BIND 10 #2082: bad ASIO update
BIND 10 Development
do-not-reply at isc.org
Sat Aug 25 20:04:26 UTC 2012
#2082: bad ASIO update
-------------------------------------+-------------------------------------
Reporter: | Owner:
fdupont | Status: new
Type: | Milestone:
defect | Sprint-20120904
Priority: high | Resolution:
Component: | Sensitive: 0
Unclassified | Sub-Project: Core
Keywords: | Estimated Difficulty: 3
Defect Severity: Very | Total Hours: 0
High |
Feature Depending on Ticket: |
Add Hours to Ticket: 0 |
Internal?: 0 |
-------------------------------------+-------------------------------------
Comment (by fdupont):
Replying to [comment:7 jinmei]:
> Could you clarify a bit more about the severity? My understanding is
> that there are other things that would make it impossible to build
> BIND 10 on Windows. What makes this particular issue so
> special/severe/urgent?
=> the current ASIO code doesn't fit with any ASIO distrib: some parts
were updated without keeping consitensy. The Windows problem is just a
symptom.
> I also wonder whether the corresponding change has been in a newer
> version of ASIO, or, if not, whether it makes sense to send the patch
> to the ASIO developer and ask to incorporate it.
=> IMHO (but only the person who updated (?) ASIO knows) there was a
partly
and inconsistent attempt to refresh the ASIO code using a more recent
base.
So there is nothing to send to ASIO developers, just to restart from the
last
ASIO base and to reapply the patches to adapt it to bind10.
> Although we have in fact modified our copy of ASIO to address some
> critical issues (for some of which I've reported to ASIO with a patch,
> but it didn't seem to go well unfortunately), in general I'd like to
> avoid diverting from the original code. It will make maintenance more
> difficult when we want to upgrade the ASIO version.
=> this was done during the last "update" or at the beginning? in the
second
case I already suggest what I believe it should be done. In the first case
there is a real issue because the fix breaks on at least a system so
should
be reconsider.
--
Ticket URL: <https://bind10.isc.org/ticket/2082#comment:8>
BIND 10 Development <http://bind10.isc.org>
BIND 10 Development
More information about the bind10-tickets
mailing list