I ran into an odd problem this weekend when recursively copying a large directory from an HP-UX system to a Linux system. I'm using scp (ssh) to perform the copy.
The directory I'm copying appears to be two very different sizes. When I type "du -s -k <dir>" on the HP-UX system, I get just over 629,000 Kbytes.
I copy the directory to the Linux system, using
scp -p -r <dir> <[EMAIL PROTECTED]:/dir>
After the copy completes (hours later), I find that I've copied over 1.6 Gigabytes to the Linux system.
Again, I used "du -s -k <dir>" on the Linux system to determine the size of the copied directory. The "-k" flag is supposed to eliminate any block size mismatch error.
My initial guess was that scp follows links, and there must be a lot of linked stuff being copied. I looked at the man page for scp, and didn't see any reference to how links or handled (or not). I didn't see any parameters that control this either, like tar provides.
There are a number of links in the data being copied, but I have trouble believing that it would result in nearly three times as much stuff!
Is this a "follow links" problem, and does anyone know a way around it?
Creating a tar file on the HP-UX system is out of the question; its filesystems are all nearly full.
Thanks for any help!
Scott C.
-- TriLUG mailing list : http://www.trilug.org/mailman/listinfo/trilug TriLUG Organizational FAQ : http://trilug.org/faq/ TriLUG Member Services FAQ : http://members.trilug.org/services_faq/ TriLUG PGP Keyring : http://trilug.org/~chrish/trilug.asc
