I expect to have multiple users working with their own copy of vagrant off their own individual laptop using the same vagrant repo for an environment (each user having their own Softlayer user name). The primary provider will be Softlayer (im not sure whether each will have their own API key/user or if they will all use the same user/key pair. I think I may need to support both depending on my customer). Ill probably set a secondary provider of Vbox or kvm for local builds of small sets of servers where only the owner of the laptop will interact with the vms.
I dont think we will be utilizing the vagrant ssh stuff much so doing things like auto port forwarding per user is not something im worried about on the primary provider. On Friday, February 21, 2014 4:52:06 PM UTC-5, Alvaro Miranda Aguilera wrote: > > how do you log into the server/machine with the boxes? > > how is different each user? or all the users use the same account? > > > > > On Fri, Feb 21, 2014 at 7:21 PM, <[email protected] <javascript:>> wrote: > >> Hi all, ive just recently been reading through the vagrant docs and have >> a few quesitons about organization and using one set of vagrant configs >> from different users on different PCs. >> >> I read that loading other vagrant files is possible so you can chain >> them, so my thoughts were to do the following structure >> >> /environment_name/ >> |-> common/ >> |-> servers/ >> |-> <hostnames ...>/ >> |-> app_groups/ >> |-> <groupname ...>/ >> >> The idea being that each level has a vagrant file, the environment >> directory includes all vagrant files servers, servers include only common, >> and app groups include specific servers. >> >> The rationale for such a setup is that I want to use this for dev,prod, >> and disaster recovery by having multiple ops users working off the same git >> repo, but im concerned that they might run 'vagrant down' in a production >> setting if I do just one giant vagrant file. My hope is with this I can >> tell them never run commands from within env dir unless in disaster >> recovery or some automated Jenkins env rebuild type scenario and always use >> individual server or app group directories otherwise. That way if they dont >> explicitly specify the host or whatever they dont break a production >> environment only a smaller subset. >> >> My concerns are primarily with vagrant states and how that would work >> with multiple users using the same code. I understand that the global state >> is per user, but what if I did a vagrant up under the hostname dir for >> server and then went into an app group that also contained it and tried to >> run a vagrant down/destroy/status would this even work or would it have >> problems with the state? What if one user ran vagrant up on a host and >> another user tried to do vagrant down/status/destroy would state cause a >> problem here? Would either of these two scenarios behave differently based >> on whether its run against Vbox/VMWare vs a cloud provider? >> >> Id appreciate any thoughts/recommendations, >> Thanks, >> >> -- >> You received this message because you are subscribed to the Google Groups >> "Vagrant" group. >> To unsubscribe from this group and stop receiving emails from it, send an >> email to [email protected] <javascript:>. >> For more options, visit https://groups.google.com/groups/opt_out. >> > > -- You received this message because you are subscribed to the Google Groups "Vagrant" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/groups/opt_out.
