Mounted union drive with service accounts causing errors on first upstream

What is the problem you are having with rclone?

I do not know if this is possible, but i want to "merge" multiple service accounts into one folder (union). So I can utilize multiple accounts of 750GB/per day.

If I understand it correctly, after a while lets say a day my Plex with ISOs are unable to load them up and many errors occurs in Google Console Developers panel. It also seems like it is always the first mount in upstreams that fails (causes errors), because when I remove for example 00-enc from upstreams in union it starts to work again and a day after it causes errors again and removing the first upstream fixes it. Do not know if I have a faulty configuration or this just wont work in theory?

What is your rclone version (output from rclone version)

rclone v1.54.0-beta.4793.2b7994e73

  • os/arch: linux/amd64
  • go version: go1.15.2

Which OS you are using and how many bits (eg Windows 7, 64 bit)

Unraid 6.8.3. 4.19.107-Unraid

Which cloud storage system are you using? (eg Google Drive)

Google drive, Team Drive

The command you were trying to run (eg rclone copy /tmp remote:tmp)

rclone mount \
        --allow-other \
        --dir-cache-time=5m \
        --vfs-cache-max-age 2h \
        --vfs-cache-max-size 5G \
        --vfs-cache-mode minimal \
        --vfs-read-chunk-size 50M \
        --vfs-read-chunk-size-limit 500M \
        --cache-dir=/tmp/rclone/union/vfs \
        --cache-tmp-upload-path=/tmp/rclone/union/upload \
        --cache-chunk-path=/tmp/rclone/union/chunks \
        --cache-workers=8 \
        --cache-writes \
        --cache-db-path=/tmp/rclone/union/db \
        --no-modtime \
        --drive-use-trash \
        --stats-one-line \
        --checkers=16 \
        --log-level INFO \
        union: \
        /mnt/union

The rclone config contents with secrets removed.

[00]
type = drive
scope = drive
service_account_file = /auth/00.json
team_drive = ----
 
[01]
type = drive
scope = drive
service_account_file = /auth/01.json
team_drive = ----
 
[00-enc]
type = crypt
remote = 00:
browseable = yes
filename_encryption = standard
directory_name_encryption = true
password = ------
password2 = -----
 
[01-enc]
type = crypt
remote = 00:
browseable = yes
filename_encryption = standard
directory_name_encryption = true
password = -----
password2 = ----
 
[union]
type = union
upstreams = 00-enc: 01-enc:
create_policy = rand
action_policy = rand

We don't advise to do anything to break the 750GB limits.