I just installed 8.1.0 and set it up to be behind my nginx proxy. This worked fine with an older install version 7.10.10, but when I do it with 8.1.0 I get “an error occurred while fetching metadata” . I enabled qa mode and looked for the graphql error and it’s a 502 bad gateway coming from the proxy I think.
Any idea how to fix this so it works? On another app that I’m proxying I had to configure that app to override the protocol and send as https even though the request was on http because the proxy is receiving on https and then passing on http.
Hope that makes sense. Any ideas?
Rich
I was able to fix this by adjusting the buffering on Nginx specifically I added these parameters in the server section:
proxy_busy_buffers_size 512k;
proxy_buffers 4 512k;
proxy_buffer_size 256k;
Only weird thing now that is happening is when I log out and try to log back in it just bounces me back to the login screen.
I am having exactly the same issue as described above.
I have installed SuiteCRM 8.1 via the bitnami docker image via docker-compose. Once everything was set up, I was able to access the SuiteCRM login page. When I entered the standard username and password and observed the network tab, I saw that graphql gets a bad gateway from the nginx reservse proxy.
I reverted back to the 7.12.7 docker image and now SuiteCRM works. I really only changed the image in the docker-compose as compared to the setup, where I get the error above.
Is there any other way besides fixing the nginx server configuration? I really don’t want to touch that, since we are running quite a few different services.
Thanks for the reply @klauswoolhouse ! I am doing this on our root server, so in principle we control the proxy (it’s another docker container). But I’d rather not modify it, since all other services are on the same root server.
@rwebb616 @webdruid @MyCRM1
Try this free open source app, nginx proxy manager, it’s a control panel (in a container) designed to make it easy to get nginx proxy working for your web apps, including SuiteCRM 8.