I think that one is because you have specified the container port in the colabora settings whilst the proxy redirects to the container so it should be straight https.
Noticed a lot of you guys are on I guess shared hosting and just wondered if you had created a seperate subdomain fot the collabora server so the DNS is right.
After that I am thinking we are looking a cert error or maybe the trusted domain section of config.php, which I have edited to included office.mydomain.com.
You can get inside of the container by docker exec -it ‘containerid’ /bin/bash
I did a domain ping from the container ping office.mydomain.com and now totally confused as who are Catalyst2 Services Ltd?
I did another check as well and in my client browser typed https://office.mydomain.com and ended up on the NextCloud page so the proxy isn’t working correct.
I am self hosting so the router is set up to port forward https http and (dns? dunno added it anyway)
Internally unless I edit the client hosts file to set nc.mydomain.com & office.mydomain the external DNS route internally will end you up on the router(public ip) rather than 192.168.1.9.
Externally it would work fine.
So my domain control panel has the nc.mydomain,com and office.mydomain.com ip of my public IP (router ip)
In windows edited the C:\Windows\System32\drivers\etc\hosts file to include.
Now when I click on a document collabora launches but gives me a message “Well, this is embarrassing, we cannot connect to your document. Please try again.”
So maybe this is just trusted domains? Off to check the logs in that docker container.
No… Think its prob to do with this line in Apache error log but not sure why or what permission is being denied.
[authz_core:error] [pid 9003] [client 192.168.1.7:54881] AH01630: client denied by server configuration: /var/www/html/data/.ocdata
No need to forward DNS (53) to your internal server.
Are you saying before you edited the hosts file the URLs internally landed you on your router admin page?
Does your router rupport DNS override? I’ve switched my dumb router out for a PFsense box and will never go back due to the advanced features for my network, but many normal routers will at least offer DNS overrides.
Yeah dunno why I did DNS but hey, yeah my public IP is the router address and the port forwarding only works one way.
So yeah it would take me to the Talk Talk internal config page.
I haven’t worked for a long time but usually with internal subnets you would run off a split DNS but hacking the hosts files for a single computer is no prob.
Its so long can not even remember if that is normal port forwarding behaviour I think so lol.
I have been doing quite a bit with Next Cloud over the last couple of days and really should of took a hint with your article but it took me till last night to install NC 9
Doesn’t matter really I could prob set up bind or DNS masq and set the router DNS to the owncloud box and still use it for dhcp.
Its virgin at the community Jason are they still static IP’s? They say they are DHCP but from experience they never change, but again I don’t do this stuff regular.
Its me who has plumed for 9, the Collaboro probs are a game breaker, just need to install 9 for what is little more than a 1 man show. In fact 10 as the missing email function isn’t really a bother either.
11 is still hot off the press and yeah I will be eventually employing it.
Eventually I will work out Collabora but I can install 9 tomoz and get to grips with 11 at my leisure.
Its been three days second day I just had one of those days where I ass hatted everything but the last install, I am pretty sure should of worked and it didn’t. I can’t install something that says “Well this is embarising…”
If you have any input as I am stumped and just going to install 9 short term.
Prob not as every time I got ‘access denied’ it was due to wrong config, as I found out later.
I had either the wrong domain names, apache conf files or DNS errors.
It was Aufs that stumped me in the end but you will prob find the errors are going into other_vhosts.log if the SNI & DNS is not correct.
I also split the log config into nextcloud.log and collaboro.log just to make things a little more easier to distinguish.
Also does your client and server have the correct DNS for the IP running example.com? example.com should be your standard nextcloud domain.
The second domain often ‘office’ is just using apache (Server Name Indication) to proxy to 9980
Somewhere along the line your config isn’t right, but without seeing whats going on in the logs its hard to say.
ps the original docker run -t -d -p 127.0.0.1:9980:9980 -e 'domain=cloud\\.nextcloud\\.com' --restart always --cap-add MKNOD collabora/code
with docker run -t -d -p 9980:9980 -e "domain=example.com:51111" --cap-add MKNOD collabora/code
should look like docker run -t -d -p 9980:9980 -e "domain=example\\.com:51111" --cap-add MKNOD collabora/code
But like I say you should be getting log info to back this up.