Sponsored Content
Top Forums UNIX for Dummies Questions & Answers may a corrupted .gz file be repaired? Post 302176918 by lbukys on Wednesday 19th of March 2008 03:03:21 PM
Old 03-19-2008
Lightbulb fixing corrupt gzip files

[This forum posting is two years late, but I'm adding it because this is a frequently asked question and somebody out there may search for it again and find some hope.]

It is very difficult to fix gzip files corrupted by FTP ASCII transfer. The problem is that (on average) 1/256 of the bytes have 3/8 of their bits flipped, and there is no way to distinguish whether one of those CR/LF bytes was supposed to be the way it is, or got that way by the ASCII transfer - so it can't be inverted in any simple way.

So-called zip recovery programs only fix CRC/checksum errors (to avoid error messages), they don't actually fix the data. That's only useful if the file is truncated or has a bad block late in the compressed data. It doesn't work for which about 1/256 of their bytes are corrupted, throughout the file.

So for 99% of the cases, give up, find the original if you can, and re-transfer in binary mode.

Despite what I said above, there is a computationally expensive way to search for the necessary repair. However, it requires custom coding of a search heuristic, and lots of computation. Therefore it's not a turn-key process. It's only feasible if you have no other backup and the data is so critical that you're willing to invest in some custom coding. I documented some details on what it took to recover a pile of 20MB gzip files containing 250MB million-line web server logs, at my web site.
 

9 More Discussions You Might Find Interesting

1. UNIX for Dummies Questions & Answers

corrupted tar file

Hi all, I've got a corrupted tar file with some filename being like ?a=n is it possible to get them repaired? (1 Reply)
Discussion started by: klintsovi
1 Replies

2. Shell Programming and Scripting

detecting corrupted file

Hello, Newbie question: How can I detect a corrupted file from a script (ksh)? Thank you, Martin (6 Replies)
Discussion started by: starless
6 Replies

3. HP-UX

passwd file corrupted

Good Day Our HP box was hacked and the passwd file has been altered,there are only 2 user accounts active,and these dont have any administrative rights.I need to edit the passwd file to correct the su and root entries. Does any body have any suggestions as to how i can do this with out the root... (10 Replies)
Discussion started by: cantona7
10 Replies

4. UNIX for Dummies Questions & Answers

Help! passwd file corrupted

Hi, I am new to UNIX, and have recently installed Suse 9.3. I have been experimenting with all of the commands and have somehow managed to modify the default shell of the root user to an invalid file. Consequently I cannot su to the root user as I receive the 'no such file or directory' error... (2 Replies)
Discussion started by: Tony Montana
2 Replies

5. HP-UX

Could be a corrupted file?

Hello! Do you know the meaning of... "crw-rw---- 1 informix informix 64 0x020001 Jan 21 2004 rifxroot" I don't know what the first "c" means. Furthermor, if I try to copy this file (rifxroot) it appears a message: "cp: cannot open rifxroot: No such device or address" I don't... (1 Reply)
Discussion started by: kaugrs
1 Replies

6. Solaris

WTMPX File corrupted

Hi All I work on solaris 8, 9 and 10 platforms and have encountered an error which is my wtmpx files appear to be corrupted as all entries contain the date 1970 (the birth of unix). Now this is obviously not the case, so my query is: 1 - Can the existing wtmpx files be manipulated to... (6 Replies)
Discussion started by: drestarr96
6 Replies

7. Red Hat

.bash_profile file corrupted

Hi, Unexpectedly i entered wrong entries in .bash_profile for my user which has administrative permissions. So, i am getting errors for every command. I dont have backup file also, so any body can help me how to recover it. Regards, Mastan (7 Replies)
Discussion started by: mastansaheb
7 Replies

8. Web Development

Mysql table is marked as crashed and should be repaired

140312 13:43:54 /usr/libexec/mysqld: Table './***/phpbb_posts' is marked as crashed and should be repaired Its mysqld.log in var/log alot of messages, but before around hour i tried to "repaid table" from within phpmyadmin, but appears it has no effect.. why? How to fix? (1 Reply)
Discussion started by: postcd
1 Replies

9. UNIX for Dummies Questions & Answers

After Ftp'ing file to destination how to check the file if it is in correct ASCII and not corrupted

