diff --git a/docs/rfcs/035-safekeeper-dynamic-membership-change.md b/docs/rfcs/035-safekeeper-dynamic-membership-change.md
index 8619f83ff58f..aca3b720c9af 100644
--- a/docs/rfcs/035-safekeeper-dynamic-membership-change.md
+++ b/docs/rfcs/035-safekeeper-dynamic-membership-change.md
@@ -356,7 +356,7 @@ Node management is similar to pageserver:
 2) GET `/control/v1/safekeeper` lists safekeepers.
 3) GET `/control/v1/safekeeper/:node_id` gets safekeeper.
 4) PUT `/control/v1/safekeper/:node_id/scheduling_policy` changes status to e.g.
-   `offline` or `decomissioned`. Initially it is simpler not to schedule any
+   `offline` or `decommissioned`. Initially it is simpler not to schedule any
     migrations here.
 
 Safekeeper deploy scripts should register safekeeper at storage_contorller as
diff --git a/libs/pageserver_api/src/controller_api.rs b/libs/pageserver_api/src/controller_api.rs
index 1248be0b5c60..497df9e6a849 100644
--- a/libs/pageserver_api/src/controller_api.rs
+++ b/libs/pageserver_api/src/controller_api.rs
@@ -438,7 +438,8 @@ pub enum SkSchedulingPolicy {
     Active,
     Activating,
     Pause,
-    Decomissioned,
+    #[serde(alias = "Decomissioned")]
+    Decommissioned,
 }
 
 impl FromStr for SkSchedulingPolicy {
@@ -449,10 +450,10 @@ impl FromStr for SkSchedulingPolicy {
             "active" => Self::Active,
             "activating" => Self::Activating,
             "pause" => Self::Pause,
-            "decomissioned" => Self::Decomissioned,
+            "decommissioned" | "decomissioned" => Self::Decommissioned,
             _ => {
                 return Err(anyhow::anyhow!(
-                    "Unknown scheduling policy '{s}', try active,pause,decomissioned"
+                    "Unknown scheduling policy '{s}', try active,pause,decommissioned"
                 ));
             }
         })
@@ -466,7 +467,7 @@ impl From<SkSchedulingPolicy> for String {
             Active => "active",
             Activating => "activating",
             Pause => "pause",
-            Decomissioned => "decomissioned",
+            Decommissioned => "decommissioned",
         }
         .to_string()
     }
diff --git a/storage_controller/migrations/2026-03-08-000001_fix_decommissioned_typo/down.sql b/storage_controller/migrations/2026-03-08-000001_fix_decommissioned_typo/down.sql
new file mode 100644
index 000000000000..44fdc838407b
--- /dev/null
+++ b/storage_controller/migrations/2026-03-08-000001_fix_decommissioned_typo/down.sql
@@ -0,0 +1 @@
+UPDATE safekeepers SET scheduling_policy = 'decomissioned' WHERE scheduling_policy = 'decommissioned';
diff --git a/storage_controller/migrations/2026-03-08-000001_fix_decommissioned_typo/up.sql b/storage_controller/migrations/2026-03-08-000001_fix_decommissioned_typo/up.sql
new file mode 100644
index 000000000000..c4a643ef11f6
--- /dev/null
+++ b/storage_controller/migrations/2026-03-08-000001_fix_decommissioned_typo/up.sql
@@ -0,0 +1 @@
+UPDATE safekeepers SET scheduling_policy = 'decommissioned' WHERE scheduling_policy = 'decomissioned';
diff --git a/storage_controller/src/heartbeater.rs b/storage_controller/src/heartbeater.rs
index fe916aa36aec..0fe1e200b078 100644
--- a/storage_controller/src/heartbeater.rs
+++ b/storage_controller/src/heartbeater.rs
@@ -326,7 +326,7 @@ impl HeartBeat<Safekeeper, SafekeeperState> for HeartbeaterTask<Safekeeper, Safe
 
