AW: [SCM] Samba Shared Repository - branch master updated

[Matthias Dieter Wallnöfer]

Jelmer,

I've added this according to the other cases - if you don't like this, then we 
should fix the other functions as well.

Cheers,
Matthias




________________________________
Von: Jelmer Vernooij <jelmer at samba.org>
An: Matthias Dieter Wallnöfer <mdw at samba.org>
CC: samba-technical at lists.samba.org
Gesendet: Dienstag, den 7. Dezember 2010, 16:19:46 Uhr
Betreff: Re: [SCM] Samba Shared Repository - branch master updated

On Tue, 2010-12-07 at 16:06 +0100, Matthias Dieter Wallnöfer wrote:
> -----------------------------------------------------------------------
> 
> Summary of changes:
>  source4/libnet/libnet_samsync_ldb.c |   12 +++++++-----
>  1 files changed, 7 insertions(+), 5 deletions(-)
> 
> 
> Changeset truncated at 500 lines:
> 
> diff --git a/source4/libnet/libnet_samsync_ldb.c 
>b/source4/libnet/libnet_samsync_ldb.c
> index 9626341..07e085e 100644
> --- a/source4/libnet/libnet_samsync_ldb.c
> +++ b/source4/libnet/libnet_samsync_ldb.c
> @@ -222,9 +222,12 @@ static NTSTATUS samsync_ldb_handle_domain(TALLOC_CTX 
>*mem_ctx,
>      /* TODO: Account lockout, password properties */
>      
>      ret = dsdb_replace(state->sam_ldb, msg, 0);
> -
> -    if (ret) {
> -        return NT_STATUS_INTERNAL_ERROR;
> +    if (ret != LDB_SUCCESS) {
> +        *error_string = talloc_asprintf(mem_ctx,
> +                        "Failed to modify domain record %s: %s",
> +                        ldb_dn_get_linearized(msg->dn),
> +                        ldb_errstring(state->sam_ldb));
> +        return NT_STATUS_INTERNAL_DB_CORRUPTION;
Isn't this a bit.. extreme? I understand that it's odd that a replace
call didn't go right, but it seems wrong to immediately attribute it to
database corruption (and get the user worried). 

CHeers,

Jelmer