[SERVER-40340] PeriodicRunner jobs should persist between MongoE restarts Created: 26/Mar/19 Updated: 06/Dec/22 Resolved: 03/Apr/19 |
|
| Status: | Closed |
| Project: | Core Server |
| Component/s: | Mobile |
| Affects Version/s: | None |
| Fix Version/s: | None |
| Type: | Bug | Priority: | Major - P3 |
| Reporter: | Benjamin Caimano (Inactive) | Assignee: | Backlog - Service Architecture |
| Resolution: | Won't Fix | Votes: | 0 |
| Labels: | None | ||
| Remaining Estimate: | Not Specified | ||
| Time Spent: | Not Specified | ||
| Original Estimate: | Not Specified | ||
| Assigned Teams: |
Service Arch
|
| Operating System: | ALL |
| Participants: |
| Description |
|
It turns out that MongoE can restart with a new ServiceContext, which henrik.edin and I believe will dump the Jobs that the PeriodicRuner was populated with. It should be trivial to move ctor the PeriodicRunner to a new service context, but we do not do it now. (Admittedly, I haven't researched behind this ticket very hard. It may be a won't fix.) |
| Comments |
| Comment by Andrew Morrow (Inactive) [ 02/Apr/19 ] |
|
I think I agree. The restart behavior should really be that: everything goes down, and everything comes up anew. |
| Comment by Mira Carey [ 01/Apr/19 ] |
|
ben.caimano, I think this is definitely a won't fix. If mongoe is restarting, and dumping the old service context, I can't imagine how we'd come back up without re-running the same code we do on initial startup. I can sort of imagine leaving some subsystems up (logging or the secure allocator), but at least for things attached to the service context those will have to get added in again. What's the setup here where we stop mongoe and destroy the service context but somehow come back up again? Can you flesh that out a little for me? |