-
On scenarios where an Orchestrator cluster is managing hundredths of MySQL pods, and you need to be able to disable recovery for a specific one, you need to either keep a list of all pods in `RecoverM…
-
There is a race condition in the relation between contrail-keystone-auth and contrail controller, that leaves the cluster in broken state.
It's impossible to do a automated installation of the clus…
-
# 💡 Summary #
This ticket goes back to the olden days of the tool.
There was a feature suggestion to add a PowerShell GUI to ScubaGear to lower the barrier for user entry.
With the introduction…
-
## Use Case
I'm a puppet partner and PE customers often start to evaluate bolt and eventually switch to PE plans. Then they are surprised that the (expensive) "enterprise" solution has less feature…
-
Not sure if this project is meant to be used as a starter for a workflow based project?
**1**
It's a nice and simple project layout; but I suppose that quit some refactoring is needed when you w…
-
**What**
Get the refresh token flow working.
**Why**
Currently when a token expires a user has to login again to get a new one. For surf that is less of a problem since we have a long expiration…
-
## Background
* Using a slightly patched 3.0.1 version of orchestrator. Patches don't affect functionality described below.
### Partial Screenshot from `/web/audit-recovery-steps/`
A rec…
-
As described web/status shows all the active orchestrator nodes and which one is active.
What is not shown is which db server orchestrator is actually using to store it's data. Providing this informat…
-
I have 3 Percona MySQL 5.6.29-76.2-log instances in separate VirtualBox VMs running CentOS 7.0.
The prior replication configuration was:
```
mysql56b-2
+ mysql56b-1
+ mysql56b-3
```
Upon …
-
The current orchestrator code is a complete mess.
Related to #119, #122, #123
On the lowest level we have `OrchestratorWSManager`. This manages the websocket and its socketIO interface.
Then …