I did my first backup last night, and tried a recovery of a couple files
today, but it didn't seem to work.
It only seemed to write like around 40 blocks or so, maybe (of 32k each).
Well, the dd (for testing) came up like this:
mt rewind
dd if=/dev/rmt/1hn bs=32k count=1 > header.dat # tape label, apparently
mt fsf 1 # skip to next archive
dd if=/dev/rmt/1hn bs=32k skip=1 of=out1.tar.gz
# the first time this would only recover 2 blocks
dd if=/dev/rmt/1hn bs=32k of=out2.tar.gz
# this would recover 38-40 blocks, depending on how i did the previous dd
# sometimes I would do skip the skip=1 part...
It would only recover part of the file. almost as if that's all that was
dumped to tape.
I mean, /net/useruv is 268842m, so doing a small test (/usr/local/bin/tar
cvvf - /net/useruv|gzip -c > /tmp/_net_useruv.tar.gz) came up with the
following size of 273435486, yet the report said:
host disk L ORIG-KB OUT-KB COMP% MMM:SS KB/s
MMM:SS KB/s
arsd /net/useruv 0 10 32 320.0 0:00 107.7
0:02 27.3
Why?
It didn't backup nearly enough at all, only 10k it looks like.
This is on a Solaris 8 box with 256M ram and a DLT IV.
On a different system (at another location) I am running amanda on a linux
system with 128m and a DDS-3 drive (Python 04106)
I'm currently redoing the tape type just in case.
Any help would be greatly appreciated...
lance
