forked from eac/resque-durable
-
Notifications
You must be signed in to change notification settings - Fork 1
deepthawtz/resque-durable
Folders and files
Name | Name | Last commit message | Last commit date | |
---|---|---|---|---|
Repository files navigation
== What's this? == Resque/Durable allows important jobs to survive when a encountering a failure with Redis or the system in general. It does this by adding traditional database-backed audits for durable jobs enqueued in Resque. When an audited job isn't marked as complete after a certain amount of time (10 minutes by default), it's considered failed, and will be re-enqueued with exponential backoff until it succeeds or is removed. == Usage == See /examples == When things go wrong == Audits stick around, and will be retried, until completed or expired by the monitoring script (expiration is configurable). Due to the backoff delay, a typical job won't be retried more than 10 times in 24 hours, or 20 times in a week. If a queue becomes backed up for too long, jobs may become double enqueued. The first enqueued job will mark the audit as complete, and the second version of the job won't be worked on. This strategy greatly reduces the possibility, but doesn't gaurentee, the same job isn't performed twice.
About
Audited Resque jobs with automatic failure recovery.
Resources
Stars
Watchers
Forks
Packages 0
No packages published
Languages
- Ruby 100.0%