If you add the external domain to the zonegroup's hostnames and endpoints, then it will be able to respond to that domain. This is assuming that the error message is that the URL is not a valid bucket. We ran into this issue when we upgraded from 10.2.5 to 10.2.9. Any domain used to access RGW that want in the hostnames and endpoints stopped working. Adding each domain to those lists fixed it.
On Sun, Oct 22, 2017, 9:31 AM Rudenko Aleksandr <ARudenko@xxxxxxx> wrote:
_______________________________________________Hi guys,
We use Ceph as S3-compatible object store and we have our self-development web-interface for our customers on different domain.
Now we use Hammer(FCGI + Apache as RGW frontend) but we have plans for upgrade Ceph from hammer to luminous.
In luminous release FCGI frontend was dropped and Civetweb frontend always checks CORS.
As I said above, our web-interface works on different domain and by default Civetweb returns 403 оn HTTP OPTIONS request from our web-interface.
If I PUT CORS for some bucket, everything works fine. But isn’t good idea because bucket owner can PUTs your own CORS and overwrite our default CORS and lost access to bucker from our web-interface. It’s not cool)
In past we rewrite http response header by Apache rules for our web-interface and pass CORS check. But now it’s impossible to solve on balancer level.
What is right way?
---
Best regards,
Aleksandr Rudenko
ceph-users mailing list
ceph-users@xxxxxxxxxxxxxx
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
_______________________________________________ ceph-users mailing list ceph-users@xxxxxxxxxxxxxx http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
