Merge ~silverdrake11/landscape-bundles:scalable-edge into landscape-bundles:scalable-edge

[Kevin Nasto]

Updated rabbitmq charm option to prevent timeouts and update new postgres charm options

[Spencer Runde (spencerrunde)]

After restarting Landscape and Postgres units, do you find that you get 503s due to schema check failures? Restarting the Landscape services inside the unit with `systemctl restart landscape*` fixes it, and I can verify that the Postgres connections aren't saturated.

Perhaps this is just a quality-of-life issue around waiting for the Postgres unit to come up before running schema checks. I've also noticed that the failed schema check results in `active (exited)` for the systemd processes, which doesn't properly reflect a failed state in juju for the Landscape unit.

[Kevin Nasto (silverdrake11)]

I haven't encountered the issue before. What do you mean after restarting the Landscape and Postgres units? Like running "lxc restart" or sudo reboot?

[Spencer Runde (spencerrunde)]

Right, `lxc restart` is how I did it.

Unmerged commits

f48db91... by Kevin Nasto on 2024-05-30

Rabbitmq timeout issue and new postgres charm options

diff --git a/bundle.yaml b/bundle.yaml
index a56d9d1..ea516b7 100644
--- a/bundle.yaml
+++ b/bundle.yaml
@@ -25,10 +25,15 @@ applications:
     channel: 14/edge
     num_units: 1
     constraints: mem=2048
+    options:
+      plugin_plpython3u_enable: True
+      experimental_max_connections: 500
   rabbitmq-server:
     channel: latest/edge
     charm: ch:rabbitmq-server
     num_units: 1
+    options:
+      consumer-timeout: 259200000
 relations:
     - [landscape-server, rabbitmq-server]
     - [landscape-server, haproxy]