# Rancher High Availability **URL:** \ **Category:** Rancher 1.x\ **Created:** [November 8, 2016, 12:00pm UTC](https://forums.suse.com/t/rancher-high-availability/4554 "2016-11-08T12:00:15Z")\ **Posts on this page:** 4\ **Page:** 1
**Author:** ![phooper0001](https://avatars.discourse-cdn.com/v4/letter/p/b77776/32.png) [@phooper0001](https://forums.suse.com/u/phooper0001)\ **Post date:** [November 8, 2016, 12:00pm UTC](https://forums.suse.com/t/rancher-high-availability/4554/1 "2016-11-08T12:00:15Z")
Hi, I’m having real troubles getting Rancher HA running ( 3 Hosts ) I have 1. All ports open both ingress and egress ( 0.0.0.0/0 ) on all hosts ( hosts are in the SAME AWS region but separate AZs ) 2. External database which HAS connections established 3. 16GB RAM on the hosts 4. 50GB volumes on the hosts 5. I’ve followed the specific documentation around setting up HA - have run the generated script against the latest container and keep seeing errors : time=“2016-11-08T11:52:12Z” level=info msg=“Waiting for 3 host(s) to be active” component=cert time=“2016-11-08T11:52:14Z” level=info msg=“Waiting for 3 host(s) to be active” component=cert time=“2016-11-08T11:52:17Z” level=fatal msg=“Failed while waiting for 3 host(s) to be active: Get [http://172.17.0.1:18080/v1/projects/1a7/hosts?state=active:](http://172.17.0.1:18080/v1/projects/1a7/hosts?state=active:) read tcp 172.17.0.1:52114-\>172.17.0.1:18080: read: connection reset by peer” component=service Can someone tell me please what rancher is doing on boot up I assume there is cross container networking at the host level but what containers are trying to communicate with each other - is there a list ? Also is there any cross host communication going on - if so what ? The documentation is generally pretty good, however when you get down to this level I can’t find anything to help me out so any guidance would be much appreciated. This has been burning me for the best part of a week now, and I really don’t want to use another Orchestrator 🙂 Thanks Paul ---
**Author:** ![phooper0001](https://avatars.discourse-cdn.com/v4/letter/p/b77776/32.png) [@phooper0001](https://forums.suse.com/u/phooper0001)\ **Post date:** [November 8, 2016, 2:54pm UTC](https://forums.suse.com/t/rancher-high-availability/4554/2 "2016-11-08T14:54:29Z")
It ended up having nothing to do with security groups. Running the below before starting the services fixed it rm -rf /var/lib/rancher/state/.physical\_host\_uuid rm -rf /var/lib/rancher/state/.docker\_uuid rm -rf /var/lib/rancher/state/containers/.healthcheck rm -rf /var/lib/rancher/cache/\* ---
**Author:** ![Kraucer\_Fernandes\_Ma](https://sea2.discourse-cdn.com/flex022/user_avatar/forums.suse.com/kraucer_fernandes_ma/32/1732_2.png) [@Kraucer\_Fernandes\_Ma](https://forums.suse.com/u/Kraucer_Fernandes_Ma)\ **Post date:** [November 14, 2016, 2:37pm UTC](https://forums.suse.com/t/rancher-high-availability/4554/3 "2016-11-14T14:37:03Z")
I want to help you with this. Do you have any detailed documentation/tutorial to share? My problem is: [https://github.com/rancher/rancher/issues/6329](https://github.com/rancher/rancher/issues/6329) Error log: time=“2016-10-19T08:21:50Z” level=info msg=“Can not launch agent right now: Server not available at [http://192.168.169.2:18080/ping:](http://192.168.169.2:18080/ping:)” component=service I made the attempt with a single host and the problem is the same. ---
**Author:** ![aemneina](https://sea2.discourse-cdn.com/flex022/user_avatar/forums.suse.com/aemneina/32/1323_2.png) [@aemneina](https://forums.suse.com/u/aemneina)\ **Post date:** [November 16, 2016, 1:11am UTC](https://forums.suse.com/t/rancher-high-availability/4554/4 "2016-11-16T01:11:42Z")
@phooper0001 take a look at the ports rancher ha nodes use to communicate here: [http://docs.rancher.com/rancher/v1.2/en/installing-rancher/installing-server/multi-nodes/#requirements](http://docs.rancher.com/rancher/v1.2/en/installing-rancher/installing-server/multi-nodes/#requirements)