[lustre-discuss] lustre file system errors with large user UIDs

[Dilger, Andreas]

I haven't heard of anything similar. You might try enabling full Lustre debugging ("lctl set_param debug=-1" on both the client and MDS) running strace on your commands to see where the error is generated. Then dump the Lustre debug log with "lctl dk /tmp/debug" on the client/MDS and look for "-2" (=-EPERM) error returns.

Cheers, Andreas

On Jun 25, 2017, at 09:45, John S. Urban <urbanjost at comcast.net<mailto:urbanjost at comcast.net>> wrote:

We recently added a lustre file server.  All appears to be going well except when users with a UID >  2**19  (ie. 524 288,) try to access the file
System; although no similar problems appear mounting other types of file systems (we have many others).  So the problem appears to be
Specific to the lustre mounts.  Users with the large UIDs get  “permission denied” if they try to access the file system, and commands like ls(1)
Show question marks for most fields (permission, owner, group, …).  Is there a non-default mount option required to use lustre with large UID
Values?

_______________________________________________
lustre-discuss mailing list
lustre-discuss at lists.lustre.org<mailto:lustre-discuss at lists.lustre.org>
http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.lustre.org/pipermail/lustre-discuss-lustre.org/attachments/20170626/4f0c6c77/attachment.htm>