[Pulp-list] HELP - Converting from crontab to interval scheduler

Jay Dobies jason.dobies at redhat.com
Tue Mar 29 14:08:15 UTC 2011

Hash: SHA1

On 03/29/2011 10:05 AM, Jason L Connor wrote:
> Hi All,
> We are converting from using the crontab for scheduled repo syncs to
> using a new interval scheduler (start_time, interval, iterations).
> I'm looking for ways to properly handle this conversion for existing
> repo schedules. Here some of the ideas I've had:
>      1. Migrate the database and parse the repo schedule and convert it
>         into an interval schedule that, as closely as possible,
>         resembles the current crontab schedule. This has the drawback
>         that we currently don't have the ability to parse the schedules
>         in any meaningful way and I've no idea how long it would take to
>         write/port one.
>      2. Migrate the database and blank out any schedules we find,
>         logging the fact, and writing messages to stderr for the user to
>         check the log. This seems kinda dirty, but it's probably the
>         easiest in terms of getting finished on time this sprint.
>      3. Same as #2, but back up the original crontab schedule into a new
>         repo field so that we can properly migrate it in the future.
>      4. <You idea here>
> I'm not super happy about any of these options, so feedback is both
> welcome and needed at this point.

Scheduled repo sync hasn't worked in a while, so I'm going to say we
take advantage of the fact that we're pre-alpha and just clean up any
crontab entries and mention in the release notes that they have to
re-add repo schedules.

- -- 
Jay Dobies
RHCE# 805008743336126
Freenode: jdob
Version: GnuPG v2.0.14 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org/


More information about the Pulp-list mailing list