Errors and Troubleshooting

Docker Errors

Connectivity issues between Docker Desktop and UI_V2 may lead to issues with experiments completing. When an error message is seen in the console, check which data file was running, then manually Delete all containers in docker desktop and restart both applications. Last, check data files to ensure that no intermediate files (.tar extension) were left behind.

The most common connectivity timeout error may occur when the computer logs out or enters sleep mode partway through a run. This issue becomes more frequent when Docker Desktop is not restarted between runs.

The first time the application uses a tool, the container is pulled from dockerhub (which is updated via github). This first pull event may be slow but afterwards, it will be faster. One issue that may occur here is once a container is pulled, it will not automatically update to the latest version. To update to the latest version, you must navigate to the “Images” tab in Docker Desktop, then “clean up” or remove images. Once the application is run again, it will automatically update to the latest version.

Two docker containers with the same name can not be run at the same time, ensure that all files have unique names and no docker containers are running or stopped before starting an experiment (these must be deleted).

Recovering Data

When a docker container exits on its own, its experiment was completed successfully. When left running indefinitely, it has failed. To retrieve any data from partial runs, see the message console to find the location or “Working Directory” of the run. Data is deleted upon exit of the application and must be retrieved before then.

Docker Setup on Windows

Docker requires WSL2 to be enabled. This should be automatically enabled, is not enter “Settings”, then on the “General” page, select the box titled “Use the WSL 2 based engine”. Then select “Apply and Restart”.

Current Issues Exist with DEIMos in the workflow

DEIMos generates a slightly different output from mzMine, autoCCS requires the mzMine values. DEIMos has received some modifications to allow it to work, however some small differences exist. DEIMos is best suited for single field usage at this time.

DEIMos is a very efficient and accurate tool that also outperforms mzMine in terms of speed. However, the current version is not entirely compatible with usage in a docker container and as such, it runs slower than expected and may run into memory issues. We hope this can be resolved in future versions of this application. We also hope to incorporate additional DEIMos functions in the future.