Hi Folks, While transferring file from FTP software like Filezilla the files gets corrupted. Is there any way I can check if the recently transferred file is in ASCII and not corrupted. I have tried using file -i filename command which does tell if the file character set is ASCII or binary... (6 Replies)
Discussion started by: Khan28
6 Replies
par2(1) 						      General Commands Manual							   par2(1)

NAME
par2 - PAR 2.0 compatible file verification and repair tool. SYNOPSIS
par2 c(reate) [options] <par2 file> [files] par2 v(erify) [options] <par2 file> [files] par2 r(epair) [options] <par2 file> [files] Or: par2create [options] <par2 file> [files] par2verify [options] <par2 file> [files] par2repair [options] <par2 file> [files] DESCRIPTION
par2 is a program for creating and using PAR2 files to detect damage in data files and repair them if necessary. It can be used with any kind of file. OPTIONS
-b<n> Set the Block-Count -s<n> Set the Block-Size (Don't use both -b and -s) -r<n> Level of Redundancy (%) -c<n> Recovery block count (don't use both -r and -c) -f<n> First Recovery-Block-Number -u Uniform recovery file sizes -l Limit size of recovery files (Don't use both -u and -l) -n<n> Number of recovery files (Don't use both -n and -l) -m<n> Memory (in MB) to use -v [-v] Be more verbose -q [-q] Be more quiet (-qq gives silence) -- Treat all remaining CommandLine as filenames EXAMPLES
With PAR 2.0 you can create PAR2 recovery files for as few as 1 or as many as 32768 files. If you wanted to create PAR1 recovery files for a single file you are forced to split the file into muliple parts and RAR is frequently used for this purpose. You do NOT need to split files with PAR 2.0. To create PAR 2 recovery files for a single data file (e.g. one called test.mpg), you can use the following command: par2 create test.mpg If test.mpg is an 800 MB file, then this will create a total of 8 PAR2 files with the following filenames (taking roughly 6 minutes on a PC with a 1500MHz CPU): test.mpg.par2 - This is an index file for verification only test.mpg.vol00+01.par2 - Recovery file with 1 recovery block test.mpg.vol01+02.par2 - Recovery file with 2 recovery blocks test.mpg.vol03+04.par2 - Recovery file with 4 recovery blocks test.mpg.vol07+08.par2 - Recovery file with 8 recovery blocks test.mpg.vol15+16.par2 - Recovery file with 16 recovery blocks test.mpg.vol31+32.par2 - Recovery file with 32 recovery blocks test.mpg.vol63+37.par2 - Recovery file with 37 recovery blocks The test.mpg.par2 file is 39 KB in size and the other files vary in size from 443 KB to 15 MB. These par2 files will enable the recovery of up to 100 errors totalling 40 MB of lost or damaged data from the original test.mpg file when it and the par2 files are posted on UseNet. When posting on UseNet it is recommended that you use the "-s" option to set a blocksize that is equal to the Article size that you will use to post the data file. If you wanted to post the test.mpg file using an article size of 300 KB then the command you would type is: par2 create -s307200 test.mpg This will create 9 PAR2 files instead of 8, and they will be capable of correcting up to 134 errors totalling 40 MB. It will take roughly 8 minutes to create the recovery files this time. In both of these two examples, the total quantity of recovery data created was 40 MB (which is 5% of 800 MB). If you wish to create a greater or lesser quantity of recovery data, you can use the "-r" option. To create 10% recovery data instead of the default of 5% and also to use a block size of 300 KB, you would use the following command: par2 create -s307200 -r10 test.mpg This would also create 9 PAR2 files, but they would be able to correct up to 269 errors totalling 80 MB. Since twice as much recovery data is created, it will take about 16 minutes to do so with a 1500MHz CPU. The "-u" and "-n" options can be used to control exactly how many recovery files are created and how the recovery blocks are distributed amoungst them. They do not affect the total quantity of recovery data created. The "-f" option is used when you create additional recovery data. e.g. If you have already created 10% and want another 5% then you migh use the following command: par2 create -s307200 -r5 -f300 test.mpg This specifies the same block size (which is a requirement for additional recovery files), 5% recovery data, and a first block number of 300. The "-m" option controls how much memory par2 uses. It defaults to 16 MB unless you override it. CREATING PAR2 FILES FOR MULTIPLE DATA FILES When creating PAR2 recovery files form multiple data files, you must specify the base filename to use for the par2 files and the names of all of the data files. If test.mpg had been split into multiple RAR files, then you could use: par2 create