[Lustre-discuss] Problem with lustre 2.0.0.1, ext3/4 and big OSTs (>8Tb)

[Joan J. Piles]

We have tried recompiling ldiskfs with ext4 enabled, and so far it seems 
to create the file systems without any further problem.

The only known issue we found is in the Release Notes:

> Enabling ext 4 allows LUNs larger than 8 TB to be used in the Lustre 
> file system.
> When ext4 is enabled, by default, in a system at scale, servers become 
> overloaded
> (cause unknown). This results in clients timing out and attempting to 
> reconnect,
> an action which the server does not accept. Eventually, the server 
> evicts the client
> due to a lock timeout.
> Workaround: Do not enable ext4 in Lustre 2.0.0.

What number of clients "a system at scale" means? We are expecting to 
have at most 1500 processes in 150 nodes accessing the filesystem. Is 
this big enough to trigger the issue?

Since is is going to be production system, using an experimental version 
is out of question. Should we sitck to 1.8 and forget about 2.0? Shall 
there be soon a 2.0.0.x release adressing these issues?

Thanks,

El 15/03/11 16:22, Kevin Van Maren escribió:
>
> Lustre 2.0 did not enable ext4 by default, due to known issues.  You 
> can rebuild the Lustre server,
> with "--enable-ext4" on the configure line, to enable it.  But if you 
> are going to use 12TB LUNs,
> you should either sick with v1.8.5 (stable), or pull a newer version 
> from git (experimental).
>
> Kevin
>
>


-- 
--------------------------------------------------------------------------
Joan Josep Piles Contreras -  Analista de sistemas
I3A - Instituto de Investigación en Ingeniería de Aragón
Tel: 976 76 10 00 (ext. 5454)
http://i3a.unizar.es -- jpiles at unizar.es
--------------------------------------------------------------------------