BIND 10 #2410: race condition(?) in dhcp[46]_test.py

[BIND 10 Development]

#2410: race condition(?) in dhcp[46]_test.py
-------------------------------+-----------------------------------------
            Reporter:  jelte   |                        Owner:
                Type:  defect  |                       Status:  new
            Priority:  medium  |                    Milestone:  New Tasks
           Component:  dhcp    |                     Keywords:
           Sensitive:  0       |              Defect Severity:  N/A
         Sub-Project:  DHCP    |  Feature Depending on Ticket:
Estimated Difficulty:  0       |          Add Hours to Ticket:  0
         Total Hours:  0       |                    Internal?:  0
-------------------------------+-----------------------------------------
 while looking at testing with bamboo, we saw regular failures in
 dhcp6_test.py:

 {{{
 FAIL: test_portnumber_0 (__main__.TestDhcpv6Daemon)
 ----------------------------------------------------------------------
 Traceback (most recent call last):
   File "/test/bamboo-home/xml-data/build-
 dir/TESTPROJBIND10-TESTPLANBIND10-JOB1/src/bin/dhcp6/tests/dhcp6_test.py",
 line 156, in test_portnumber_0
     self.assertEqual( str(error).count("Failed to parse port number"), 1)
 AssertionError: 0 != 1
 }}}

 (and other similar ones from the same set)

 an earlier message may be related:
 {{{
  F/test/bamboo-home/xml-data/build-
 dir/TESTPROJBIND10-TESTPLANBIND10-JOB1/src/bin/dhcp6/tests/dhcp6_test.py:172:
 ResourceWarning: unclosed file <_io.FileIO name=13 mode='wb'>
 }}}
 (and something about no data on either stdout and stderr, but I can't find
 the exact message in the logs right now)

 Sometimes it works, sometimes it does not, I think I saw the same for
 dhcp4 at some point.

 I suspect there is a potential race condition here, where sometimes read
 is done before the process got to printing. Not sure why this is regularly
 hit here, and not on our main test bots (perhaps the initial wait is long
 enough there).

 Either way that unclosed file should probably be adressed anyway :)

-- 
Ticket URL: <http://bind10.isc.org/ticket/2410>
BIND 10 Development <http://bind10.isc.org>
BIND 10 Development