diff --git a/CHANGELOG.rst b/CHANGELOG.rst
index 003e942..2b6ed37 100644
--- a/CHANGELOG.rst
+++ b/CHANGELOG.rst
@@ -5,6 +5,19 @@ Community RouterOS Release Notes
 .. contents:: Topics
 
 
+v2.0.0-a1
+=========
+
+Release Summary
+---------------
+
+First prerelease for a new major release with a breaking change in the behavior of ``community.routeros.api``.
+
+Breaking Changes / Porting Guide
+--------------------------------
+
+- api - due to a programming error, the module never failed on errors. This has now been fixed. If you are relying on the module not failing in case of idempotent commands (resulting in errors like ``failure: already have such address``), you need to adjust your roles/playbooks. We suggest to use ``failed_when`` to accept failure in specific circumstances, for example ``failed_when: "'failure: already have ' in result.msg[0]"`` (https://github.com/ansible-collections/community.routeros/pull/39).
+
 v1.2.0
 ======
 
diff --git a/changelogs/changelog.yaml b/changelogs/changelog.yaml
index 534cbce..0e41513 100644
--- a/changelogs/changelog.yaml
+++ b/changelogs/changelog.yaml
@@ -81,3 +81,18 @@ releases:
     - 37-api-validate-cert-options.yml
     - ansible-core-_text.yml
     release_date: '2021-06-29'
+  2.0.0-a1:
+    changes:
+      breaking_changes:
+      - 'api - due to a programming error, the module never failed on errors. This
+        has now been fixed. If you are relying on the module not failing in case of
+        idempotent commands (resulting in errors like ``failure: already have such
+        address``), you need to adjust your roles/playbooks. We suggest to use ``failed_when``
+        to accept failure in specific circumstances, for example ``failed_when: "''failure:
+        already have '' in result.msg[0]"`` (https://github.com/ansible-collections/community.routeros/pull/39).'
+      release_summary: First prerelease for a new major release with a breaking change
+        in the behavior of ``community.routeros.api``.
+    fragments:
+    - 2.0.0-a1.yml
+    - 39-api-fail.yml
+    release_date: '2021-07-31'
diff --git a/changelogs/fragments/2.0.0-a1.yml b/changelogs/fragments/2.0.0-a1.yml
deleted file mode 100644
index fcbecba..0000000
--- a/changelogs/fragments/2.0.0-a1.yml
+++ /dev/null
@@ -1 +0,0 @@
-release_summary: First prerelease for a new major release with a breaking change in the behavior of ``community.routeros.api``.
diff --git a/changelogs/fragments/39-api-fail.yml b/changelogs/fragments/39-api-fail.yml
deleted file mode 100644
index f361485..0000000
--- a/changelogs/fragments/39-api-fail.yml
+++ /dev/null
@@ -1,8 +0,0 @@
-breaking_changes:
-- >-
-  api - due to a programming error, the module never failed on errors. This has now been fixed.
-  If you are relying on the module not failing in case of idempotent commands
-  (resulting in errors like ``failure: already have such address``), you need to adjust your roles/playbooks.
-  We suggest to use ``failed_when`` to accept failure in specific circumstances,
-  for example ``failed_when: "'failure: already have ' in result.msg[0]"``
-  (https://github.com/ansible-collections/community.routeros/pull/39).