Emergency update to 8.19 known issues (elastic#4922)

* Update known-issues.asciidoc

* Update docs/en/observability/apm/known-issues.asciidoc

Co-authored-by: Visha Angelova <[email protected]>

* Update docs/en/observability/apm/known-issues.asciidoc

Co-authored-by: Visha Angelova <[email protected]>

* Update known-issues.asciidoc

---------

Co-authored-by: Visha Angelova <[email protected]>

Author: theletterf

docs/en/observability/apm/known-issues.asciidoc

@@ -21,6 +21,29 @@ _Versions: XX.XX.XX, YY.YY.YY, ZZ.ZZ.ZZ_
 // If applicable, link to fix
 ////
 
+[discrete]
+== APM Integration / Server might get removed after upgrading to 8.19.0 and 9.1.0
+
+_Elastic Stack versions: 8.19.0 and 9.1.0_
+_Environments: ECH, ECE, ECK, and self-managed when running Fleet_
+
+The APM Integration might sometimes get removed from the Elastic Agent on Cloud policy.
+
+When this happens, APM intake and OTLP intake through APM Server / APM Integration will stop working.
+
+*Workaround*
+
+To work around this issue you can either:
+
+* Restart the Integration servers through Force Restart in the Cloud Admin UI.
+* Save a copy of the Elastic APM Integration policy within the Elastic Agent on Cloud policy in the Fleet UI:
+** Go to *Kibana* > *Fleet* > *Elastic Cloud agent policy*.
+** Select the *...* icon, then *Edit Integration*.
+** Add a blank space to the *Description*, then remove it.
+** Select *Save Integration*.
+
+In both cases, the settings of APM Integration are maintained. 
+
 [discrete]
 == Tail Sampling may not compact / expired TTLs as quickly as desired, causing increased storage usage.