I'm on a Mac (which is already a problem; I'm unfamiliar) and this Mac
backed up a lot of data (4 TB, 900,000 files) that was in native Apple
HFS+ format. That 4 TB HFS RAID array crashed and burned, so now the
last copy of this data is in our TSM server. It exists there as HFS+
files with HFS+ style metadata. We'd like to restore this data directly
to a new CIFS share that lives on a NetApp, which will be a much better
place for it in the future.
When I attempt a test restore of a single file dsm.sys, I get:
ANS5036I DIAG: Error for file:
Parameter error
ANS5036I DIAG: Error for file: /Volumes/CIFSshare/test/dsm.sys
Parameter error
ANS5036I DIAG: Session function thread, fatal error, signal 11
(/Volumes/CIFSshare/test/dsm.sys was the full filename I was trying to
restore one test file to.)
Then it crashed and created a core file.
Anybody had to do this before? Is the Mac ISP Client (which is v7.1.5)
trying to set HFS+ style ACLs on a CIFS filesystem during the restore,
and failing? The SKIPACL option is only for backup/archive, not for
restore. Fishing for ideas here.
I want to avoid an intermediate step of restoring to an external 6 TB
HFS+ USB drive, and then using Unix cp to copy it to CIFS. That would be
a double copy of 4 TB / 900,000 files. It's already going to take a
while to restore this, and I don't want to double my time to do this.
Roger Deschner University of Illinois at Chicago rogerd AT uic DOT edu
======I have not lost my mind -- it is backed up on tape somewhere.=====
|