Good morning,
I’m trying to setup access to Cubbit’s S3 space with no success. I know how it works and trying with other software like Veeam, Iperius, etc everything works. Using the same parameters after selecting S3 compatible I can’t login and I get error “Failed to connect: A WebException with status NameResolutionFailure was thrown.” What am I doing wrong? I thank whoever will help me. Andrea
Welcome to the forum @Andrea_Mizzi
could happen if you used documentation like https://www.cubbit.io/s3-compatible-backup
which says to “Just change the endpoint” to s3.cubbit.io, which is an unknown name.
Their site has that error a lot. Which endpoint are you using for Duplicati and the others?
When on the Destination screen, try Test connection. Is that what you tried before?
If not, what was it? For now, it might be best to use no destination advanced options too.
Hello
I admit to being a bit surprised by this information; I tried to do a backup to Cubbit following this issue:
and never succeeded with last Duplicati beta (2.0.7).
I am currently in the process of testing a library upgrade to solve it for next canary, so I am wondering if I am toiling for nothing 
This forum topic was from no server connection. @Andrea_Mizzi any news about later testing?
One person got the cited issue, but I don’t think there’s much data on how widespread issue is.
I’m assuming cited issue requires a server connection in order to get as far as the next issue…
From what I see (error message about wrong Xml format) it seems a compatibility issue.
From what I have seen on the S3 protocol, it is not a simple thing. I suspect that Cubbit being somewhat a new enter on the market use new features that are not supported by the (old) drivers of Duplicati. So far updating Minio to latest seems to get results (that is, a successful backup). At the moment I am (slowly, painfully) testing this new driver with old servers and configuration. No use to be able to support Cubbit if it breaks all the rest.
Good morning,
the initial configuration seems correct and the program connects to the cloud but unfortunately it doesn’t complete it and the error I am attaching appears.
I tried veeam and hyperbackup on the NAS and everything works fine. I don’t know.
The file you linked to is not readable for me but it don’t matter much, your reply confirms that Duplicati in its current state is unable to handle this provider, thanks.
Clipboard images can be posted to the forum. The downloaded after.jpg file shows this:
which doesn’t tell details, but looks like a dblock put failing and retrying default 5 times,
each time with a new name as expected, but with same content. Size and Hash match.
With apologies to new semi-compatible S3 market entrants, I think compatibility burden is theirs.
Here’s me responding to one such entrant, who needed a change in order to work with Duplicati:
OTOH version upgrades might be forced by the desired/required migration from .NET Framework.
With limited resources, it’s IMO always a question of priorities when deciding where to spend time.
Cubbit by itself is not interesting me. I don’t care about investigating these specific problems at all. Cubbit problems showing that Duplicati project has not kept with the times in updating drivers is interesting to me and I am willing to invest time on it (that is, attempting to update the Minio backend).
and some of these updates might show general benefits, so single cases like Cubbit might benefit.
Not having them be the main driver sounds good. I mentioned the .NET > 4 migration, and there’s
also TLS 1.3 problems which could worsen as Windows 11 rolls out (or its fixes might offset them).
It’s more on topic for the recent FTPS truncations, but I’ve found technical explanations on causes.
OpenSSL had a lot of those. If someone’s complaining about SChannel I’m not sure where to read. Mentioning new TLS here because some drivers may have had to do workarounds for the problem.
So yes the world changes and we have to follow. Still looking for ways to fit willing helper-people in, possibly in other areas. Maybe I can work on that some, trying to get some PRs very well prepared.
Staying selective on PR generation would also help reduce the burden of too many sitting in queue.
That would weigh against wanting to get new people started on small PRs, so it’s all a tangled goal.
How to work on untangling is going to be a continued challenge, but maybe we can find some ways.
I’m encouraged by @tkohhh and @Jojo-1000 work and potential (maybe gap-filling) so will pursue.
Interest areas and skill sets factor in too. There isn’t enough resource for a huge amount of training.
Volunteers may have to read code on their own, and probably help get their PRs very well prepared.
Developer category forum talk can help, and also serve as some training (and easier than PR talks).
Hints from you are also encouraged, as Duplicati has proven a big challenge for a single maintainer.
Definitely want to keep you doing what you like (same for all volunteers), so it’s a jigsaw puzzle here.
Thanks for putting aside my concern that Cubbit was a driver even if it might wind up a beneficiary…