Machine Agent must monitor Service Unit Agent health
Bug #617556 reported by
Gustavo Niemeyer
This bug affects 2 people
| Affects | Status | Importance | Assigned to | Milestone | |
|---|---|---|---|---|---|
| pyjuju |
Confirmed
|
Wishlist
|
Unassigned | ||
Bug Description
The Machine Agent (MA) must observe the state of Service Unit Agents which are controlling the formulas initialized in the running machine the MA is responsible for. When a Service Unit Agent dies, the Service Unit should be recovered in a reasonable way (TBD).
This is an extension to the initial MA work defined in bug #617042, and should be achieved through the ServiceUnitState API defined in bug #617541.
| description: | updated |
| summary: |
- Machine Agent must monitor Service Units + Machine Agent must monitor Service Unit Agent health |
| Changed in ensemble: | |
| status: | New → Confirmed |
| importance: | Undecided → Low |
| tags: |
added: agents removed: spike |
| Changed in ensemble: | |
| milestone: | none → budapest |
| importance: | Low → High |
| Changed in ensemble: | |
| milestone: | budapest → dublin |
| Changed in ensemble: | |
| milestone: | dublin → none |
Revision history for this message
| Kapil Thangavelu (hazmat) wrote | #1 |
| Changed in juju: | |
| importance: | High → Wishlist |
To post a comment you must log in.
with the move to upstart, this is less of a concern for the machine agent.