I'm having some trouble estimating dump sizes.  I have two Solaris 7
machines where I compiled tar 1.13.25 (they were previously running a
packaged 1.13 version, but the index files generated by that version were
corrupt).  It appears that when tar is told to get an estimate, it
immediately fails, and these systems appear to be the first ones to report
back to the server in the log files (even being faster than the tape label
re-writing).

amstatus reports that these disks have `no estimate', and the dumps fail.

[...]
taper: read label `Daily-06' date `20011207'
got result for host system1.csom.umn.edu disk /export/d2: 0 -> -1K, 1 -> -1K, -1 -> -1K
got result for host system1.csom.umn.edu disk /home: 0 -> -1K, 1 -> -1K, -1 -> -1K
got result for host system1.csom.umn.edu disk /opt: 0 -> -1K, 1 -> -1K, -1 -> -1K
got result for host system1.csom.umn.edu disk /usr/openwin: 0 -> -1K, 1 -> -1K, -1 -> 
-1K
got result for host system1.csom.umn.edu disk /usr: 0 -> -1K, 1 -> -1K, -1 -> -1K
got result for host system1.csom.umn.edu disk /var: 0 -> -1K, 1 -> -1K, -1 -> -1K
got result for host system1.csom.umn.edu disk /: 0 -> -1K, 1 -> -1K, -1 -> -1K
got result for host system2.csom.umn.edu disk /home: 0 -> -1K, 1 -> -1K, -1 -> -1K
got result for host system2.csom.umn.edu disk /var: 0 -> -1K, 1 -> -1K, -1 -> -1K
got result for host system2.csom.umn.edu disk /usr: 0 -> -1K, 1 -> -1K, -1 -> -1K
got result for host system2.csom.umn.edu disk /: 0 -> -1K, 1 -> -1K, -1 -> -1K
taper: wrote label `Daily-06' date `20020109'
got result for host linuxwww.csom.umn.edu disk /: 0 -> 618610K, 1 -> 40130K, -1 -> -1K
[...]

I suspect I must have built a bad version of tar somehow.  Anyone know what
the problem might be, or how to test the build of tar that I have without
going through a whole amdump cycle?

-- 
        Mike Hicks   [mailto:[EMAIL PROTECTED]]
   Unix Support Assistant    | Carlson School of Management
Office: 1-160  Phone: 6-7909 |   University of Minnesota

Attachment: msg09179/pgp00000.pgp
Description: PGP signature

Reply via email to