Open wdbaruni opened 2 days ago
root@bacalhau-vm-prod-0:/data# pwd
/data
root@bacalhau-vm-prod-0:/data# ll
total 72
drwxr-xr-x 8 root root 4096 Sep 30 16:27 ./
drwxr-xr-x 22 root root 4096 Sep 30 16:31 ../
drwxr-xr-x 2 root root 4096 Nov 9 2023 .bacalhau/
-rwxr-xr-x 1 root root 65 Sep 30 16:31 bacalhau.run*
drwxr-xr-x 3 root root 4096 Sep 30 16:10 compute/
-rw-r--r-- 1 root root 332 Sep 30 16:10 config.yaml
drwxr-xr-x 5 root root 4096 Sep 30 16:38 ipfs/
-rw------- 1 root root 1597 Nov 9 2023 libp2p_private_key
drwx------ 2 root root 16384 Nov 9 2023 lost+found/
drwxr-xr-x 3 root root 4096 Sep 30 16:10 orchestrator/
drwx------ 2 root root 4096 Nov 9 2023 plugins/
-rw------- 1 root root 13 Sep 30 16:10 repo.version
-rw-r--r-- 1 root root 825 Sep 30 16:31 secrets.sh
-rw------- 1 root root 197 Sep 30 16:31 system_metadata.yaml
-rw------- 1 root root 1679 Sep 30 16:27 user_id.pem
root@bacalhau-vm-prod-0:/data# which bacalhau
/usr/local/bin/bacalhau
root@bacalhau-vm-prod-0:~# bacalhau --repo=/data version
Flag --repo has been deprecated, Use --data-dir=<path> to set this configuration
CLIENT SERVER LATEST UPDATE MESSAGE
v1.5.0-alpha10 v1.5.0-alpha10 1.4.0
Checking the instance now, it appears the key has content in it, was this a one time issue or were manual steps taken to mitigate it?
I had to delete user_id.pem
. Not sure why that happened. The logs don't say much and I am not sure what state the repo was at before the migration. It does say the file was created today and was empty
My hunch right now is: The key in the repo was empty before the migration ran and thus a new (valid) one was never initialized, but that's a pretty weak hunch.
but how did the node run with an empty user_id.pem
before using v1.4? Did we change anything related to the user key in v1.5?
Deployment failed in prod network. For some reason
user_id.pem
is empty and causing nil panic