Deprecate the second argument to Maintenance::error() in favor of a new
Maintenance::fatalError() method. This is intended to make it easier to
review flow control in maintenance scripts.
Change-Id: I75699008638f7e99b11210c7bb9e2e131fca7c9e
Follow-up of Ib56cf57e56f03ce6f1c4b52e3e36f228413c1056,
Job::getQueuedTimestamp() returns a UNIX timestamp, so use
the same here for comparison.
Change-Id: I595f962b71337fc9021963c55ee7dd2443149a46
This can handle dropping queued and re-enqueueing abandoned jobs.
Also use the DB cache table to avoid re-enqueuing duplicate jobs
if this has to get used several times for a queue during a week.
Change-Id: Ib56cf57e56f03ce6f1c4b52e3e36f228413c1056
