[OpenAFS] ktc_principal, smbname and Windows

[Bruno Lopes]

Our logins are based on a 'first.last' name scheme, with first and last 
being substituted for other names when that one is already taken,
the largest being 20 chars long.
I also remember some limitation on the pre-Windows 2000 login name in 
Active Directory, but it was restricted to that, pre-W2K, so we never 
took much notice, since there is no such thing here.
So I think there was never any 'blocker' to large usernames...

Rodney M Dyer wrote:

> Mr. Lopes,
>
> It looks like you are trying the same thing we did.  We never 
> encountered your problem because all of our usernames are 8 characters 
> or less.  This is because we have a history in UNIX (Sun OS and 
> Solaris) and most unix'es don't allow, or don't promote the use of 
> usernames longer than 8.  This is probably just for that reason.  Eg. 
> programmers hardcoding limits in unix code so UNIX has to live with 
> the legacy until all the code gets fixed.
>
> This has caused some aggravations because Novell and Microsoft allow 
> longer names.  Even Microsoft limits usernames to 14 characters.  In 
> this respect I'm surprised that you would have longer names, I mean 
> how are you getting logged onto the workstations?
>
> Rodney
>
>
> At 03:28 PM 4/16/2003 +0100, Bruno Lopes wrote:
>
>> Hello
>> We are using afs in our college to store the windows roaming 
>> profiles, and to do that we altered some code to aklog in order for 
>> it so work when running as SYSTEM and put the token on the correct 
>> cache ( strncpy'd the username to the smbname field of aclient in 
>> auth_to_cell)
>> Our problem now is the limitation on the smbname field , being 16 
>> char long only ( I believe), when some of our usernames are longer 
>> than that, resulting in the ticket not being placed on the correct 
>> place.
>> My questions are:
>> Why 16? Is there a reason for that number outside of afs code? It 
>> seems to be used only to tell who does the ticket belong to...
>> Can it be increased? If it can, where do we need to change it?
>> Has anyone else notice this problem?
>> Thank You
>>
>> Bruno Lopes
>>
>> _______________________________________________
>> OpenAFS-info mailing list
>> OpenAFS-info@openafs.org
>> https://lists.openafs.org/mailman/listinfo/openafs-info
>
>
> _______________________________________________
> OpenAFS-info mailing list
> OpenAFS-info@openafs.org
> https://lists.openafs.org/mailman/listinfo/openafs-info