[Patches] The way to remove gensec_update_ev()
metze at samba.org
Wed Jun 14 15:39:30 UTC 2017
just for the record:
Contains an async version of spnego.c, it already passed:
make -j test FAIL_IMMEDIATELY=1 SOCKET_WRAPPER_KEEP_PCAP=1 TESTS="spnego"
I'm running private autobuilds now...
Then I just need to put the 128 commits into some
more useful once:-)
Am 14.06.2017 um 02:42 schrieb Andrew Bartlett:
> On Wed, 2017-06-14 at 02:04 +0200, Stefan Metzmacher wrote:
>> Am 14.06.2017 um 01:09 schrieb Andrew Bartlett:
>>> On Wed, 2017-06-14 at 00:04 +0200, Stefan Metzmacher wrote:
>>>> here's the next chunk for the LDAP server.
>>>> Also available under:
>>>> It just passed a private autobuild.
>>>> Please review and push:-)
>>> Could we have some tests added, specifically that disrupt the
>>> connection during the async processing? I'm thinking of something
>>> might do part of the handshake, ideally against a trusted domain
>>> (but I
>>> realise that isn't finished yet) and then drop the socket?
>> I don't see an easy way to write a test for it.
>> But if you have a close look at the
>> "s4:ldap_server: add call->wait_send/recv infrastructure" commit:
>> You'll see that ldapsrv_call_read_next() (and also
>> ldapsrv_call_writev_start()) is deferred until the
>> waiting is over, which means we won't detect a connection drop
>> until we write or read from the socket.
>> So the outside behavior of the connection is still the same,
>> it's blocked during a bind call. The difference is that
>> other connections are no longer also blocked.
>>> In the meantime I will look over the patches, and I thank you for
>>> continued attention to this area.
> I've had a first look over the patches, and I'll look over again.
> Absent any major concerns I plan to get this into master in the next
> day or so.
> Andrew Bartlett
-------------- next part --------------
A non-text attachment was scrubbed...
Size: 836 bytes
Desc: OpenPGP digital signature
More information about the samba-technical