diff --git a/.github/workflows/nightly-gnu.yml b/.github/workflows/nightly-gnu.yml
index 757539340..e1a7a996f 100644
--- a/.github/workflows/nightly-gnu.yml
+++ b/.github/workflows/nightly-gnu.yml
@@ -360,13 +360,16 @@ jobs:
   # live scripts (AppRole login, three-node Raft leader failover).
   #
   # GitHub-hosted ubuntu-latest, deliberately not the self-hosted sm-standard
-  # fleet: the HA failover script needs a working Docker daemon, and the
-  # self-hosted fleet is heterogeneous — a docker-dependent workflow has been
-  # burned by it before (see the banner in e2e-s3tests.yml, rustfs/backlog#1149).
+  # fleet: both lanes need a working Docker daemon, and the sm-standard-2 pool
+  # does not provide one. #8112 rerouted these lanes there and both died in
+  # seconds on "Docker is not available" while the package itself built and
+  # published fine, which kept the nightly chain from firing (its gate
+  # requires a successful build run). See also the banner in e2e-s3tests.yml
+  # and rustfs/backlog#1149.
   kms-vault-lane:
     needs: resolve-source
     name: KMS live Vault lane
-    runs-on: sm-standard-2
+    runs-on: ubuntu-latest
     timeout-minutes: 90
     env:
       FORCE_JAVASCRIPT_ACTIONS_TO_NODE24: "true"
@@ -466,7 +469,7 @@ jobs:
   kms-vault-ha-failover:
     needs: resolve-source
     name: KMS Vault HA failover lane
-    runs-on: sm-standard-2
+    runs-on: ubuntu-latest
     timeout-minutes: 60
     env:
       FORCE_JAVASCRIPT_ACTIONS_TO_NODE24: "true"