test.mpg.rar.par2 test.mpg.part*.rar The files filename "test.mpg.rar.par2" says what you want the par2 files to be called and "test.mpg.part*.rar" should select all of the RAR files. VERIFYING AND REPAIRING When using par2 recovery files to verify or repair the data files from which they were created, you only need to specify the filename of one of the par2 files to par2. e.g.: par2 verify test.mpg.par2 This tells par2 to use the information in test.mpg.par2 to verify the data files. par2 will automatically search for the other par2 files that were created and use the information they contain to determine the filenames of the original data files and then to verify them. If all of the data files are ok, then par2 will report that repair will not be required. If any of the data files are missing or damaged, par2 will report the details of what it has found. If the recovery files contain enough recovery blocks to repair the damage, you will be told that repair is possible. Otherwise you will be told exactly how many recovery blocks will be required in order to repair. To carry out a repair use the following command: par2 repair test.mpg.par2 This tells par2 to verify and if possible repair any damaged or missing files. If a repair is carried out, then each file which is repaired will be re-verified to confirm that the repair was successful. MISSNAMED AND INCOMPLETE DATA FILES If any of the recovery files or data files have the wrong filename, then par2 will not automatically find and scan them. To have par2 scan such files, you must include them on the command line when attempting to verify or repair. e.g.: par2 r test.mpg.par2 other.mpg This tells par2 to scan the file called other.mpg to see if it contains any data belonging to the original data files. If one of the extra files specified in this way is an exact match for a data file, then the repair process will rename the file so that it has the correct filename. Because par2 is designed to be able to find good data within a damaged file, it can do the same with incomplete files downloaded from UseNet. If some of the articles for a file are missing, you should still download the file and save it to disk for par2 to scan. If you do this then you may find that you can carry out a repair in a situation where you would not otherwise have sufficient recovery data. You can have par2 scan all files that are in the current directory using a command such as: par2 r test.mpg.par2 * WHAT TO DO WHEN YOU ARE TOLD YOU NEED MORE RECOVERY BLOCKS If par2 determines that any of the data files are damaged or missing and finds that there is insufficient recovery data to effect a repair, you will be told that you need a certain number of recovery blocks. You can obtain these by downloading additional recovery files. In order to make things easy, par2 files have filenames that tell you exactly how many recovery blocks each one contains. Assuming that the following command was used to create recovery data: par2 c -b1000 -r5 test.mpg Then the recovery files that are created would be called: test.mpg.par2 test.mpg.vol00+01.par2 test.mpg.vol01+02.par2 test.mpg.vol03+04.par2 test.mpg.vol07+08.par2 test.mpg.vol15+16.par2 test.mpg.vol31+19.par2 The first file in this list does not contain any recovery data, it only contains information sufficient to verify the data files. Each of the other files contains a different number of recovery blocks. The number after the '+' sign is the number of recovery blocks and the number preceding the '+' sign is the block number of the first recovery block in that file. If par2 told you that you needed 10 recovery blocks, then you would need "test.mpg.vol01+02.par2" and "test.mpg.vol07+08.par". You might of course choose to fetch "test.mpg.vol15+16.par2" instead (in which case you would have an extra 6 recovery blocks which would not be used for the repair). NOTES
This version of par2 does not support recording path information for files. Whilst you can create recovery files for files from multiple locations, it will expect all files to be in the current directory when verifying and repairing. This limitation will be corrected in an update. REED SOLOMON CODING
PAR2 uses Reed Solomon Coding to perform its calculations. For details of this coding technique try the following link: ``A Tutorial on Reed-Solomon Coding for Fault-Tolerance in RAID-like Systems'' <http://www.cs.utk.edu/~plank/plank/papers/CS-96-332.html> AUTHOR
par2 was written by Peter Brian Clements <peterbclements@users.sourceforge.net>. This man page was contributed by Andres Salomon <dilinger@voxel.net> for the Debian GNU/Linux system (but may be used by others). May 2004 par2(1)
All times are GMT -4. The time now is 12:12 AM.
Unix & Linux Forums Content Copyright 1993-2022. All Rights Reserved.
Privacy Policy