The most likely explanation here is simply a timing issue At a guess, your pull task is running at the moment in time when we are in the process of updating the download site you're pulling from. If you time it just right - or just wrong, depending on your viewpoint - it's possible for the files in the download site to be out of sync, giving rise to this error.
When you run the task again, the update of the download site has completed and all the files are correct, so the pull succeeds. This is completely normal and (I would imagine) happens many, many times a day.
To alleviate this, you might want to schedule the pull task to run more frequently - say hourly - so that if the first pull does fail, you won't be missing the latest files for too long.