Recovering files from corrupt TAR archive...
I am a broke college student and cannot afford to purchase the Advanced TAR Repair tool (that ironically only runs on Windows machines) to repair a 4GB tar-ball (uncompressed) that has become corrupted. Old versions of 'tar --recover' does not work and the tar program quits without recovering a SINGLE file. Here is some output of my attempts to recover data:
Code:
$ tar --version Code:
$ ./tar-1.13.25/src/tar --version |
Can you view the archive in "mc"? Perhaps not given its size, but it is worth a try. It may be that the wrong tar program was used to produce it rather than being corrupted.
Exactly how and where was this tar file produced? What does the "file" command say about this tar file? I wonder if the source for the tar program used to produce the file is available. Also, I wonder if the tar file is actually uncompressed. The message about the 64 bit headers would be seen if you forgot to use the -j option. There is a program called "star" that you may have better luck with. It seems to have more options. |
Quote:
Code:
tar: This does not look like a tar archive Quote:
Code:
$ split -b 500M huge-tar-ball.tar my-backup-files Some answers to your other questions: Quote:
new-try/new-try.tar: data Quote:
Quote:
|
HORRAY! I found out, thanks to Aaron M. Renn (creator of The gzip Recovery Toolkit ), that a useful utlility cpio can do the trick quite easily! Here is what I did:
Code:
$ cpio -ivd -H tar < new-try/new-try.tar |
Quote:
Before I re-built my server I backed all my websites through webmins FileSystem Backup which uses TAR. Now that the server is built I tried to restore using webmin the 2.5gb and 3.5gb TAR archives. I seem to be getting the same error as John above, I've tried the Advanced Recovery Software and that got me nowhere, and i've also tried using Johns solution above, that also got me nowhere. Does anyone have any ideas? Please? I'm desperate and can't afford to lose 15+ websites. I've just realised that all the backups i've been running using the webmins that Backup solution is corrupt. Many THanks Regards Bruno |
I have the same problem and I've tried Advancerepair and the guys solution none of the work, :(
I have 2 precious rare collections I want to use, but so far this broken tar is .... killing me $ cpio -ivd -H tar < new-try/new-try.tar Test - Recover/gzrt-0.4$ cpio -ivd -H tar < Jennifer.tar.gz cpio: invalid header: checksum error Nothing happens... I just wanna fix this error and NEVER use tar for a backup |
Quote:
I had the same problem with 2gb of my webspace, try to get the company hosted those site to see if there's any backup A.S.A.P... it seems that there backup only works, while the cpanel backups are currupted! |
Quote:
You can verify a tar when you take it (some limitations), but that doesn't protect against media failure later on. Nothing except multpile copies does. I recommend dar - amongst many nice features it has CRC; as per the doco Quote:
|
Quote:
I gave up yesterday, after 4 hours, i've never had such issues with rar or any other compression for that matter. I will check out dar Thanks Again |
All times are GMT -5. The time now is 09:07 AM. |