My TODO before RC1
ricky.nance at weaubleau.k12.mo.us
Mon Sep 10 13:50:17 MDT 2012
This looks like an issue Andrew Bartlet is already working on. I was
supposed to get him more logs last night, but my VPN was being cranky, so
hopefully tonight I can get him the right info to fix that problem.
On Sep 10, 2012 12:04 PM, "Stefan (metze) Metzmacher" <metze at samba.org>
> Am 10.09.2012 19:02, schrieb Stefan (metze) Metzmacher:
> > Am 10.09.2012 14:59, schrieb Andrew Bartlett:
> >> Metze,
> >> You asked for my TODO list before (and after, at this rate) RC1 to see
> >> if others might be able to help.
> >> It is (in rough order):
> >> - Call for last changes to our WHATSNEW
> >> - Find out if what (else!) folks need to be in RC1 beyond what is
> >> already discussed.
> >> - Sort out ACLs after classicupgrade. Essentially I think the issue is
> >> that the idmap isn't reflexive. I'm trying to implement the posix ACL
> >> hash stuff we discussed a couple of months ago.
> >> - have 'waf dist' run source3/autogen.sh and incorporate the outputs.
> >> - this would be easier if that script would run with builddir !=
> >> srcdir so we could just point it at a blank directory and collect the
> >> outputs (but certainly don't have the energy for that).
> >> - Merge the outstanding patches from Christof Schmitt
> >> <christof.schmitt at us.ibm.com>.
> >> - The PAC patch looks good. However, as has been ovbious over the
> >> past week, I'm a stickler for testing, and I'm still waiting for the
> >> tests for the PAC patch. I prepared the framework in winbind.pac but I
> >> was leaving this to Chrisof to finish.
> >> - The rfc2307 idmap patch seems like a reasonable idea, except for the
> >> user/group suffixes (I still dislike them as a concept, even if they are
> >> required). However, I again would like some tests - particularly
> >> because we already have folks complaining that idmap_ad doesn't work,
> >> and re-using the tests in that configuration could prove that either way
> >> (idmap_ad is also modified by the patch).
> >> - Do the DNS tests that I promised Kai earlier this evening
> >> - Have the WAF and autoconf config.h be identical (enough). The great
> >> work by Björn Jacke <bjacke at samba.org> to merge and finish my quota
> >> patches means we are close, with only some waf quota checks to go.
> >> - Run a wintest before doing RC1.
> >> - Look into the patch for ntlm_auth TLS channel binding. (There is no
> >> practical way i can do this for RC1, but it's almost a year since the
> >> patch was posted)
> >> - Look into the whole 'map untrusted to domain' saga properly. Make
> >> winbind return a special NTSTATUS return if the 'authoriative' flag is
> >> not set, rather than basing things on the known list of domains.
> >> - many other things I can't think of at this time of night.
> >> I plan RC1 on Wednesday, and I don't have a clone army, so clearly I
> >> won't get most of this done. Any assistance would be most welcome.
> > I've patches which move provision to 'samba-tool domain provision'
> > and move upgradeprovision to samba_upgradeprovision.
> > But somehow they fail in the openldap related provision tests...
> > Any idea?
> [1540/1549 in 1h25m31s] samba4.blackbox.provision-backend
> REASON: _StringException: _StringException: Administrator password will
> be set randomly!
> You are not root or your system do not support xattr, using tdb backend
> for attributes.
> not using extended attributes to store ACLs and other metadata. If you
> intend to use this provision in production, rerun the script as root on
> a system supporting xattrs.
> Looking up IPv4 addresses
> No IPv4 address will be assigned
> Looking up IPv6 addresses
> No IPv6 address will be assigned
> FAILED (1 failures, 0 errors and 0 unexpected successes in 0 testsuites)
> A summary with detailed information can be found in:
More information about the samba-technical