Task #3719 (closed)
Opened 13 years ago
Closed 13 years ago
BUG: Server synchronisation problem
Reported by: | cblackburn | Owned by: | jamoore |
---|---|---|---|
Priority: | blocker | Milestone: | OMERO-Beta4.3 |
Component: | General | Version: | n.a. |
Keywords: | n.a. | Cc: | jamoore, jburel |
Resources: | n.a. | Referenced By: | n.a. |
References: | n.a. | Remaining Time: | 0.0d |
Sprint: | 2011-01-13 (3) |
Description
After 30 minutes of inactivity the server seems to get into a problem with synchronisation. Any requests for sessions after this point raise exceptions.
The attached log is a test made on Mac OS with all power saving options disabled. The critical point is in the fragment below:
2011-01-03 17:18:45,740 INFO [ome.services.sessions.SessionManagerImpl] (l.Server-3) closeSession called and no more references: 960e9822-44c7-4440-bdb2-78b001225228 2011-01-03 17:18:45,740 INFO [ome.services.sessions.state.SessionCache] (l.Server-3) Destroying session 960e9822-44c7-4440-bdb2-78b001225228 due to : Remove session called 2011-01-03 17:18:45,740 INFO [ org.perf4j.TimingLogger] (l.Server-3) start[1294075111991] time[13749] tag[omero.session] 2011-01-03 17:18:45,743 INFO [ ome.services.blitz.fire.SessionManagerI] (l.Server-3) Reaping 1 clients for 960e9822-44c7-4440-bdb2-78b001225228 2011-01-03 17:48:05,927 INFO [ome.services.sessions.state.SessionCache] (l.Server-7) Waiting for synchronization 2011-01-03 17:48:15,953 INFO [ome.services.sessions.state.SessionCache] (l.Server-0) Waiting for synchronization 2011-01-03 17:48:25,957 INFO [ ome.services.util.ServiceHandler] (l.Server-0) Executor.doWork -- ome.services.sessions.SessionManagerImpl.createSession 2011-01-03 17:48:25,957 INFO [ ome.services.util.ServiceHandler] (l.Server-0) Args: [null, ome.tools.spring.InternalServiceFactory@1e3a43] 2011-01-03 17:48:26,058 INFO [ome.services.sessions.state.SessionCache] (l.Server-0) Waiting for synchronization
Attachments (2)
Change History (5)
Changed 13 years ago by cblackburn
Changed 13 years ago by cblackburn
comment:1 Changed 13 years ago by cblackburn
comment:2 Changed 13 years ago by jmoore
- Owner set to jmoore
- Status changed from new to assigned
comment:3 Changed 13 years ago by jmoore
- Remaining Time set to 0
- Resolution set to fixed
- Status changed from assigned to closed
(In [8762/omero]) Scheduler broken since r8715 due to missed ContextRefreshedEvent (Fix #3717, #3719)
Since the internal handler had a null context, the check for
event.getContext() != ctx always returned true.
master.out following kill -QUIT <Blitz pid> attached.