Remote Backup did not work correct

Support and Discussion General Discussion
1

Renote backup did not work if one of the cyberpanels is behind a firewall.

I have a live server fo a customer (live1) and da development server (dev2).

live1 is hosted in the internet and dev2 inside my companys network - protected by perimeter firewall, nir reachable from outside.

I did

  1. API Access enabled on live1 and dev2
  2. set SSH Port back to 22
  3. Start “Remote Backup” on dev2
  4. Enter IP of live1 and Password of User “admin” from live1
  5. dev2 connects to live1 and pulls list of all websites
  6. I select one website and click Transfer Button
  7. on the left side i got:

############################
Starting remote Backup
Start date: 02.02.2022_17-23-48
############################

[02.02.2022_17-23-58] Currently generating local backups for: domain.tld
[02.02.2022_17-23-58] Starting local backup for: domain.tld
[02.02.2022_17-24-00] Waiting for backup to complete…
[02.02.2022_17-26-20] Backup Completed for: domain.tld
[02.02.2022_17-26-20] Local Backup Completed for: domain.tld
[02.02.2022_17-26-20] Sending /home/backup/transfer-6819/backup-domain.tld-02.02.2022_17-23-58.tar.gz to [ip of FIREWALL]
[02.02.2022_17-28-32] Sent /home/backup/transfer-6819/domain.tld-02.02.2022_17-23-58.tar.gz to [ip of firewall]
[02.02.2022_17-28-32] #############################################
[02.02.2022_17-28-32] Backups are successfully generated and received on: [ip of firewall]

  1. Right side shows:

############################
Starting Backup Restore
Start date: 02.02.2022_17-28-34
############################

[02.02.2022_17-28-34] Backup Restore complete
completed[success]

…but there is no website on dev2

So it´s clear what happened; live1 trys to SSH to dev2 and can´rt reach.

This is a BUG by design - live1 server never got the credentials or allowness to connect via ssh directly to dev2. The initial ssh tunnel from dev2 to live1 is not being used to transfer the .tar.gz file.

In my constellation my firewall blocks direct access to my dev2 computer.

Better is the solution for remote backup, tha the client (dev2) waits for a flag-file set in the /home/backup directory of live1, that backup file is ready to transfer and than pulls this file. Pushing from live1 to dev 2 is not a good idea. My 5 cent …

But ok, if you say I have to live with it:

PLEASE remove the BUG in the logs saying !“Restore Complete” and “Backups are successfully generated and received on:”

Non of these 2 steps are successfull.

Best regards,
Kai