Look at the build logs for a specific step that takes most of the 1-hour duration. Usually a long hang is followed by an agent timeout, and the job gets killed automatically. You can post the step.
A possible cause could be concurrency limts.
Did you change your image? If you did, revert and test
It is also possible that your run was impacted by a transient network issue, credential failures, services outages, etc.. From https://status.dev.azure.com/ there is no outage so you can try again.
You can mark it 'Accept Answer' and 'Upvote' if this helped you
(Please note: If you have Priority Community support please wait for a dedicated Microsoft support representative to assist you, as they have access to the necessary backend resources. If you have not yet opened a support case, we recommend reaching out through the support channel available under your subscription level.)
Regards,
Abiola