System.IO.IOException: Device or resource busy #262
-
|
My P2G docker (running in unraid if that matters) gets an exception when it attempts to cleanup old workout files after uploading. This seems similar to issue #221 but I would assume I have the latest code since there are no updates to the docker available. My config file is generic and everything works, it just attempts to reload my last 5 workouts during every interval. I also went into the docker and tried to delete the files, but I think it's redownloading each time so essentially creating a loop. Here's the log:
|
Beta Was this translation helpful? Give feedback.
Replies: 3 comments 1 reply
-
|
Hmmm this is interesting, You could try switching to the stable tag and see if the issue goes away. That would help me narrow down if this is something new introduced on the latest version vs. something environmental: |
Beta Was this translation helpful? Give feedback.
-
|
Same deal with the stable branch: Welcome! P2G is starting up... |
Beta Was this translation helpful? Give feedback.
-
|
I was finally able to reproduce, but only by mounting the If Can you confirm your docker run command (or compose file) is not mounting The following docker compose reproduces the error: version: "3.5"
services:
p2g-console:
container_name: p2g-console
image: philosowaffle/peloton-to-garmin:stable
environment:
- TZ=America/Chicago
volumes:
- ./configuration.local.json:/app/configuration.local.json
- ./data:/app/output
- ./data/working:/app/working # this should not be a mounted volume, its a tmp dir that needs to get deletedlog output: |
Beta Was this translation helpful? Give feedback.
-
|
Huzzah! That did the trick. I am running this in unRAID so it's always a bit flaky to add "unofficial" dockers and I saw the need for a working directory so just assumed I needed to mount it as a volume. Everything worked fine in the previous configuration it would just constantly try and re-upload the last 5 workouts. Thanks for figuring this out and really love this tool. |
Beta Was this translation helpful? Give feedback.
I was finally able to reproduce, but only by mounting the
app/workingdirectory in my container which is a non-recommended configuration.If
app/workingis provided as a mounted volume to the container, then the Docker process is maintaining a lock on the folder and will not allow it to be deleted.Can you confirm your docker run command (or compose file) is not mounting
app/workingas a volume?The following docker compose reproduces the error: