Even if information might show not to be needed in the end, you must be aware of the fact that other people around the world that are reading your post know nothing about what you have, do or see. When you want adequate help you better try to try to give them all you can think they might need then to not give them what you think they will not need.
That might be difficult, because you even did not give me the things I asked for explicitly. I had three (3) questions You only gave something on the third one.
I am still wating for information on the first two and in the mean time will add a new one: How do you access system Y (login using what, other ways of access)?
alaios wrote:
> Dear all,
> I was copying my files on a remote server, lets’ call it X, for some
> time and I knew that these files were backedup to a server Y.
>
> It looks like that now the backup process is broken and might be two -
> three files not the same in both servers (X and Y)
>
> I was wondering if you can suggest me a tool to search for file
> differences (compare files?) between the two servers .
Why do you want to search for the differences? Why not just tell rsync
to repeat the backup?
On 11/07/2012 02:46 PM, alaios wrote:
> I thought not more information would be needed as I think this boils
> down to “Find a way to compare two folder contents”.
as an example of why more info (like that which Henk asked for) is required:
if X is openSUSE (or other Linux) and Y is openSUSE (or other Linux)
(his first question), and you have access to both (his second question)
and you are copying with rsync with switches to retain date/etc (his
third question) then you could make a little script which would ls each
and compare the two outputs with diff, to find differences… simple.
but, if X is openSUSE (or other Linux) and Y is Windows
Server/Solaris/SPARC/etc then that won’t work.
and, his fourth question: if you access via remote desktop and need a
GUI to do your server admin work, then that simple script gets . . .
> Hi,
> I thought not more information would be needed as I think this boils
> down to “Find a way to compare two folder contents”.
No, an answer is impossible unless you say what operating system each
machine runs and what type of access you have on each. For example, are
they Linux, and do you have ssh access to them?
–
Cheers / Saludos,
Carlos E. R.
(from 11.4 x86_64 “Celadon” (Minas Tirith))
It looks like that both systems run same os and kernel
3.2.0-32-generic #51-Ubuntu SMP
in both systems I have a access to files through ssh. Both directories return exactly the same number of megabytes which I think is a first indicator (even though not enough) that files might be identical.
So what is the missing part here for proving that also?
I would like to thank you in advance for your help
On 2012-11-09 16:26, alaios wrote:
>
> It looks like that both systems run same os and kernel
>
> 3.2.0-32-generic #51-Ubuntu SMP
>
>
> in both systems I have a access to files through ssh. Both directories
> return exactly the same number of megabytes which I think is a first
> indicator (even though not enough) that files might be identical.
Then I would log on both machines via ssh and run a checksum on the
files, which you can then compare visually. Or create a checksum file
(man md5sum) which you download to another machine and run there, if it
is a bunch of files.
–
Cheers / Saludos,
Carlos E. R.
(from 11.4 x86_64 “Celadon” (Minas Tirith))
I ẃould like to thank you for your reply. The problem I see is that the files are so many that comparing one by one would take ages. In that case either should I use a graphic interface (that supports ssh) or even better that would pinpoint files that are different (and that because I guess that only few files would have differences if any=)
Mass actions like your’s seldom ask for a GUI program. They ask for a program that can run from the CLI and thus is fit for batch work. A good script might be the solution, using something like ssh and sftp (over ssh) to fetch those files pair by pair, compare them and delete them again from local storage, on to the next pair. Maybe it runs for a few hours, but who cares, just do it at night and go for a drink and some sleep.
But for that you need somebody rather fluent in scripting and the usage of ssh from the batch. And that person needs to sit at your system because itt needs a lot of try and error. That means either you, or a good friend.
EDIT: Sorry Martin, didn’t see your suggestion. But yes all sorts of scripted soutions come to one’s mind.
On 2012-11-12 13:56, alaios wrote:
>
> I ẃould like to thank you for your reply. The problem I see is that
> the files are so many that comparing one by one would take ages. In that
> case either should I use a graphic interface (that supports ssh) or even
> better that would pinpoint files that are different (and that because I
> guess that only few files would have differences if any=)
No, you use md5sum to generate a list of files with their checksums.
When you do the checking in the other computer with that file, it
automatically checks all those filenames.
Am 12.11.2012 13:56, schrieb alaios:
> case either should I use a graphic interface (that supports ssh) or even
> better that would pinpoint files that are different
If you look for a gui tool: krusader should be able to do that
–
PC: oS 12.2 x86_64 | i7-2600@3.40GHz | 16GB | KDE 4.8.5 | GeForce GT 420
ThinkPad E320: oS 12.2 x86_64 | i3@2.30GHz | 8GB | KDE 4.9.3 | HD 3000
eCAFE 800: oS 11.4 i586 | AMD Geode LX 800@500MHz | 512MB | lamp server
Only works if it is a safe assumption that a file present means a file
intact. Anything short of a checksum (md5 or otherwise) or something
that verifies the contents like that (rsync can do this too) is going to
fall short if a single bit of the file is not identical on both sides.
Also, you’d at least want to sort the output since ordering isn’t the
same for every system/filesystem depending on a bunch of variables.
Good luck.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.19 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://www.enigmail.net/
On 2012-11-12 14:26, hcvv wrote:
>
> Mass actions like your’s seldom ask for a GUI program. They ask for a
> program that can run from the CLI and thus is fit for batch work. A good
> script might be the solution, using something like ssh and sftp (over
> ssh) to fetch those files pair by pair, compare them and delete them
> again from local storage, on to the next pair. Maybe it runs for a few
> hours, but who cares, just do it at night and go for a drink and some
> sleep.
That requires heavy network usage, which I thought was to be avoided.
And if what you want to check is for errors in transit, you can create
new errors…
–
Cheers / Saludos,
Carlos E. R.
(from 11.4 x86_64 “Celadon” (Minas Tirith))
Hi I tried the rsync solution as md5sum could not run recursively, and requires probably scripting to make it work.
The dry run rsync produces (edited some lines regarding host and username)
rsync --dry-run -rvce "ssh" ./ username@host:/storage/volume325/Data
The authenticity of host 'hidden' can't be established.
ECDSA key fingerprint is 62:12:d3:f2:7e:25:05:89:f4:43:fa:11:c3:ef:62:04.
Are you sure you want to continue connecting (yes/no)? yes
Warning: Permanently added hidden to the list of known hosts.
user@hosts's password:
sending incremental file list
check.md5
sent 1870426 bytes received 871 bytes 73.19 bytes/sec
total size is 1016380260676 speedup is 543142.14 (DRY RUN)
as you can see from the rsync just file a fille called check.md5 (it was created when I was trying the md5 solution proposed).
What I do not understand is the send bytes 1870426 why are so large when the check.md5 file is 0 bytes.
Could you please confirm me also that I understand right the output of rsync and all the files are indeed alike?
Am 15.11.2012 10:56, schrieb alaios:
> What I do not understand is the send bytes 1870426 why are so large
> when the check.md5 file is 0 bytes.
That means that only the file check.md5 is the sole difference between
the two folders.
As for the amount of data sent: How do you think rsync can compare files
without sending data about the files you want to compare?
–
PC: oS 12.2 x86_64 | i7-2600@3.40GHz | 16GB | KDE 4.8.5 | GeForce GT 420
ThinkPad E320: oS 12.2 x86_64 | i3@2.30GHz | 8GB | KDE 4.9.3 | HD 3000
eCAFE 800: oS 11.4 i586 | AMD Geode LX 800@500MHz | 512MB | lamp server