dataDir and dataLogDir Swapped about zookeeper-docker HOT 10 CLOSED

https://issues.apache.org/jira/browse/ZOOKEEPER-2960

from zookeeper-docker.

@zanes2016 thanks for you report. Is this reproducable with Zookeeper running on the host instead of Docker?

from zookeeper-docker.

@31z4 I haven't tried that yet since we run everything in docker. If it helps I can try to install/configure zookeeper to run on an Ubuntu 16.04 host.

from zookeeper-docker.

Yeah, that would be really helpful!

from zookeeper-docker.

@zanes2016 I cannot reproduce this with zookeeper:3.4.11 docker image.

# echo conf | nc localhost 2181
clientPort=2181
dataDir=/data/version-2
dataLogDir=/datalog/version-2
tickTime=2000
maxClientCnxns=60
minSessionTimeout=4000
maxSessionTimeout=40000
serverId=0

# cat /conf/zoo.cfg
clientPort=2181
dataDir=/data
dataLogDir=/datalog
tickTime=2000
initLimit=5
syncLimit=2

# ls /datalog/version-2
log.1

# ls /data
myid       version-2

from zookeeper-docker.

Isn't reproduced with zookeeper:3.5.

# ls /datalog/version-2/
log.100000001

# ls /data
myid       version-2

Also, it seems that you were running 3.5 instead of 3.4 because of the log file name.

from zookeeper-docker.

3.3 is also fine.

from zookeeper-docker.

I've noticed that I only see this when run in HA/cluster mode (not single node). I'm also deploying this with Ansible, but that would be surprising if the docker_container Ansible module was causing this issue. And I am certain that I've been specifying zookeeper 3.4.

from zookeeper-docker.

Would be great to know more about how exactly do you run it. If you use docker-compose or swarm mode, please show your stack.yml.

from zookeeper-docker.

Fixed in 3.4.12.

from zookeeper-docker.