- Update the documentation of 'max-query-restarts' in the man page.

diff --git a/doc/Changelog b/doc/Changelog
index 0cc46722dba2ff571ea87715bd38fe74ae52381e..9e38cc51537c9e43fc0035815af6265f43648c1f 100644 (file)
--- a/doc/Changelog
+++ b/doc/Changelog
@@ -1,3 +1,6 @@
+13 April 2026: Yorgos
+       - Update the documentation of 'max-query-restarts' in the man page.
+
 10 April 2026: Wouter
        - Fix for EDNS client subnet so that it does not store SERVFAIL in
          the global cache after a failed lookup, such as timeouts. A failure

diff --git a/doc/unbound.conf.rst b/doc/unbound.conf.rst
index 09278be1299d4c8750fa88ff665cb24a9c337b25..a0daeb236cb97233fc92232e8a816b55e554c3e2 100644 (file)
--- a/doc/unbound.conf.rst
+++ b/doc/unbound.conf.rst
@@ -3269,6 +3269,10 @@ These options are part of the ``server:`` section.
     Hard limit on the number of times Unbound is allowed to restart a query
     upon encountering a CNAME record.
     Results in SERVFAIL when reached.
+    This applies to chained CNAME records but not sporadic CNAME records that
+    could be encountered in the lifetime of the query's resolution effort.
+    When a CNAME chain concludes, the counter keeping track of this limit is
+    reset.
     Changing this value needs caution as it can allow long CNAME chains to be
     accepted, where Unbound needs to verify (resolve) each link individually.