DietPi-Backup stops services and does not restart them after successfull backup

Troubleshooting
1

Hello guys!

I use the “DietPi-Backup” function for regular automatic backups. I have set it up to perform backups daily at 03:30 AM and to retain a maximum of the last three backups.

From time to time, I notice that while the backup runs successfully, the services that were stopped beforehand are not automatically restarted. This doesn’t happen every time, but at irregular intervals. I’ve noticed it with the “Jellyfin” service, which then becomes unreachable. I can see that the service hasn’t restarted in the “DietPi-Dashboard” under the “Services” section. The start time there is then very close to 03:30 AM. This issue doesn’t just affect “Jellyfin,” but also services like “minidlna” or “docker,” for example.

The destination for backups is a connected external HDD via USB 3.0 with an extra origin power supply (WD My Book with 18 TB).

Is this a normal behaviour? Or how can I avoid that services get unreachable after automatic backups?

Greetings from GER :waving_hand:

Creating a bug report/issue

I have searched the existing open and closed issues

Required Information

  • DietPi version | CORE=9 SUB=13 RC=2
  • Distro version | bookworm
  • Kernel version | 6.1.0-37-amd64 6.1.140-1 (2025-05-22)
  • Architecture | amd64
  • SBC model | Native PC (x86_64) Lenovo M720Q
  • Power supply used | Lenovo Power Supply 90 Watt
  • SD card used | No, 128 GB SSD (brand: Crucial)

Additional Information (if applicable)

  • Software title | dietpi-backup
  • Was the software title installed freshly or updated/migrated? Preinstalled
  • Can this issue be replicated on a fresh installation of DietPi? No, only sometimes
2

Usually services should be restarted automatically afterwards. You would need to check service logs. Maybe there is a hint for the service not starting correctly.

3

But the rest of the system is stell repsonsive? You can execute commands and stuff?

4

@Joulinar Could you please give me advice which log should I check?

@Jappe Yes, I can connect via SSH for example or I can log in to the dietpi dashboard.

5

You can use following to check service logs

journalctl -u <service-name>

Replace <service-name> with the service in question

1 Like
6

Thanks! Next time Services fail to restart after Automatic backup I wil use journalctl -u command to check for example jellyfin or docker.

7

Does dietpi support the program called “needrestart” that will restart services if an update happens or something shuts them down?

8

Something like this is not needed. There is a reason for the service not being restarted after backup finished. This we have to find out. Installing software to work around is not a solution and is hiding the real cause.

1 Like
9

Today it happened: Automatic Backup was successfull but some services did not restart. For example minidlna, docker oder Jellyfin.

For each service I run journalctl -u, result attached here.
minidlna.txt (3,3 KB)
jellyfin.txt (5,9 KB)
docker.txt (5,4 KB)

10

Hmm there is no sign for a restart of the services. Might be the backup did not finished successfully? Are there other services that have been restarted correctly?

11

Yes, there are services that have been restarted correctly. For example dietpi-dashboard (#200 of software list) and pihole-dashboard. Other services are not reachable.

12

dietpi-dashboard and PiHole are not stopped during the backup. You can check the logs and should see that they are continuously running.

13

Screenshot_20250612_183538_Termius
Screenshot_20250612_183538_Termius2340×1080 187 KB

Dietpi-Backup says last backup is completed. That’s strange.

14

the backup process is more than just creating “the backup” themselves. In addition to the normal backup, there is also the stopping and starting of the services. Unfortunately, this is not included in the actual backup log.

1 Like
15

Why not run the backup interactively and see what fails? You can use script to record the console output.

1 Like
16

yes next step would be to record the complete backup to see what happens. For this, the daily backup via dietpi-backup had to be deactivated, because we do the executions via crontab temporarily.

The crontab entry then looks like this

30 3 * * * /boot/dietpi/dietpi-backup 1 > /tmp/dietpi-backup_$(date +\%Y-\%m-\%d).log 2>&1

This creates a log file in /tmp/ and tags it with a daily timestamp.

1 Like
17

Today I created a cronjob with crontab -e for 03:30 am. Daily backup via dietpi-backup is deactivated now. Afterwards I will interpret the logs and if I am not able to understand them, I will upload them here.

18

the interesting part should be at the end of the file