BIND 10 #2856: memory manager initialization
BIND 10 Development
do-not-reply at isc.org
Fri Aug 2 06:48:35 UTC 2013
#2856: memory manager initialization
-------------------------------------+-------------------------------------
Reporter: jinmei | Owner: muks
Type: task | Status:
Priority: medium | reviewing
Component: shmem | Milestone:
manager | Sprint-20130806
Keywords: | Resolution:
Sensitive: 0 | CVSS Scoring:
Sub-Project: DNS | Defect Severity: N/A
Estimated Difficulty: 4 | Feature Depending on Ticket:
Total Hours: 0 | shared memory data source
| Add Hours to Ticket: 0
| Internal?: 0
-------------------------------------+-------------------------------------
Changes (by vorner):
* owner: vorner => muks
Comment:
Hello
Replying to [comment:18 muks]:
> Do you agree with the following (it's not very clear from the design
doc):
>
> When the memmgr is not in the `READY` state, if it gets notified of a
new auth subscribing to the readers group, it is fine to assume the new
auth instance is using the new mapped file and not the old one. For making
sure there is no race, memmgr should make `SegmentInfo` updates in the
main thread itself (which also handles communications) and only have the
builder in a different thread.
Thinking about that, yes, you are right. It's correct to put them all to
the current set.
So it is OK to merge this (it might be good to add the ↑ text to a
comment/docstring, though ‒ but that doesn't need another review round).
--
Ticket URL: <http://bind10.isc.org/ticket/2856#comment:19>
BIND 10 Development <http://bind10.isc.org>
BIND 10 Development
More information about the bind10-tickets
mailing list