Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!sdd.hp.com!zaphod.mps.ohio-state.edu!wuarchive!ukma!dftsrv!jagubox.gsfc.nasa.gov!jim From: jim@jagubox.gsfc.nasa.gov (Jim Jagielski) Newsgroups: comp.unix.aux Subject: TAR and CPIO (Was:Re: SCSI tape drive) Message-ID: <3914@dftsrv.gsfc.nasa.gov> Date: 14 Nov 90 18:51:12 GMT References: <681@creatures.cs.vt.edu> <1990Nov14.104544.14142@panix.uucp> Sender: news@dftsrv.gsfc.nasa.gov Reply-To: jim@jagubox.gsfc.nasa.gov (Jim Jagielski) Distribution: comp Organization: NASA/Goddard Space Flight Center Lines: 43 In article <1990Nov14.104544.14142@panix.uucp> alexis@panix.uucp (Alexis Rosen) writes: > >The problem is, "working" doesn't mean "releasable". Tar and cpio will fail >with any large job (trivial ones work). Dump and restore work, but I've seen >problems there too- restore not wanting to restore and suchlike. > I've heard a number of times that tar and cpio "fail" with large jobs... what is meant by "fail"? During the backup? During the restore? Too many files? Files too big? Too many links (symbolic or otherwise)? My system is relatively medium in size : / 53398 blocks used of 102508 total. (52.09%): /dev/dsk/c0d0s0 /usr2 59426 blocks used of 154418 total. (38.48%): /dev/dsk/c5d0s3 /usr 71524 blocks used of 190782 total. (37.49%): /dev/dsk/c0d0s2 and I backup each partition using cpio and the st driver. I've not hit a snag yet when backing-up and have not required extensive restoring yet, all- though I have done a cpio -pdmuv from /usr2 to /usr/usr2 when I reformatted /usr2 to BSD Fast File (after I updated to 2.0) and then back again to /usr2 without problems (/ and /usr are on a Wren 170, /usr2 is Quantum 80). Have I been lucky or haven't I reached the size limit yet? By the way, isn't A/UX distributed in cpio format? Therefore, cpio must be able to restore A/UX entirely from the distribution medium... true, it's less than 50Megs, but I wouldn't call it "trivial". Anyone know if any of the 3rd party tar/cpio replacements ("CTar", "LoneTar"(?) ) have been ported to A/UX? Finally, what about pax? -- ======================================================================= #include =:^) Jim Jagielski NASA/GSFC, Code 711.1 jim@jagubox.gsfc.nasa.gov Greenbelt, MD 20771 "Kilimanjaro is a pretty tricky climb. Most of it's up, until you reach the very, very top, and then it tends to slope away rather sharply."