Unfortunately we don't have that functionality. the _ID was desinged to give permanent urls. However it's a simple fix on http server, simply generate the names of repos and write 100 redirects so old urls will point to the new ones. It's much faster than changing CI/end user configs.
on 15 Jun, 2018 03:04 PM
Thanks for the super fast reply.
I am trying to figure out what http server Rhodecode is using. (I just took over the Rhodecode administration role).
I have found an apace instance running, but stopping it doesn't seems to bring down Rhodecode. So I am a bit confused...
This is an ubuntu server...
on 15 Jun, 2018 03:27 PM
Before the move it was:
"maybe rhodecode is accessed directly without an HTTP server."
Could you elaborate on this, please? This very much might be the case...
No nginx installed.
Apache2 is installed, but if I stop it, I can still access Rhodecode.
Does Rhodecode has it's own web server kind of thing? How can we access it without a webserver?
And if that is the case, probably the suggested solution to create redirects for the old urls is not an option, does it?
on 15 Jun, 2018 05:05 PM
Thanks again for sticking with me!
Indeed there is another Apache server running as proxy, responsible to translate the FQDN to the IP of the rhodecode server.
So now the idea is that the redirection should be modified, not to talk to rhodecode directly, but to the apache server (on port 8080). And configure the apache server on the rhodecode server to redirect these old urls to the new variant?