Age | Commit message (Collapse) | Author | Files | Lines |
|
Guenther
(This used to be commit 1b9e526bf60372e5b3731e98dbfcc029b04e4b01)
|
|
Guenther
(This used to be commit df94fbfe96200ed521fd377a01b6b7b7a7ef88d8)
|
|
Guenther
(This used to be commit ebcbdb5fbdb30005a60431687a738df5f26d3472)
|
|
Guenther
(This used to be commit d8fe7652bda2a984d830b83762950377a4849d64)
|
|
Guenther
(This used to be commit 846d81c0ade7a1b56366feb4338312c24dc4351b)
|
|
Guenther
(This used to be commit 2114103aedbe2134624caa3aba42b6f7721e6d45)
|
|
Guenther
(This used to be commit 109e05b5117b1722a190765f84563cea24273de1)
|
|
Guenther
(This used to be commit 5922a1228653709e1ff99be50e902663e152c706)
|
|
some random bytes.
Andrew Bartlett
(This used to be commit 606c82f6254b8dc6fb32740a7bb55a1989081fe7)
|
|
require top-quality entropy. We don't want to waste system enropy
generating challenges (which simply need to be unpredictable, not
secret) or when generating UUIDs.
Rework generate_random_buffer() to use /dev/urandom less often, only
to seed the existing RC4 based PRNG. (With an exception to ensure we
don't waste this setup cost for very small entropy requests).
Perhaps we should be using heimdal's code for this instead?
This should drasticly reduce our entropy use, particularly in the
build farm (automated Samba build on hosts without much other source
of entropy).
Andrew Bartlett
(This used to be commit 6a5630d37191542022f02fae519227b7829ef620)
|
|
(This used to be commit 7416a1d606a315174b782d67463f8cec878d18bb)
|
|
code,
which also does.
(This used to be commit 083b606496308741958bb9fc6b3e50a582857677)
|
|
(This used to be commit aafd3ea03f52b08b6697f3778728659c790d03b4)
|
|
(This used to be commit b0b4668333b53991b04cae8905836c766e50aa10)
|
|
(This used to be commit 84a202754004ec618aa2663a4614d80eb2c7ce60)
|
|
system/network.h because we stripped down includes.
(This used to be commit 262c1c23a61f1f4fae13e0a61179fe98b682cecf)
|
|
metze
(This used to be commit 398940031ba4b8208d80c0e589b7235120cf3594)
|
|
metze
(This used to be commit eceed3f6d6d8df2abd264b7cc7b74ad3e1fc0758)
|
|
compilers.
Instead, as jelmer suggests, remove the depricated marker on fstring,
for as long as it is used in the samba3 winbind protocol.
Andrew Bartlett
(This used to be commit b15df834ae6506a5c585ec1ede8bf513339d9b13)
|
|
of the build farm I can break with this...
Andrew Bartlett
(This used to be commit b7e4b4f95cc3363329b2b14a4295bcb47bb30c53)
|
|
from Timur Bakeyev.
(This used to be commit 7565c4a0dd304312898839df6789f7338cc1ea63)
|
|
as part of the setup inf file.
Andrew Bartlett
(This used to be commit 6c8987464e198430885b9e71b54fed6758886fdd)
|
|
(This used to be commit 9f1e4a44b55feca9ad873f7c0a29c99e76d2ec57)
|
|
(This used to be commit 6f88c41c31271fd4b634b6393dc4ca44563a88d2)
|
|
hopefully gives more info why this fails some bigendian
platforms
metze
(This used to be commit 1d2bc79aa5841bbdbaf003005a161bbf294c7366)
|
|
Guenther
(This used to be commit a5e4d13eef9f20fdd304bff7279e56e09aa070f2)
|
|
Guenther
(This used to be commit e9e6ee4f73c5ed88cbf4585aa7e1a40e06cc131f)
|
|
Guenther
(This used to be commit 99408cf20c6feb745cd2dd56c37015cfa11e9b3d)
|
|
Guenther
(This used to be commit 617091fb5aa708f4bfef3312d1442c3fc627bd30)
|
|
Guenther
(This used to be commit 1ebcceb922bdb566e6a548aa1ad816eb8e9e26e9)
|
|
security_descriptor_create().
Guenther
(This used to be commit 7dd0d28d254f78891b0807492baafa188b42df16)
|
|
Guenther
(This used to be commit 3a0d22f572bac4f53902cf1fcb6534fe88e6edab)
|
|
Guenther
(This used to be commit ceb2ebfbf1eeb80bd34beadbba3a3c7a04da306a)
|
|
Guenther
(This used to be commit 7d8f53b1c73dc4025821d96d8f675b6866407acb)
|
|
Guenther
(This used to be commit 176614423ea57e853211c43b9853203243c6a978)
|
|
as this brings in the socket_wrapper.h in unexpected
code and we endup with a missing 'swrap_close' while linking
metze
(This used to be commit 507d5ca7d994f0771dbb66e61d2d22e391508658)
|
|
(here it's "system/time.h")
metze
(This used to be commit 00467c25930a527eb9831cfd893bd8e217ca4f74)
|
|
(This used to be commit 8fc632baaacf19602e9fc1b5b5dac47082c7ba8b)
|
|
metze
(This used to be commit 76f1f7d7017977e49d100627db728cef1a8a95c0)
|
|
mail shortly.
Andrew Bartlett
(This used to be commit 42c07d0d74a54f469ff6c8229454fc933f3bad66)
|
|
named.conf example.
Andrew Bartlett
(This used to be commit 7ecedd050038855c65d57ac66e8f6d8890038cf0)
|
|
Andrew Bartlett
(This used to be commit 9cc19bcc37eaef380c83c5a69cef38f351a0c72f)
|
|
Andrew Bartlett
(This used to be commit f189553db73c7d93d30cfa36073d86be14427f13)
|
|
Andrew Bartlett
(This used to be commit 69c31936104bd3ed19a622a0c7ef0fb36f185b17)
|
|
In response to bug #4892 by Matthias Wallnöfer <mwallnoefer@yahoo.de>
allow the objectclass module to reconstruct the objectclass hierarchy,
rather than using templates.
The issue being fixed in particular is that 'top' was not being set on
containers.
This should ensure we do this right for all objects.
Andrew Bartlett
(This used to be commit eeaa6f82492f2ec5f6595d93451c1429c4f6947e)
|
|
Andrew Bartlett
(This used to be commit 075147ff1c2679b74d6e28c3ca401453b684bee0)
|
|
into release branch.
Andrew Bartlett
(This used to be commit 413af0aa02df1d0b8b3998253d4375364cc6899a)
|
|
Andrew Bartlett
(This used to be commit 6961ab9291d719fa556d116b60d186cf1110c249)
|
|
stolen the text from the first TP. Comments welcome...
Andrew Bartlett
(This used to be commit 3db7721ae16c65d93b13f55a84946fb5d2faa5b7)
|
|
correct
Samba4 tarballs harder to screw up.
Andrew Bartlett
(This used to be commit b17e1c783216ed0b0975c1a2e7f2a6f95cd1ac33)
|