If this is a duplicate please tell me, but/because others seem to have had my/a similar error
But their solution seems not to be mine.
My system is a NC 15 fresh install in subfolder /nextcloud on raspbian stretch with apache2 and app social activated.
I get the following error:
[public] Error: Exception: The requested uri(/.well-known/webfinger) cannot be processed by the script ‘/nextcloud/public.php’) at
0. /var/www/html/nextcloud/lib/private/AppFramework/Http/Request.php line 810
getRawPathInfo()
/var/www/html/nextcloud/public.php line 45
getPathInfo()
/.well-known/webfinger
from XX.XXX.XXX.XXX at 2019-01-02T18:59:24+00:00
Dein Web-Server ist nicht richtig eingerichtet um “/.well-known/webfinger” aufzulösen. Weitere Informationen findest Du in der Dokumentation.
I worked through this https://docs.nextcloud.com/server/14/go.php?to=admin-setup-well-known-URL
(as mentioned in the admin-panel) and for caldav/carddav the redirecting sends me to the correct dav-url. at least my browser tells my so.
but not for webfinger. No redirecting, just: missing resource.
Could you please post your web server config (the file which covers .well-known/caldav and .well-known/webfinger) so we can check that?
Maybe we see something.
I actually don’t need the app (but i’d like to).
Right now, i can post, but i cant follow other fediverse-instances. If i try, my log shows a error message as mentioned in my first post. (and of course the instance i wanted to follow shows a error message).
I searched a bit now and stumbled across this github issue:
You could try the commands daita posted:
[…]what do you have when opening https://your.domain.tld/.well-known/webfinger?resource=acct:YOUR-USERNAME@your.domain.tld
And check the reachability from your server it self via command line command: curl -H "Accept: application/ld+json" -X GET https://your.domain.tld/nextcloud/apps/social/@YOUR-USERNAME/
Unfortunately, the update to v0.1.2 did not solve the issue.
I noticed another strange behaviour: sometimes my admin-panel tells me, /.well-known/caldav+carddav were not properly setup up. one minute later, they are.
I do have a similar issuesince I moved my NC instances behind a HaProxy instance. The SSL handling is done by HaProxy and the request is internall forwarded over http. When the page then requests /caldav then NC responds with a 301 redirect to a http URL instead of a https URL. I wonder how I could force NC to always use https URLs when it builds redirect responses like this.
Hmm… after thinking a bit longer about it, I think this is a webserver issue. Because the redirects to the /.well-known URLs are handled by .htaccess files (if you are using apache) or by the nginx configuration (if you are using nginx).
With nginx configuration I could help. Which webserver are you using?