diff --git a/bin/tests/system/resolver/tests.sh b/bin/tests/system/resolver/tests.sh
index ab94c17c10..c2c2fd4d6f 100755
--- a/bin/tests/system/resolver/tests.sh
+++ b/bin/tests/system/resolver/tests.sh
@@ -53,16 +53,14 @@ grep -F "no servers could be reached" dig.out.ns1.test${n} >/dev/null || ret=1
 if [ $ret != 0 ]; then echo_i "failed"; fi
 status=$((status + ret))
 
-# 'resolver-query-timeout' is set to 5 seconds in ns1, which is lower than the
-# current single query timeout value MAX_SINGLE_QUERY_TIMEOUT of 9 seconds, so
-# the "hung fetch" timer should kick in, interrupt the non-responsive query and
-# send a SERVFAIL answer.
+# 'resolver-query-timeout' is set to 5 seconds in ns1, so named should
+# interrupt the non-responsive query and send a SERVFAIL answer before dig's
+# own timeout fires, which is set to 7 seconds.
 n=$((n + 1))
 echo_i "checking no response handling with a longer than resolver-query-timeout timeout ($n)"
 ret=0
 dig_with_opts +tcp +tries=1 +timeout=7 noresponse.example.net @10.53.0.1 a >dig.out.ns1.test${n} || ret=1
 grep -F "status: SERVFAIL" dig.out.ns1.test${n} >/dev/null || ret=1
-grep -F "EDE: 22 (No Reachable Authority)" dig.out.ns1.test${n} >/dev/null || ret=1
 if [ $ret != 0 ]; then echo_i "failed"; fi
 status=$((status + ret))