Rclone not finishing transfers to idrive

rclone, latest version on Rocky linux 8.6.

We have created an account with idrive (S3 compatible cloud storage - IDrive® e2), partly because you can get cheap S3-compatible storage there with a very high transfer rate (uploads almost always run at 800-900 MBit with our GBit WAN connection).

The problem is: According to the progress bar, large uploads over 50 GB run without problems up to 100% and then remain there without completing the process. We've already waited more than 1 hour, nothing happens anymore and the files don't appear in idrive either.

30 minutes later:

hi,

a debug log should provide a clue, as with the rest of the help and support template.

Agree, especially a debug log with --dump headers to trace the communication with IDrive.

Here is the complete log:

Thanks!

My quick and immediate impression is that IDrive doesn't respond to this request (within the last 3 minutes of the log):

2022/09/24 19:30:06 DEBUG : >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
2022/09/24 19:30:06 DEBUG : HTTP REQUEST (req 0xc000603000)
2022/09/24 19:30:06 DEBUG : POST /kvm/RAM/rfc0ninefrchi7nfv15otthm78?uploadId=8c8097eb-2e6d-4dda-863e-701665eeafc1 HTTP/1.1
Host: w8p7.fra.idrivee2-33.com
User-Agent: rclone/v1.59.2
Content-Length: 954636
Authorization: XXXX
X-Amz-Content-Sha256: f3fa19dd350a71b7a51067956240b8b6c750a1ca61cd3cb31389c28653078fa0
X-Amz-Date: 20220924T193006Z
Accept-Encoding: gzip

2022/09/24 19:30:06 DEBUG : >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>

I don't know the S3 API well enough to say exactly what is happening, but I guess rclone tries to "commit" the transfer to iDrive and then never receives the "OK, committed successfully" from IDrive.

Please post the remaining items in the help and support template:

  • output from rclone version
  • output from rclone config show idrive_enc: and rclone config show idrive:

then I will ping somebody with more knowledge.

PS:
It would also be helpful to know if you see the same with this slightly different command (removed a leading / and --fast-list):

rclone move /KVM-backups/RAM.qcow2.zst idrive_enc:kvm/RAM --log-level DEBUG --dump headers --log-file=rclone.log

[root@rl8 ~]# rclone version
rclone v1.59.2

  • os/version: rocky 8.6 (64 bit)
  • os/kernel: 4.18.0-372.16.1.el8_6.0.1.x86_64 (x86_64)
  • os/type: linux
  • os/arch: amd64
  • go/version: go1.18.6
  • go/linking: static
  • go/tags: none

rclone config show idrive::
[idrive]
type = s3
provider = Other
env_auth = false
access_key_id = xxx
secret_access_key = xxx
endpoint = w8p7.fra.idrivee2-33.com
acl = private

rclone config show idrive_enc:
[idrive_enc]
type = crypt
remote = idrive:
filename_encryption = standard
directory_name_encryption = false
password = *** ENCRYPTED ***
password2 = *** ENCRYPTED ***

Thanks, did you see the same with the modified command?

@ncw Can you check my analysis above and share any ideas to proceed?

Thank you for continuing to work on this.

And, no, the slightly changed command didn't help either. The messages in the log were also the same.

@peter999 Raghava from IDrive; recently we had similar complaints from other user and after observing the patterns, we found an edge case and fixed it. So it should be fine now; please check and let us know if you have any issues.

Thanks,

1 Like

This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.