BIND 10 #2222: Implement counters into Xfrout (2/3)
BIND 10 Development
do-not-reply at isc.org
Tue Sep 25 01:48:46 UTC 2012
#2222: Implement counters into Xfrout (2/3)
-------------------------------------+-------------------------------------
Reporter: | Owner: vorner
naokikambe | Status: reviewing
Type: | Milestone:
enhancement | Sprint-20120925
Priority: | Resolution:
medium | Sensitive: 0
Component: | Sub-Project: DNS
xfrout | Estimated Difficulty: 4
Keywords: | Total Hours: 0
Defect Severity: N/A |
Feature Depending on Ticket: |
Add Hours to Ticket: 0 |
Internal?: 0 |
-------------------------------------+-------------------------------------
Changes (by naokikambe):
* owner: naokikambe => vorner
Comment:
Hello,
Replying to [comment:12 vorner]:
> Actually, I was not worried about the comment of the test, but about the
test
> itself. What is the purpose of the test? Is there some scenario in which
it
> could fail? I mean ‒ the test passes if it increments and it passes if
it does
> not increment.
I think that test makes sense in a very rare situation.
axfr_running can be 1 while it is between XFROUT_XFR_TRANSFER_STARTED and
XFROUT_XFR_TRANSFER_DONE. In my environment that was:
{{{
2012-09-25 09:55:08.114 INFO [b10-xfrout.xfrout]
XFROUT_XFR_TRANSFER_STARTED AXFR client [::1]:56174: transfer of zone
example.org/IN has started
...
2012-09-25 09:55:08.117 INFO [b10-xfrout.xfrout] XFROUT_XFR_TRANSFER_DONE
AXFR client [::1]:56174: transfer of example.org/IN complete
}}}
The time difference was only 0.03 seconds.
OTOH in this situation the stats module collects statistics from xfrout at
least every second. In the above case axfr_running cannot be 1 because
transferring is already done when querying statistics. However this test
would make sense if transferring takes more than one second. For that we
should use so large size of zone. Would we like to do that in lettuce?
Otherwise I think we shouldn't do that because this test doesn't make
sense in most situation.
So may I remove this test from `xfrin_notify_handling.feature`?
Regards,
--
Ticket URL: <http://bind10.isc.org/ticket/2222#comment:13>
BIND 10 Development <http://bind10.isc.org>
BIND 10 Development
More information about the bind10-tickets
mailing list