Nicolas, thanks for the feedback and the very interesting link to your multi-thread cron version! We've added a warning when the max number of database connections is set lower than the cron threads limit. If that happens the cron master will simply fail to start more cron thread jobs (cursor request will raise). Jobs already executing will complete normally, and trigger a re-schedule of the database crons, which will eventually cause processing to resume correctly, once the connections are freed.
BTW, one aspect that is key for us is the multi-process capabilities: cron processing must work properly even if several OpenERP server processes are using the same database - automatically sharing the work between them - but never duplicating it! Hence the peculiar database rows locking strategy - implementing per-task mutex -- https://code.launchpad.net/~openerp-dev/openobject-server/trunk-threaded-cron-vmt/+merge/68064 Your team OpenERP R&D Team is subscribed to branch lp:~openerp-dev/openobject-server/trunk-threaded-cron-vmt. _______________________________________________ Mailing list: https://launchpad.net/~openerp-dev-gtk Post to : [email protected] Unsubscribe : https://launchpad.net/~openerp-dev-gtk More help : https://help.launchpad.net/ListHelp