         let mut heartbeat_futs = FuturesUnordered::new();
         for (node_id, sk) in &*safekeepers {
-            if sk.scheduling_policy() == SkSchedulingPolicy::Decomissioned {
+            if sk.scheduling_policy() == SkSchedulingPolicy::Decommissioned {
                 continue;
             }
             heartbeat_futs.push({
diff --git a/storage_controller/src/service/safekeeper_reconciler.rs b/storage_controller/src/service/safekeeper_reconciler.rs
index 7dbbd3afe4cd..abfade45d809 100644
--- a/storage_controller/src/service/safekeeper_reconciler.rs
+++ b/storage_controller/src/service/safekeeper_reconciler.rs
@@ -100,7 +100,7 @@ pub(crate) async fn load_schedule_requests(
     for (op_persist, timeline_persist) in pending_ops_timelines {
         let node_id = NodeId(op_persist.sk_id as u64);
         let Some(sk) = safekeepers.get(&node_id) else {
-            // This shouldn't happen, at least the safekeeper should exist as decomissioned.
+            // This shouldn't happen, at least the safekeeper should exist as decommissioned.
             tracing::warn!(
                 tenant_id = op_persist.tenant_id,
                 timeline_id = op_persist.timeline_id,
diff --git a/storage_controller/src/service/safekeeper_service.rs b/storage_controller/src/service/safekeeper_service.rs
index 689d341b6ad2..69bff4179a94 100644
--- a/storage_controller/src/service/safekeeper_service.rs
+++ b/storage_controller/src/service/safekeeper_service.rs
@@ -879,7 +879,7 @@ impl Service {
                     .safekeeper_reconcilers
                     .start_reconciler(node_id, self);
             }
-            SkSchedulingPolicy::Decomissioned
+            SkSchedulingPolicy::Decommissioned
             | SkSchedulingPolicy::Pause
             | SkSchedulingPolicy::Activating => {
                 locked.safekeeper_reconcilers.stop_reconciler(node_id);
@@ -1192,13 +1192,13 @@ impl Service {
             .map(|&id| NodeId(id as u64))
             .collect::<Vec<_>>();
 
-        // Validate that we are not migrating to a decomissioned safekeeper.
+        // Validate that we are not migrating to a decommissioned safekeeper.
         for sk in new_safekeepers.iter() {
             if !cur_sk_set.contains(&sk.get_id())
-                && sk.scheduling_policy() == SkSchedulingPolicy::Decomissioned
+                && sk.scheduling_policy() == SkSchedulingPolicy::Decommissioned
             {
                 return Err(ApiError::BadRequest(anyhow::anyhow!(
-                    "safekeeper {} is decomissioned",
+                    "safekeeper {} is decommissioned",
                     sk.get_id()
                 )));
             }
diff --git a/test_runner/regress/test_safekeeper_migration.py b/test_runner/regress/test_safekeeper_migration.py
index ba067b97de6a..7113ff29270a 100644
--- a/test_runner/regress/test_safekeeper_migration.py
+++ b/test_runner/regress/test_safekeeper_migration.py
@@ -118,9 +118,9 @@ def expect_fail(sk_set: list[int], match: str):
     assert len(sk_set) == 2
 
     decom_sk = [sk.id for sk in env.safekeepers if sk.id not in sk_set][0]
-    env.storage_controller.safekeeper_scheduling_policy(decom_sk, "Decomissioned")
+    env.storage_controller.safekeeper_scheduling_policy(decom_sk, "Decommissioned")
 
-    expect_fail([sk_set[0], decom_sk], "decomissioned")
+    expect_fail([sk_set[0], decom_sk], "decommissioned")
 
 
 def test_safekeeper_migration_common_set_failpoints(neon_env_builder: NeonEnvBuilder):
diff --git a/test_runner/regress/test_storage_controller.py b/test_runner/regress/test_storage_controller.py
index e11be1df8c7c..3a6f977a33a6 100644
--- a/test_runner/regress/test_storage_controller.py
+++ b/test_runner/regress/test_storage_controller.py
@@ -3752,8 +3752,8 @@ def storcon_heartbeat():
 
     wait_until(storcon_heartbeat)
 
-    # Now decomission it
-    target.safekeeper_scheduling_policy(inserted["id"], "Decomissioned")
+    # Now decommission it
+    target.safekeeper_scheduling_policy(inserted["id"], "Decommissioned")
 
 
 @run_only_on_default_postgres("this is like a 'unit test' against storcon db")
@@ -3803,8 +3803,8 @@ def safekeeper_is_active():
 
     wait_until(safekeeper_is_active)
 
-    # Now decomission it
-    target.safekeeper_scheduling_policy(inserted["id"], "Decomissioned")
+    # Now decommission it
+    target.safekeeper_scheduling_policy(inserted["id"], "Decommissioned")
 
 
 def eq_safekeeper_records(a: dict[str, Any], b: dict[str, Any]) -> bool: