[OpenAFS] Linked cells CellServDB syntax ?

stephen@physics.unc.edu stephen@physics.unc.edu
Fri, 16 Aug 2013 09:48:28 -0400 (EDT)


  This message is in MIME format.  The first part should be readable text,
  while the remaining parts are likely unreadable without MIME-aware tools.

--8323329-1088357621-1376660913=:6001
Content-Type: TEXT/PLAIN; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 8BIT

Bueller?
Bueller?...

On Tue, 13 Aug 2013, stephen@physics.unc.edu wrote:

> To followup to my own email... what's the secret to getting linked cells 
> working on Windows?
>
> I'm playing with 1.7.26 on Windows 7 Enterprise.
>
> I've tried bidirectional linking and unidirectional linking.
>
> The behavior I'm seeing is that when I'm a member of newcell and click on a 
> mountpoint for a volume in oldcell, I get the contents of the local machine's 
> dynroot ('newcell', 'oldcell', and 'thirdcell' (another cell I accessed)).
>
> Clicking on one of those "folders" just leads to a recursive rabbithole of 
> the same folder with the 3 items. I got bored after ~10 levels.
>
> Cheers,
> Stephen
>
> On Mon, 12 Aug 2013, stephen@physics.unc.edu wrote:
>
>> Thanks!
>> 
>> I didn't realize the unix client didn't allow bidirectional linkage (docs 
>> for linked cells are hard to find).
>> 
>> Anyway, I have gotten this to work now but I had to use the reverse order 
>> that you suggested.
>> 
>>> newcell oldcell #name
>>> oldcell #name
>> 
>> Otherwise I still got the original error 2.
>> 
>> Cheers,
>> Stephen
>> 
>> On Mon, 12 Aug 2013, Derrick Brashear wrote:
>> 
>>> the unix client will not allow bidirectional linkage, and you must name 
>>> the cell being linked to before the
>>> cell linking to it, in order, in the CSDB.
>>> 
>>> e.g.
>>> 
>>> oldcell #whatever
>>> newcell oldcell #whatever.
>>> 
>>> 
>>> 
>>> 
>>> On Mon, Aug 12, 2013 at 5:45 PM, <stephen@physics.unc.edu> wrote:
>>>       Hi,
>>>
>>>       Any known bugs in the parsing of CellServDB syntax for linked cells? 
>>> I ask because I seem to be
>>>       experiencing one and am trying to determine if it's a known problem 
>>> or PEBKAC (problem existing
>>>       between keyboard and chair).
>>>
>>>       I'm trying linked cells as per the docs I found at
>>>       <http://docs.openafs.org/ReleaseNotesWindows/Linked_Cells.html>. 
>>> Upon starting the client, I get
>>>       the following messages.
>>>
>>>       Starting AFS services: openafs afsd.
>>>       Adding cell 'oldcell.unc.edu': error 2
>>>       Adding cell 'newcell.unc.edu': error 2
>>>       afsd: All AFS daemons started.
>>>
>>>       Client is "Linux tasty 3.2.0-51-generic #77-Ubuntu SMP Wed Jul 24 
>>> 20:18:19 UTC 2013 x86_64
>>>       x86_64 x86_64 GNU/Linux"
>>>
>>>       Openafs version is 1.6.1+ubuntu0.2 -- just 'cuz that's the newest in 
>>> the repo.
>>>
>>>       Here's the syntax I've tried:
>>>             newcell.unc.edu oldcell.unc.edu #New Cell
>>>
>>>       152.2.4.5                        #afsdb1.newcell.unc.edu
>>>       152.2.5.6                        #afsdb2.newcell.unc.edu
>>>       152.2.6.7                        #afsdb3.newcell.unc.edu
>>>             oldcell.unc.edu newcell.unc.edu #Old Cell
>>>
>>>       152.2.1.2                        #db1.oldcell.unc.edu
>>>       152.2.2.3                        #db1.oldcell.unc.edu
>>>       152.2.3.4                        #db1.oldcell.unc.edu
>>>
>>>       Interestingly, I can configure linked cells correctly via "fs 
>>> newcell -linkedcell". So, is this
>>>       something obvious I'm doing wrong or should I open a bug report?
>>>
>>>       Cheers,
>>>       Stephen
>>>       _______________________________________________
>>>       OpenAFS-info mailing list
>>>       OpenAFS-info@openafs.org
>>>       https://lists.openafs.org/mailman/listinfo/openafs-info
>>> 
>>> 
>>> 
>>> 
>>> --
>>> Derrick
>>> 
>>> 
>> 
>
--8323329-1088357621-1376660913=:6001--