| To: | Michael Monnerie <michael.monnerie@xxxxxxxxxxxxxxxxxxx> |
|---|---|
| Subject: | Re: rsync from xfs to nfs+xfs and ACL problems |
| From: | Christoph Hellwig <hch@xxxxxxxxxxxxx> |
| Date: | Wed, 19 May 2010 07:51:51 -0400 |
| Cc: | xfs@xxxxxxxxxxx, Christoph Hellwig <hch@xxxxxxxxxxxxx> |
| In-reply-to: | <201005191346.38699@xxxxxx> |
| References: | <201005181350.15308@xxxxxx> <201005191304.42314@xxxxxx> <20100519112034.GA29064@xxxxxxxxxxxxx> <201005191346.38699@xxxxxx> |
| User-agent: | Mutt/1.5.19 (2009-01-05) |
On Wed, May 19, 2010 at 01:46:38PM +0200, Michael Monnerie wrote: > On Mittwoch, 19. Mai 2010 Christoph Hellwig wrote: > > XFS stores ACLs different than other Linux filesystems because it's > > on-disk compatible to IRIX. While the other filesystems store the > > ACL in the same format that is used for the user interface XFS > > stores it in a separate trusted extended attribute. rsync should > > and > > traditionally has ignored those attributes, but it seems something in > > rsync changed recently so that it tries to copy these attributes in > > the trusted.* namespace, which fails on NFS. > > Thank you. Does that mean I can simply ignore that errors, and that > everything is OK even over NFS? Thinks should be okay - rsync probably has copied the ACLs via the normal system namespace. You can verify this by doing a getfacl on a nfs file that should havfe ACLs. |
| <Prev in Thread] | Current Thread | [Next in Thread> |
|---|---|---|
| ||
| Previous by Date: | Re: XFS on 2.6.26: reading the first 4K of a large file takes ages, Christoph Hellwig |
|---|---|
| Next by Date: | [XFS updates] XFS development tree branch, master, updated. v2.6.33-10192-ge158ccf, xfs |
| Previous by Thread: | Re: rsync from xfs to nfs+xfs and ACL problems, Michael Monnerie |
| Next by Thread: | Re: rsync from xfs to nfs+xfs and ACL problems, Linda Walsh |
| Indexes: | [Date] [Thread] [Top] [All Lists] |