mirror of
https://github.com/rsyslog/rsyslog.git
synced 2025-12-19 09:10:42 +01:00
Andre Lorbach
8ecf0aeb34
Also saving failed msgs now when kafka produce fails. No message loss should be possible anymore. Also removed old kafka tests and readded new kafka tests. If kafka instances cannot be started (happens from time to time when kafka/zookepper sync fails), tests will be skipped instead of failing. Closes https://github.com/rsyslog/rsyslog/issues/1559 Closes https://github.com/rsyslog/rsyslog/issues/1584 Closes https://github.com/rsyslog/rsyslog/issues/1515 May fix https://github.com/rsyslog/rsyslog/issues/1230 Closes https://github.com/rsyslog/rsyslog/issues/1052
69 lines
2.3 KiB
Bash
Executable File
69 lines
2.3 KiB
Bash
Executable File
#!/bin/bash
|
|
# added 2017-05-18 by alorbach
|
|
# This test only tests what happens when kafka cluster fails
|
|
# This file is part of the rsyslog project, released under ASL 2.0
|
|
export TESTMESSAGES=1000
|
|
export TESTMESSAGESFULL=2000
|
|
|
|
echo ===============================================================================
|
|
echo \[sndrcv_kafka_fail.sh\]: Create kafka/zookeeper instance and static topic
|
|
. $srcdir/diag.sh download-kafka
|
|
. $srcdir/diag.sh stop-zookeeper
|
|
. $srcdir/diag.sh stop-kafka
|
|
. $srcdir/diag.sh start-zookeeper
|
|
. $srcdir/diag.sh start-kafka
|
|
. $srcdir/diag.sh create-kafka-topic 'static' '.dep_wrk' '22181'
|
|
|
|
echo \[sndrcv_kafka_fail.sh\]: Give Kafka some time to process topic create ...
|
|
sleep 5
|
|
|
|
echo \[sndrcv_kafka_fail.sh\]: Stopping kafka cluster instance
|
|
. $srcdir/diag.sh stop-kafka
|
|
|
|
echo \[sndrcv_kafka_fail.sh\]: Starting receiver instance [omkafka]
|
|
export RSYSLOG_DEBUGLOG="log"
|
|
. $srcdir/diag.sh init
|
|
. $srcdir/diag.sh startup sndrcv_kafka_rcvr.conf
|
|
. $srcdir/diag.sh wait-startup
|
|
|
|
echo \[sndrcv_kafka_fail.sh\]: Starting sender instance [imkafka]
|
|
export RSYSLOG_DEBUGLOG="log2"
|
|
. $srcdir/diag.sh startup sndrcv_kafka_sender.conf 2
|
|
. $srcdir/diag.sh wait-startup 2
|
|
|
|
echo \[sndrcv_kafka_fail.sh\]: Inject messages into rsyslog sender instance
|
|
. $srcdir/diag.sh tcpflood -m$TESTMESSAGES -i1
|
|
|
|
echo \[sndrcv_kafka_fail.sh\]: Starting kafka cluster instance
|
|
. $srcdir/diag.sh start-kafka
|
|
|
|
echo \[sndrcv_kafka_fail.sh\]: Sleep to give rsyslog instances time to process data ...
|
|
sleep 5
|
|
|
|
echo \[sndrcv_kafka_fail.sh\]: Inject messages into rsyslog sender instance
|
|
. $srcdir/diag.sh tcpflood -m$TESTMESSAGES -i1001
|
|
|
|
echo \[sndrcv_kafka_fail.sh\]: Sleep to give rsyslog sender time to send data ...
|
|
sleep 5
|
|
|
|
echo \[sndrcv_kafka_fail.sh\]: Stopping sender instance [imkafka]
|
|
. $srcdir/diag.sh shutdown-when-empty 2
|
|
. $srcdir/diag.sh wait-shutdown 2
|
|
|
|
echo \[sndrcv_kafka_fail.sh\]: Sleep to give rsyslog receiver time to receive data ...
|
|
sleep 5
|
|
|
|
echo \[sndrcv_kafka_fail.sh\]: Stopping receiver instance [omkafka]
|
|
. $srcdir/diag.sh shutdown-when-empty
|
|
. $srcdir/diag.sh wait-shutdown
|
|
|
|
# Do the final sequence check
|
|
. $srcdir/diag.sh seq-check 1 $TESTMESSAGESFULL -d
|
|
|
|
echo \[sndrcv_kafka_fail.sh\]: stop kafka instance
|
|
. $srcdir/diag.sh delete-kafka-topic 'static' '.dep_wrk' '22181'
|
|
. $srcdir/diag.sh stop-kafka
|
|
|
|
# STOP ZOOKEEPER in any case
|
|
. $srcdir/diag.sh stop-zookeeper
|