLIMIT}" "${YETUS_ARGS[@]}")
# Currently, flaky list is calculated only for master branch.
UNDERSCORED_BRANCH_NAME=$(echo ${BRANCH_NAME} | tr '.-' '_')
> Up yetus proclimit on nightlies
> -------------------------------
>
> Key: HBASE-19901
> URL: https://issues.apache.org/jira/browse/HBASE-19901
> Project: HBase
> Issue Type: Bug
> Reporter: stack
> Assignee: stack
> Priority: Major
> Attachments: HBASE-19901.master.001.patch
>
>
> We're on 0.7.0 now which enforces limits meant to protect against runaway
> processes. Default is 1000 procs. HBase test runs seem to consume almost 4k.
> Up our proclimit.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
