[MISC] Sync main to 16/edge

.github/renovate.json5

@@ -15,6 +15,5 @@
       allowedVersions: '<2.0.0',
     },
   ],
-  customManagers: [
-  ],
+  customManagers: [],
 }

.github/workflows/integration_test.yaml

@@ -84,17 +84,8 @@ jobs:
     needs:
       - collect-integration-tests
     runs-on: ${{ matrix.job.runner }}
-    timeout-minutes: 226  # Sum of steps `timeout-minutes` + 5
+    timeout-minutes: 216  # Sum of steps `timeout-minutes` + 5
     steps:
-      - name: Free up disk space
-        timeout-minutes: 10
-        run: |
-          printf '\nDisk usage before cleanup\n'
-          df --human-readable
-          # Based on https://github.com/actions/runner-images/issues/2840#issuecomment-790492173
-          rm -r /opt/hostedtoolcache/
-          printf '\nDisk usage after cleanup\n'
-          df --human-readable
       - name: Checkout
         timeout-minutes: 3
         uses: actions/checkout@v5

.github/workflows/release.yaml

@@ -40,7 +40,7 @@ jobs:
       - ci-tests
     uses: canonical/data-platform-workflows/.github/workflows/release_charm_edge.yaml@v35.0.2
     with:
-      track: ${{ needs.tag.outputs.track }}
+      track: 16
       artifact-prefix: ${{ needs.ci-tests.outputs.artifact-prefix }}
     secrets:
       charmhub-token: ${{ secrets.CHARMHUB_TOKEN }}

spread.yaml

@@ -82,6 +82,9 @@ backends:
       sudo passwd -d runner
 
       ADDRESS localhost
+
+      sudo mkdir -p /var/snap/lxd/common/lxd/storage-pools
+      sudo mount --bind /mnt /var/snap/lxd/common/lxd/storage-pools
     # HACK: spread does not pass environment variables set on runner
     # Manually pass specific environment variables
     environment:

src/charm.py

@@ -1040,7 +1040,11 @@ def _reconfigure_cluster(self, event: HookEvent | RelationEvent) -> bool:
                 return False
             if ip_to_remove in self.members_ips:
                 self._remove_from_members_ips(ip_to_remove)
-        self._add_members(event)
+        try:
+            self._add_members(event)
+        except Exception:
+            logger.debug("Deferring on_peer_relation_changed: Unable to add members")
+            return False
         return True
 
     def _update_member_ip(self) -> bool:

templates/patroni.yml.j2

@@ -53,6 +53,7 @@ bootstrap:
     retry_timeout: 10
     maximum_lag_on_failover: 1048576
     synchronous_mode: true
+    synchronous_mode_strict: false
     synchronous_node_count: {{ synchronous_node_count }}
     postgresql:
       use_pg_rewind: true

tests/integration/ha_tests/helpers.py

@@ -241,7 +241,9 @@ async def is_cluster_updated(
 
     # Verify that no writes to the database were missed after stopping the writes.
     logger.info("checking that no writes to the database were missed after stopping the writes")
-    total_expected_writes = await check_writes(ops_test, use_ip_from_inside)
+    for attempt in Retrying(stop=stop_after_attempt(3), wait=wait_fixed(5), reraise=True):
+        with attempt:
+            total_expected_writes = await check_writes(ops_test, use_ip_from_inside)
 
     # Verify that old primary is up-to-date.
     logger.info("checking that the former primary is up to date with the cluster after restarting")