Testing library separation: now not in launcher but isolated

[anna.git] / dynamic / launcher / gx / 00001 / dynamic.suffix

diff --git a/dynamic/launcher/gx/00001/dynamic.suffix b/dynamic/launcher/gx/00001/dynamic.suffix
new file mode 100644 (file)
index 0000000..97afc7f
--- /dev/null
+++ b/dynamic/launcher/gx/00001/dynamic.suffix
@@ -0,0 +1,23 @@
+# Procedure arguments:
+#
+# <timeout>|<digits to sequence>|<CCR-Initial path file>[|CCR-Termination path file>]
+#
+# <timeout> is the first step to control the test cases execution. A value of 0 omits
+#  this first step (good to avoid huge load of timers during high performance testing).
+#
+# All the tests not fulfilled (those where the corresponding answers are not received)
+#  will be eternally keep 'in-progress' state. You could consider them failed after
+#  a specific time. A non-zero timeout step will ensure that this situation is updated
+#  as failed on expiration. Diameter application request timeouts also expire but are
+#  not processed here (not neccessary because reception conditions are enough to block
+#  the flow when something is wrong). Anyway, you could check ADML counters executing
+#  the command: ./operation show-oam
+# The main problem with a zeroed first timeout step, is that cycle repeats are not
+#  possible except if you reset the pool (hard reset) when close to finish, something
+#  very dificult to monitor.
+
+# Only CCR-Initial:
+5000|7|CCR-I.xml
+
+# CCR-Initial + CCR-Termination:
+5000|7|CCR-I.xml|CCR-T.xml