I have coned my production VM (an LXD container actually, but for my scanrio it doesn’t matter really I suppose) to a local machine in order to use this clone as test instance going forward.
After ignoring the warning (“no valid certificate”) which is not unexpected I am facing the “Sorry, we will be back soon” error and I am looking for the way around this.
I assume it may have to do with the site-name (erp.mydomain.com) and or the fact letsencrypt is still enabled in some way.
Can anybody kindly suggest any possible routes to my goal (having an http *without the ‘s’) local site with the same content (code and data) as the one I have cloned it from).
You could (potentially) use a letsencrypt wildcard certificate. This requires that you manually create the cert using the DNS method - which in itself implies that you need to have control over the actual DNS zone file.
I have tested/used this (for webpages but not yet for ERPNext - although it should be the same since it IS a webpage) successfully using PowerDNS(specifically with their pdnsutil), and then place the resulting .pem files into the appropriate location on your server’s filesystem using scp or rsync
The instance that is not accessible at the moment is a purely local one that should be used for local testing only.
So, it does not have a domain, and it does not need a domain really and consequently it can not be provided with an SSL certificate neither is there any need for it (as access is never coming from outside of it’s local network).
The solution I need is how to sort “un-SSL” the instance and likewise make it accessible in an unencrypted way via the mere local IP address . I was wondering whether it may be sufficient (or at least a good start) to rename the site from erp.mydomain.com to something lilke site1.local (but would not know how the exact process for this was (just renaming the folder and adjusting the currentsite.txt file?)
cd frappe-bench && bench config dns_multitenant off
sudo service nginx reload
2.) rename site
mv sites/erp.mydomain.com sites/site2.local
do the same change in sites/currentsite.txt
bench restart
sudo service nginx reload
all this does not change anything in the behavior. I can make my way to the website against the “Connection not secure” warnings of the browser and am confronted with the frappe “Sorry, we will be back soon” error.
Next idea I’ll follow up upon is to try to remove the lets encrypt files which I suspect to find in /etc/letsencrypt. Maybe there is a recommended way to undo the encryption which I am not aware of this far.