Hi folks, I finally managed to complete the deployment of a High Availiability reverse proxy for server hosted in the community cage[1]. While the switch should be invisible to everybody, and likely bring nothing (except resilience and a better architecture on the system, and maybe more reason for me to complain about glusterd [2]), I plan to switch the web applications one by one from the old reverse proxy to the new one. Main website and download would not be impacted by that. So next week, I will start by changing fstat.gluster.org, see how it go, and start to switch others. So if you see anything weird with fstat.gluster.org, please ping me on irc. Then I will move softserve, analysis and likely jenkins, and later move those to be on the internal lan, so we can save some IP v4. Gerrit still need to have 1 ssh port exposed, not sure how to solve that cleanly. [1] https://github.com/gluster/gluster.org_ansible_configuration/blob/m aster/playbooks/deploy_proxy_ha.yml for the playbook, the roles are in the same repo. [2] https://github.com/gluster/gluster.org_ansible_configuration/blob/m aster/roles/cert_syncer/tasks/main.yml#L41 -- Michael Scherer Sysadmin, Community Infrastructure and Platform, OSAS
Attachment:
signature.asc
Description: This is a digitally signed message part
_______________________________________________ Gluster-devel mailing list Gluster-devel@xxxxxxxxxxx http://lists.gluster.org/mailman/listinfo/gluster-devel
