Description of problem: As of now it is little difficult to setup ROBO .Raising this as enhancement to ease the deployment of ROBO using gdeploy Version-Release number of selected component (if applicable): How reproducible: Steps to Reproduce: 1. 2. 3. Actual results: Expected results: Additional info:
Kasturi, can you please list what are the pain points as of now? And what changes you would like to see?
As of now we totally have four conf files which we use for deploying ROBO setup. 1)Once hosted-engine is deployed on one first node, user has to go and edit hosted-engine machine to permit root login and execute engine-robo.conf to change database values, install nagios-server-addons and configure nagios. 2) Then execute second-host.conf to add second node 3) Then execute third-host.conf on third node. It would be nice if we can have a conf file which just takes three hosts details , FQDN and mac for the hosted engine and finish complete installation with out any manual intervention in between. When any one of the step fails, script should stop executing at that step. May be a cleanup script which cleans up in case any of the step fails in between.I can provide the steps for cleanup which i got from RHEV folks. 4) Once the deployment is done go to UI , then create a new network and assign that network to glusternw role and assign this new network to a nic. I am not sure if this can be done from gdeploy. will we be able to do this if there is a RESTAPI available?
As per discussions with Sahina, steps 2 & 3 are no longer needed. Sahina can you please comment on those?
Steps 2 & 3 can now be done via the UI using additional host deploy. What's the ask here, Kasturi? Do you want all 3 hosts and engine installed via a single gdeploy conf file? Current steps: 1. Setup HC nodes via hc.conf 2. Deploy hosted engine using hosted-engine --deploy 3. Setup passwordless ssh to engine host, and customize it using a gdeploy conf file 4. Add additional hosts via UI
(In reply to Sahina Bose from comment #6) > Steps 2 & 3 can now be done via the UI using additional host deploy. > > What's the ask here, Kasturi? Do you want all 3 hosts and engine installed > via a single gdeploy conf file? sahina, yes > > Current steps: > > 1. Setup HC nodes via hc.conf > 2. Deploy hosted engine using hosted-engine --deploy > 3. Setup passwordless ssh to engine host, and customize it using a gdeploy > conf file > 4. Add additional hosts via UI
Commit: https://github.com/gluster/gdeploy/commit/354260b393d fixes the issue. Step 1 & 2 in comment #7 is config file change. Step 3 has to be carried out manually because passwordless ssh is a bootstrapping requirement for gdeploy.
Verified and works fine with build gdeploy-2.0.1-2.el7rhgs.noarch With the change present in comment 10, ROBO setup can be done in four steps. 1) Run gdeploy on first host which will create volumes, deploys hosted-engine on first host. 2) Run engine-robo.conf on HEVM. 3) Import storage domains. 4) Add additional hosts from UI.
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://rhn.redhat.com/errata/RHEA-2017-0483.html