Run one deployment orchestrator per deployment per cluster#1865

Open

joshk wants to merge 85 commits intomainfrom distributed-deployment-orchestrator

+1,327-172

Commits on Feb 4, 2025

Run one deployment orchestrator per deployment per cluster

joshk
authored and
nshoes
committed
Support devices telling the orchestrator that they might of updated

joshk
authored and
nshoes
committed
Minor cleanup

joshk
authored and
nshoes
committed
Don't forcefully clear inflight updates

joshk
authored and
nshoes
committed
Remove inflight updates where the firmware matches the device

joshk
authored and
nshoes
committed
Only message an orchestrator if the device has a deployment

joshk
authored and
nshoes
committed
Some cleanup, some docs, some encapsulation

joshk
authored and
nshoes
committed
Use singular where the event is for one deployment

joshk
authored and
nshoes
committed
Prefer Phoenix.Socket.Broadcast

joshk
authored and
nshoes
committed
Use init over handle_continue(:boot, _)

joshk
authored and
nshoes
committed
Tests and fixes from the tests

joshk
authored and
nshoes
committed
Test scheduling updating as devices come online

joshk
authored and
nshoes
committed
Remove a commented out test which has been covered

joshk
authored and
nshoes
committed
Bump the Orchestrator timer interval to 90 secs

joshk
authored and
nshoes
committed
Fix the need to have a second start_link function on the orchestrator

joshk
authored and
nshoes
committed
Remove an unneeded subscribe in a test

joshk
authored and
nshoes
committed
Allow the orchestrator to ignore the device-online

joshk
authored and
nshoes
committed
Only send the "device finished updating" broadcast if it did update

joshk
authored and
nshoes
committed
The original Monitor shouldn't have been changed

joshk
authored and
nshoes
committed
Make it easier to start a cluster in dev

joshk
authored and
nshoes
committed
Only start the clustered orchestrator on the web nodes

joshk
authored and
nshoes
committed
Switch from Horde to ProcessHub

joshk
authored and
nshoes
committed
Tweak the log message when a device updates.

joshk
authored and
nshoes
committed
If the payload is nil, use an empty map

joshk
authored and
nshoes
committed
Simplify the Monitor to a one off OrchestratorRegistration

joshk
authored and
nshoes
committed
Fix the new orchestrator tests

joshk
authored and
nshoes
committed
Add a child_spec to the OrchestratorRegistration module

joshk
authored and
nshoes
committed
Remove the Distributed.Supervisor, the extra layer isn't needed

joshk
authored and
nshoes
committed
Clean up Distributed.Orchestrator.start_link/1

joshk
authored and
nshoes
committed
Add some doc comments around the deployments orchestrator config in application.ex

joshk
authored and
nshoes
committed
Have the Orchestrator undertake threshold checks before scheduling a device to update

joshk
authored and
nshoes
committed
Nate doesn't like counting IDs, he feels it excludes other information, he likes to be inclusive.

joshk
and
nshoes
committed
Fix a missing alias after merging in main

joshk
authored and
nshoes
committed
Fix a warning from dialyzer that this code can never be reached

joshk
authored and
nshoes
committed
Remove a superfluous broadcast, and move an audit to a better place

joshk
authored and
nshoes
committed
Handle fwup error messages

joshk
authored and
nshoes
committed
Clear a devices inflight update during connection if it isn't updating

joshk
authored and
nshoes
committed
Fix skip the queue (use phx-click not phx-submit)

joshk
authored and
nshoes
committed
Test deployments starting/stopping orchestrator when updating to active or inactive
nshoes
committed
publish
nshoes
committed
Rework the duplicated libcluster config
joshk
committed
[publish]
nshoes
committed

Commits on Feb 5, 2025

Commits on Feb 6, 2025

Commits on Feb 7, 2025

Commits on Feb 8, 2025

Commits on Feb 9, 2025

New ProcessHub version allows for clean GenServer shutdown
joshk
committed

Commits on Feb 10, 2025

Commits on Feb 11, 2025