Re: dd / netcat / md5sums: hashes do not match
From: Casper H.S. Dik (Casper.Dik_at_Sun.COM)
Date: 09/20/03
- Previous message: Security Alert: "SSRT3603 wu-ftpd off by one vulnerability (rev.1)"
- In reply to: Mark Hittinger: "Re: dd / netcat / md5sums: hashes do not match"
- Next in thread: Mark Hittinger: "Re: dd / netcat / md5sums: hashes do not match"
- Reply: Mark Hittinger: "Re: dd / netcat / md5sums: hashes do not match"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ] [ attachment ]
Date: 20 Sep 2003 10:49:59 GMT
bugs@pu.net (Mark Hittinger) writes:
>Casper H.S. *** <Casper.***@Sun.COM> writes:
>>>B# nc -l -p 6789 | dd of=./hda1.img bs=512 conv=noerror,sync
>>>A# dd if=/dev/hda1 bs=512 conv=noerror,sync | nc <B's IP> 6789
>>I wouldn't use "sync" because sync causes blocks to be
>>padded with NULLs; this is certain to corrupt
>>the file.
>If his hard drive has a bad spot you need sync to fill the bad read with
>nulls. If you don't do this then the blocks on the output side will be
>mis-aligned after the bad spot.
But you should *NEVER* use sync on pipes; the above usage is just wrong
(that is, the writing with "sync" is wrong as you read from a pipe; reading
the "disk" with sync is fine)
>Now that we know he's doing this on a live file system I guess the mystery
>is solved.
Indeed. He should just run fsck afterwards.
Casper
- Previous message: Security Alert: "SSRT3603 wu-ftpd off by one vulnerability (rev.1)"
- In reply to: Mark Hittinger: "Re: dd / netcat / md5sums: hashes do not match"
- Next in thread: Mark Hittinger: "Re: dd / netcat / md5sums: hashes do not match"
- Reply: Mark Hittinger: "Re: dd / netcat / md5sums: hashes do not match"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ] [ attachment ]