Updating core certs in ISD 17 via Coordinator GUI took some time to finish. Just annoying, not an error, but probably we can improve the algorithm we use for batches.
Some machines didn't pick up the new core certificates for at least some hours. This could be an error.
There is no "Update about to expire certs" command or GUI option. This would be a huge improvement. Or a configuration option to automatically do it before e.g. 3 days of expiry time.
In the coordinator configuration, link between 1001 and 1108 was misconfigured. This is either a bug, a misconfiguration or a DB corruption?
There is no easy command to diagnose our scionlab infrastructure. This would be a huge improvement.
E.g. diagnose would check this host has scion configuration (sciond, dispatcher), has access to the control service, then to all border routers in this AS, then to a predifined list of hosts inside our infrastructure, etc.
Whenever a problem was found, we can even print out the corresponding log file.
Then we run it again the the next machine that has an issue.
diagnosewould check this host has scion configuration (sciond, dispatcher), has access to the control service, then to all border routers in this AS, then to a predifined list of hosts inside our infrastructure, etc.