I'm attempting to install my first remote proxy. I've configured a zone on a Solaris 11 host dedicated for this purpose. The host / zone have no issues reaching the IPS repository. OCDoctor.sh --proxy-prereq confirms all checks are OK.
When I try to use the EC to deploy the proxy it fails (with no explanation), so I'm attempting to install it manually if only to see what is going wrong. Here's what's happening.
root@opsproxy:~/xvmoc_full_bundle# pkg publisher
pkg: Unable to set locale 'en_CA.UTF8'; locale package may be broken or
not installed. Reverting to C locale.
PUBLISHER TYPE STATUS URI
solaris (syspub) origin online proxy://http://myhost.company.com:8080/
root@opsproxy:~/xvmoc_full_bundle# *./install -p*
perl: warning: Setting locale failed.
perl: warning: Please check that your locale settings:
LC_ALL = "",
LC_COLLATE = "",
LC_CTYPE = "",
LC_MESSAGES = "",
LC_MONETARY = "",
LC_NUMERIC = "",
LC_TIME = "",
LANG = "en_CA.UTF-8"
are supported and installed on your system.
perl: warning: Falling back to the standard locale ("C").
Verifying we can reach the solaris IPS repository.
This may take a couple of minutes if it is not currently reachable.
solaris IPS publisher is reachable
Error: //email@example.com,5.11-0.175.0.0.0.2.0 not found in solaris IPS repository, exiting
Looking at my OPS repo, I have a few branches
I admit though that I don't yet fully understand what these branches represent or why my install attempt seems to be looking for a much older one.
Thanks for the link. I'll keep that in my back pocket.
The issue turned out to be that my IPS repository. I had built it manually before even installing the EC. Once the EC was installed I needed to basically shut down my own pkg/server and configure a Solaris 11 library within the EC pointing to the mount of my repository. The EC then re-sync's the repository, remounts the NFS share in a new location and reconfigures the pkg/server service on a new TCP port.
Once I had an EC-blessed repository I was able to deploy a proxy server.