[MXS-1436] Investigate Replication Manager and extract the master down logic. Created: 2017-09-18 Updated: 2017-10-24 Resolved: 2017-10-24 |
|
| Status: | Closed |
| Project: | MariaDB MaxScale |
| Component/s: | N/A |
| Affects Version/s: | None |
| Fix Version/s: | N/A |
| Type: | Task | Priority: | Major |
| Reporter: | Johan Wikman | Assignee: | markus makela |
| Resolution: | Done | Votes: | 0 |
| Labels: | None | ||
| Issue Links: |
|
||||||||
| Epic Link: | MaxScale MRM | ||||||||
| Sprint: | 2017-43, 2017-44 | ||||||||
| Description |
|
It seems that detecting that a master is down is not quite as straightforward as one could imagine. The RM apparently does not immediately conclude that a master that appears to be down actually is down, but makes various attempts before finally accepting that it indeed is down. That apparently is necessary in order to ensure that unnecessary promotions are not performed. |