[Update] Installation via manifest link can fail (Patreon)

Hello everyone, 

[Update 1] Cause is known, resolution is not

Thanks to iotech, I was pointed into the right direction. The issue is caused by Let's Encrypt`s SSL root certificate expiring and being replaced with another, valid one. The error message indicates that the Manifest format is wrong, but in the end, the SSL connection from the native Foundry VTT (Electron) client to vtta.io cannot be established. This is happening for every developer that is self-hosting it's modules on it's own webserver that has SSL encryption enabled (which it should), and not a vtta.io especiality. 

Things I know:

• Only native Foundry VTT (Electron) clients are affected by this, if you are self-hosting using the Linux/node.js package you are able to update the modules without issues. I am not sure for other environments. That's why I wasn't able to replicate the issue in the first place and other users chimed in, stating that they are neither experiencing anything on their end
• It is (somewhat transposed) stated, that the native Foundry VTT (Electron) client will probably not receive an update to reflect the changed SSL chain until v9 is released. I am currently not aware of a time schedule for this release.
• The certificate renewed on foundryvtt.com is supposed to have the correct, detected chain, and vtta.io includes the very same chain, but it still fails. 

Things I do not know:

As some of you might know, the whole infrastructure is running on Kubernetes, something that I am learning while working with it. Currently, I am at a loss on what to do (since the certificate chains from vtta.io and foundryvtt.com are exactly the same) and start learning/ reading up on this. Changes could lead to outages on the dev environment (vtta.dev). I will not be changing anything on the production environment (vtta.io), but that also means that the issues described will probably remain for now. 

Things I consider:

• To leave it as is. It seems to be a misconfiguration on the native client, the regular web accesses using browsers etc. are working fine, so in total I am almost sure that it is not a misconfiguration on my end, but a short-coming on the client. Yes, you can workaround this issue, but as stated above, that might involve complicated steps (for me, in all honesty, there are definitely more knowledge-able people around), but that might risk breaking the dev environment to a state where I perhaps cannot recover and need to redeploy the whole cluster. I am not very happy about this prospect.

I will keep you posted, but it will definitely take a moment.

-----------------------------------------------------------------------------------------

today, an issue arose that I cannot replicate: Installation of the module directly from the module browser is failing for some people, others are reporting a success. The error message is: "Installation of [module name} failed: [url] did not provide a valid manifest file".

I tested it thrice with clean installs on various Foundry VTT versions, but did not run into the issue once. While I am investigating this weird issue, you can always install the module via ZIP manually. You can retrieve the download archives from the individual modules on their respective pages, the list is found at https://www.vtta.io/assets

Thanks for your patience, 

Sebastian