Services/F1/Server/Architecture: Difference between revisions

Services/F1/Server/Architecture (view source)

55 bytes removed , 18 March 2011

Confirmed users

927

edits

@@ Line 60: / Line 60: @@
 * if the queue crashes, the data stored should not be lost. Workers should be able to resume their subscriptions
-* The job should be left in the queue while a worker runs it:
+* The job should be left in the queue until the transaction is over
-** a marker (M) is added so it's not picked by another worker
+** a worker that picks a job marks it as being processed
-** the job is deleted upon completion and successful push of the result
+** if a worker crashes while doing the job, the marker should decay after a TTL so another worker can pick up the job
+** if a consumer cannot reach the result DB, the job stays in the queue and eventually goes away after a ttl.
+** if a consumer cannot reach the session DB the job stays in the queue and eventually goes away after a ttl.
-* if a worker crashes while doing the job, M should decay after a TTL so another worker pick up the job
-* if a consumer cannot reach the result DB, the job stays in the queue and eventually goes away after a ttl.
-* if a consumer cannot reach the session DB the job stays in the queue and eventually goes away after a ttl.
 * if a consumer cannot reach the services status DB, nevermind.