mgr/dashboard: Use a persistent volume to store Loki DB

Current Behavior: The Loki configuration directs all data to /tmp/loki, which is not mounted as a persistent volume. This results in data being stored in a temporary location that gets cleared upon container restart.
Expected Behavior: Loki data should be stored in /loki, which is backed by the persistent volume mapped to /var/lib/ceph/<fsid>/loki.<id>/data, ensuring data persistence across container restarts.

Fixes: https://tracker.ceph.com/issues/69029

Signed-off-by: Aashish Sharma <[email protected]>

Author: Aashish Sharma

src/pybind/mgr/cephadm/templates/services/loki.yml.j2

@@ -6,11 +6,11 @@ server:
   grpc_listen_port: 8080
 
 common:
-  path_prefix: /tmp/loki
+  path_prefix: /loki
   storage:
     filesystem:
-      chunks_directory: /tmp/loki/chunks
-      rules_directory: /tmp/loki/rules
+      chunks_directory: /loki/chunks
+      rules_directory: /loki/rules
   replication_factor: 1
   ring:
     instance_addr: 127.0.0.1

src/pybind/mgr/cephadm/tests/test_services.py

@@ -1416,11 +1416,11 @@ def test_loki_config(self, _run_cephadm, cephadm_module: CephadmOrchestrator):
                   grpc_listen_port: 8080
 
                 common:
-                  path_prefix: /tmp/loki
+                  path_prefix: /loki
                   storage:
                     filesystem:
-                      chunks_directory: /tmp/loki/chunks
-                      rules_directory: /tmp/loki/rules
+                      chunks_directory: /loki/chunks
+                      rules_directory: /loki/rules
                   replication_factor: 1
                   ring:
                     instance_addr: 127.0.0.1