Accidential push of patch was: [SCM] Samba Shared Repository - branch master updated

Stefan Metzmacher metze at
Wed Jun 10 07:42:26 UTC 2020

Am 10.06.20 um 07:54 schrieb Jeremy Allison:
> On Wed, Jun 10, 2020 at 07:26:10AM +0200, Andreas Schneider via samba-technical wrote:
>> On Tuesday, 9 June 2020 19:25:03 CEST Andreas Schneider wrote:
>>> commit b458f8fbb7febf3584fa648128b4e1dc764059f7
>>> Author: Stefan Metzmacher <metze at>
>>> Date:   Wed Aug 7 12:11:58 2019 +0200
>>>     s3:libsmb: add a cache for cli_session_creds_prepare_krb5()
>>>     Signed-off-by: Stefan Metzmacher <metze at>
>>>     Reviewed-by: Andreas Schneider <asn at>
>>>     Reviewed-by: Volker Lendecke <vl at>
>> Ups, this was pushed by accident. Should I revert this or do we deal with this 
>> later?
> What effect does it have ?

The idea was to avoid doing kinit twice for a given cli_state and
cli_credentials combination.

But this doesn't have protection against pointer reusage, which can
really happen, there's a reason why I invented this commit for a similar
caching problem:

commit 0cec96526bf4d3209caf36c4a19632ff5d5dd112
Author: Stefan Metzmacher <metze at>
Date:   Fri Oct 4 10:02:56 2019 +0200

    smb2_server: make sure we detect stale smbXsrv_connection pointers
in smbXsrv_channel_global

    Pointer values can be reused (yes, I hit that during my testing!).
    Introduce a channel_id to identify connections and also add
    some timestamps to make debugging easier.

    This makes smbXsrv_session_find_channel() much more robust.

    Signed-off-by: Stefan Metzmacher <metze at>
    Reviewed-by: Guenther Deschner <gd at>

I think we should revert b458f8fbb7febf3584fa648128b4e1dc764059f7,
there will be better ways to avoid the kinit overhead.


-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 833 bytes
Desc: OpenPGP digital signature
URL: <>

More information about the samba-technical mailing list