A few days ago I posted:
] I do nightly dumps from several Sun and SGI workstations to an Exabyte
] 8500 drive on a Sun 4/670. I am able to restore Sun dumps without
] difficulty but I'm unable to restore an SGI dump.
]
] A typical dump from the SGI to Sun tape server, using rdump:
]
] DUMP: Date of this level 0 dump: Tue Mar 2 05:22:35 1993
] DUMP: Date of last level 0 dump: the epoch
] DUMP: Dumping /dev/rroot (/) to /dev/nrst9 on host flex1
] DUMP: mapping (Pass I) [regular files]
] DUMP: mapping (Pass II) [directories]
] DUMP: estimated 10945 tape blocks on 0.00 tape(s).
] DUMP: dumping (Pass III) [directories]
] DUMP: dumping (Pass IV) [regular files]
] DUMP: level 0 dump on Tue Mar 2 05:22:35 1993
] DUMP: Closing tape device
] DUMP: DUMP: 11061 tape blocks on 1 tape(s)
] DUMP: DUMP IS DONE
]
] When I try to restore on the Sun tape server:
]
] root@flex1(101)restore if /dev/nrst9
] Media read error: Invalid argument
]
] or if I try to rrestore on the SGI:
]
] root@flex10(104)rrestore if flex1:/dev/nrst9
] read: Invalid argument
] 22
]
] Tape read error: Unknown error
]
] Any suggestions? Do I have to use dd somehow to read the dump file from
] tape on the Sun and then run restore on this file on the SGI? If so,
] what dd parameters do I use?
]
] Please reply to me directly and I'll summarize the responses.
The overwhelming majority of responses said the problem was due to
differences in block size between the Sun and SGI. The Sun dump
command writes 512 byte blocks whereas the SGI writes 1024 byte blocks.
The solution is to specify an appropriate blocking factor on the dump
and restore commands. I was using a blocking factor of 126 for both
the Sun and SGI dumps.
I was able to restore the SGI dumps on the Sun tape server by specifying
a blocking factor of 252 (126*1024/512):
restore ibf 252 /dev/nrst9
I didn't try restoring the SGI dumps on the SGI using rrestore.
It was suggested that the problem was caused by the blocking factor
exceeding 64K bytes (126 blocks * 1024 bytes/block). The Sun blocking
factor was within the 64K limit (126 blocks * 512 bytes/block).
I've changed the SGI dumps to use a blocking factor of 63.
Nothing else regarding tape positioning, byte swapping, etc. was
necessary.
Thanks for all the suggestions.
-- Todd Pfaff \ Internet: todd@flex.eng.mcmaster.ca Dept. of Mechanical Engineering \ Voice: (416) 525-9140 x2902 McMaster University \ FAX: (416) 572-7944 Hamilton, Ontario, CANADA L8S 4L7 \
This archive was generated by hypermail 2.1.2 : Fri Sep 28 2001 - 23:07:35 CDT