How to fix /dev/tape: No such file or directory
Closed this issue · 10 comments
Hi @barcus
First of all thanks for your work 🥇
But unfortanly I have an issue with the storage daemon on a virtual maschine. (Ubuntu 20.04 LTS)
A backup job is failing after a few seconds there is an error "error wanted 2000 OK storage, got 2902 Bad storage"
The docker logs of the sd container are saying:
chown: /dev/[n]st*: No such file or directory chown: /dev/tape: No such file or directory
It is happening with the newest alpine-psql and the newest ubuntu-psql images :(
Hi @k3nd0x, thanks for your feedback
This message is just a warning, it's not related to your storage daemon issue, you can ignore it. (chown /dev/*)
I will update docker-entrypoint to avoid this warning when device does not exist.
Could be great to know why you storage daemon exit.
It looks like you use docker-compose to run the Bareos stack
Can you run docker-compose up (without -d) ? in this way bareos-sd will display an error message.
It should be link to bareos storage configuration or your docker-compose file if it exit instantly on start
@barcus Thanks for your answer :)
the SD is starting without any problems. status storage in bconsole says everything is fine. When i start my first backup job the SD container exits and stops working only the chown failes are showing up in the logs. The SD can read, write & execute to the mounted storage device.
additionally I have to say I installed the bareos-client directly on the maschine (without docker-container) The client is reachable and should not be the problem.
Can you please add this in your compose file (inside bareos-sd container) before launching an other backup job ?
command:
- "/usr/sbin/bareos-sd"
- "-u"
- "bareos"
- "-f"
- "-d"
- "200"
It will enable high level debug for storage daemon
Before a backup job:
sorry, i would like to see the error when the container exits, during the backup job i imagine :)
because bareos-sd crashed (exit 139) as we can see in your first screenshot
Hi, sorry for the late reply: This are the full logs after starting the container and after the first job:
Seems that your storage daemon does not speak to bareos-fd but bareos-dir. I don't know why but i think there is an issue in your configuration.
Take a look on my own logs !
Oh shi.. I found it. My client was unable to resolve the name of the bareos-sd ... The name of the SD is "bareos-sd" but my client is running directly on the maschine not in a docker container. So the DNS entry (or /etc/hosts entry) for bareos-sd was missing directly on the maschine. Sorry I bothered you with my config :) but thanks for your help. My backup jobs are running now 💯