Trying to switch over a teamdrive mounted by rclone with a service account. While the output shows it switches over, the config file gets updated, the mount is still using the original SA which it started with.
At the moment SA1.json has errors on files, running the command seems to switch it over to SA2.json (it does in the output and in the config) but mount will only start using SA2.json on a remount.
What is your rclone version (output from rclone version)
v1.52.3 (tested with v1.50 to latest beta, all the same for me)
Which OS you are using and how many bits (eg Windows 7, 64 bit)
Debian 10
Which cloud storage system are you using? (eg Google Drive)
Google team drive
The command you were trying to run (eg rclone copy /tmp remote:tmp)
Thank, I did (multiple times) and re-read it multiple times today. As stated above I use the rc command as stated there. If there is something obvious that I'm missing in that post let me know
As stated there the rclone rc backend/command command=set fs=clouddrive: should be used to change the service account of a running mount. It SHOULD re-authenticate after running that command but for me it doesn't. I'm doing exactly as stated there and it looks like it works except for the auth part.
The problem here seems to be that you are mounting the crypt remote and running the command on the drive remote. I don't think the backend command infra. traverses wrapped remotes. @ncw Thoughts?
Just to rule it out, can you try the following too:
Yep, that's what I expected. Will need to wait for ncw to chime in if there is something that can be done in the existing infra for this or something needs to be added to the infra.
If you want to run backend commands on the clouddrive remote then you will have to use clouddrive:Crypt2 as the target. If you use clouddrive: then rclone will create a new remote and do the command on that, not do the command on the existing clouddrive:Crypt2 remote.
So I would have expected @darthShadow s suggestion to work.
If it isn't working then it means that the right backend isn't being found.
I added a debug to show when backends are being created Creating backend with remote %q which should help debug what is going on.
Ideally we'd get the crypt backend to run translated commands on the backend it is wrapping, but I haven't figured out a sensible way of doing that yet.
OK, have a go with this - I think I worked out what the problem is - namely that crypt wasn't putting the drive backend into the cache so the rc command was never working.
I think I've fixed that and the backend lifecyle so hopefully this will work!
Testing gives me this, I assume I need to send it to test:Crypt2 as that one is not making an new backend so it uses the correct mount. For completion both to test: and test:Crypt2 added.
I don't have an SA with issues at the moment but I assume it works