Hi David,
I investigated, it turns out that these jobs failed because a cluster operator unexpectedly triggered a restart of docker on the cluster, causing these jobs to fail.
Normally no credits should be charged for this.
Cluster operators will try to find a way to do this kind of maintenance without triggering restarts, or will schedule a proper maintenance window in the future.
Related to FuseTS and MOGPR, a performance issue was also logged yesterday, but not sure if it affects you: