Hi, I wanted to know if there is a par generation tool that you can input a root directory to & will create PAR recovery volumes for each subdirectory contained.

How about searching for all files inside the root direcory and then dropping all found items in QuickPar?

The problem with that is that the par files are created in the root directory, I want each sub folder to have it's own par files.

I work with like 50 folders, & 12 files in each folder each time I back-up so it's inconvenient to have to go folder by folder with QuickPar.

Could you do this with Sweep? You feed it a command line and it executes it recursively on the directory tree.

That's true. However, creating a single par for all files results in higher efficiency.

I forgot to mention par2cmdline (http://sourceforge.net/project/showfiles.php?group_id=30568&package_id=82331&release_id=233051)

Yes, use sweep with par2cmdline.

I use that method with my music backups...

Could anyone suggest which parameters to use with par2cmdline?

& yes this is a muisc back-up to DVD-R, each folder is an album & the files are in mpc format (about 5MB per file) but I also have extraction logs, encoding logs, & CUE sheets (about 2KB per file) for each album.

sweep par2 c -r<redundancy percentage> -s<block size> *

I think this is kinda self-explanatory... This will create par2 files of all files in the directory. If you want more options to configure recovery file sizes and such, check the usage of par2.exe

This has some strange naming effects. For example, with a directory containing:

Air - Talkie Walkie.cue
Air - Talkie Walkie.log
Air - Talkie Walkie.wv

the command line generates files called "Air - Talkie Walkie.cue.par2" etc. So it most likely bases the file name off the first file in the directory list.

Wouldnt you be better off making an ISO of your completed disc and making PAR files of that? If the Disc becomes damaged, you can just rip the entire thing using ISO Buster and fix the broken parts, in theory...

That's a pretty good idea, I'll give that a try.

I'm wrestling with the same issue . I'd like to have 1 set of PARs for the entire DVD (redundancy a lot more effective). I want the PARs to be based on files, not ISO, because the ISO might not extract: see this thread (also read tgh's post on the 2nd page).
However I still want the option of using ISO (as a last alternative). Quickpar allows this: it finds PARs within an ISO (although only one set).

However since Quickpar doesn't have directory support yet, these wishes don't go together, unless I avoid directories and store all files in the root...

Westgroveg, in your situation I'd store each album folder in a ZIP file (without compression), store the ZIPs in the DVD root, together with a PAR set based on all these ZIPs. Make sure you create more PAR data than a typical ZIP file size (preferably a multiple).
To me this method is not an option because my (lossless album) folders are too large to be safe.

EDIT: rewrote post, it was not clear enough

I read it on the QuickPar forums, you might be able to find more useful information there. http://www.quickpar.org.uk/forum/

I think Ollie’s idea is a good one, recovery would probably work a lot faster if the corrupt source file is on hard disk.

After reading the thread (see link in my previous post) I no longer dare to rely on ISO only. Maybe my concern is unjustified (I hope so) but I don't bet on a horse that might be handicapped.

That thread seems to conclude that you can’t extract corrupt files from a DVD which is wrong. Also most of it has to do with corrupt PAR files.

I would extract with DVD decrypter in ISO mode, set hardware read retries to 1 & tick ignore errors. This would allow me to extract the corrupt ISO file, then I would grab my PAR files (which I would store on CD-R), copy them to the same folder as the corrupt ISO file & begin recovery.

Zipping each folder is out of the question, it’s too inconvenient

It says that extraction is impossible in case an entire sector is unreadable. Has anyone experience with extracting an ISO from a damaged DVD or CD?

This is not the case with DVD decrypter.

The sweep par2 method with parameter examples is described at www.high-quality.ch.vu - users's Audio High Quality Tutorials & Guides


Though I changed my method some time ago to create ca. 10% recovery data from all files of the entire DVD, 4.5 GB,
reason is the higher probability to restore corrupt files on the DVD.

I started writing a security article/rewrite of the par2 chapter for high-quality.ch.vu pages:
1st draw:


Security strategies against data failures, dead HDs, corrupted files on DVD.



Heat is main factor, which lets HD die.

My strategy, to avoid temperatures over 30°C, like 40°C, which are way to high:

I have big tower, and have 1 slot over and 1 slot below the HD free.
This allows natural air flow and cooling, enough for silent mode

If HD is stressed by replaygaining a lot of albums, or moving around a lot of data,
I switch on manually a fan blowing air from outside over the HD.


Recovery par2 data on HD or DVD media:


I changed my par2 way described at http://www.high-quality.ch.vu a little bit:

Now I create par2 data for all files on 1 DVD !
I copy all files to 1 direcotry, or add them in Quickpar gui,
so I have 4.5 GB data.
practical setting for slow pc, like P3-800 MHz:

block size ca. 1.5 MB,
results to ca. 3300 source blocks (of the 4.35 GB data)
redundancy, recovery data files: 10%, ca. 450 MB for each DVD,
split par2 files to 10 eg.
this will result to a computing time of 4-5 hours for P3-800.

This new way offers clearly higher probability to restore files on a damaged DVD.



Music backup strategy:

backup of the album !

1. Lossless album rip, stored on DVD 1.
2. MPC (1.14, 1.15t) --quality (7 --ms 15, 7.5 --ms 15 or best 8 --ms 15) --xlevel same album, on DVD 2.

probability, that both DVDs have errors at same music files ?
low.

MPC at those high qualities offers a nearly transparent backup, at least without annoying artefacts, even with possibility to transcode eg. to good mp3, with warranty to not get annoying or listenable transcoding artefacts, at least for casual listening, ABXing may be different.
And this at very low costs regarding media or space.






btw., any medium will die sooner or later, but big HDs sooner..
magnetical devices suffer from heat and a lot of usage.

DVDs with music are not in use so often, only when u really need it, opposite to big HD with all-in-one usage.

optical devices don't last till eternity, too, but there are cheap new developments over years, see:

CD 0.7 GB
DVD 4.35 GB
.
DVD+DL 8.5 GB
.
.
Blue-Ray ca. 30 GB ?

So, after some time re-copying will be cheap and make the data safer.

I believe this won't work since AFAIK DVDDecrypter skips bad sectors whereas we need to be replaced with dummy data instead.

I personally transfer all my files in a single drivecrypt(http://www.securstar.com/) volume and then create a single par for this. In case of damage, i copy the drivecrypt volume to my hard disk with a special program i've written that replaces bad sectors with zeros and then do the repair. I think this method provides the higher redudancy of all. More over it overcomes the fact that QuickPar does not support directories. Finally i can backup this way my files without worrying about limitations in filename lengths of cd filesystems and the lack of unicode support of cd burning software. The only drawback of this method is drivecrypt not being free.

p.s. If drivecrypt volume is damaged it can be still mounted.

I think that encrypting my back-up's is a little overkill & it would also be a little inconvenient for my purposes.



It's a shame DVD Decreptor can't do this

There is an open source disk encryption program: TrueCrypt.
http://truecrypt.sourceforge.net/
Isn't this good for you?

When you create the virtual drive, do you use encryption or compression, or the files are uncompressed and not encypted?


Yes, I feel this a key point! For ISO, is the same true? If we can extract a corrupted ISO, can it be mounted by Daemon Tools always, and can we see the file system (folders)?

I thought so too. But Peter Clements (the Quickpar author) contradicted that (cfr. again the link I gave earlier). This is actually good news and I think will go with the method of ISO + PARs based on the ISO.

@geopoul

I am just starting to learn about computer hardware so have very little knowledge about how data communication in a computer works but do you think that if the corrupt bits are removed from the file instead of replaced with null data the file becomes shorter & when recovery is attempted it will think starting at the first error everything in the file is corrupt?

That's correct.

As for DVDDecrypter there is still the problem that it supports only dvd video discs, isn't it?

edit: read my next post

No, DVD Decrypter can read any type of DVD in ISO mode not just DVD video's.

No! No! No! I was wrong! I just rechecked the link that liekloo mentioned and it seems that QuickPar can handle damaged files even if some bytes are missing! I am really impressed! I tested that myself and it works.

Well that's good to know. Then here is the final method I plan on using for creating & recovering a DVD back-up with PAR:

• Create an ISO file containing files to be backed-up
• Create PAR recovery files based on the ISO
• Burn the ISO
• Burn the PAR files
• If corruption occurs extract the corrupt ISO with DVD Decrypter
• Copy PAR files to same directory & begin recovery

It's ok, but be sure if the first ISO to burn, and the extracted ISO is same format. It's seems for me that Nero ISO format has some difference with ISObuster ISO format. I checked, and MD5 checksums was different.

Yes, I read on the QuickPAR forums that not all programs make identical ISO files.
So we will have to first burn, then extract ISO, then create PAR2s.

using burnatonce to compile and burn the ISO and isobuster to extract it you get identical MD5s... at least with CD-Rs I get this results...

Be sure you use the extract user data option in isobuster.

Don't know if this is the case for DVDs...

If we discuss Isobuster, what's the difference between "Extract CD Content" and "Extract CD Image"?

Also, what is the .tao format (Isobuster always wants to save image into that format)? It seems for me it is same as .iso, because if I simply rename .tao to .iso, that MD5 is same with ISO of DVDDecrypter.

Just a note.. Given the limitations of QuickPar, here's how I'd do it:

1. Take all the files you want to burn.
2. Run QuickPar or whatever to create a single PAR2 file from *all* of the files you're going to burn, with a whole lot of blocks in it. As many as you feel safe with.
3. Burn those files plus the single PAR2 file to the disk.

If the disk fails, then:
a) Copy off as many of the files as you can still read, and if you can get partial files then so much the better.
b) Make as good of an ISO of the disc as you can (doesn't matter if you drop data or zero it, QuickPar can cope with either), and rename the ISO to PAR2.
c) Run Quickpar and feed it that "PAR2" to recover the set.

Seems like it should work fine to me. Unless I'm missing something.

I can't see why doing both a & b? Only one of them is necessary. And why renaming iso to par2?

If i followed your scheme then in case of disc failure i would just make an iso and try to copy from the damaged disc any of par2 files. Then i would open the par2 file and feed the iso to QuickPar.

I think we now have the definitive method which I listed above, only with the modification of extracting in ISO mode with DVD Decrypter (or the program you plan on restoring the ISO with) & creating PAR files based on that, maybe even burn the PAR files with a copy of the program.

Anyone now see any potential problems with this method?

It's fine and I've started using it.
However your original idea (create ISO before burning) is worth further investigation:

Which would avoid ripping the DVD to ISO. I didn't even know this possibility existed.


Directories. Quickpar doesn't support subdirectories (yet), so you'll have to place all files in the root of the CD (DVD). If that's not a problem, this method is ideal.


a is easiest; if a fails you do b. That's what he means.
Renaming a duplicate of the ISO to *.par2 will make Quickpar consider this file as a par2 file and it will find the real par2 recovery set in it (limitation: it will find only one set).

Yes quickpar doesn't support subdirectories but this is not a problem. You can create your own directory structure any way you like it. The you do a *.* search and drag all the results into quickpar. You create the par2 files and burn them in whatever directory you like with the rest of the files. In case of disc failure you copy any of the par2 files to your hard disk and then make an iso. Double-click the par2 file, select 'Add' button to add the iso and you're done! The lack of subdirectories support is no problem.

I remember reading once in the QuickPar forums (one of the threads) that instead of creating Par2 files per album one should RAR the entire album together (or use matroska etc.) and then PAR2 them.

If someone could shed some light on this issue I'd be delighted: Is it better to PAR2 10 times for 10 albums or PAR2 once the 10 RAR files corresponding to each album? I think the more the amount of data that is PARred the better it becomes in recovery.

Also is there any known tools for MFT (master file table) backup? I recently had to go through a lot of trouble to recover the data from an almost full 300GB drive whose MFT is corrupted.

Assume that you have two files, A and B, 10 MB each and you create a 2 MB par2 file for each. Then if 3 MB of A are damaged you won't be able to do the recovery since you won't have enough recovery data (you need 3 MB but you have 2 MB). On the contrary, if you had created a 4 MB par2 file for A+B then you would have been able to perform the recovery process.

So the answer to your question is: par2 once.

I use a program created by member "bodhy" a year ago. Go to this thread for details.....I'm not sure if this program can be found anywhere, but I use it all the time and it does exactly what you're asking.

Thread 18456

Here's my method of backing up CD's to DVD-R. I'm sure it's outdated, but I don't really care:

1. Rip CD's and convert to FLAC files.
2. Sort CD's in folders based on my personal preferrence.
3. Create a single par2 recovery file for each CD using a 10% redundancy.
4. WinRAR the FLAC and PAR2 files using "store" compression method and 1% recovery data.
5. Burn *.rar files on DVD-R.

In all my years of burning CD-R and DVD-R, I've never had one go bad, nor have I ever had the need to use par2 files to recovery anything. The only time par2 files seem to be of any use is when I'm in the newgroups and a downloaded binary is corrupt or incomplete. This whole process of adding par2 files to DVD-R is probably non-sense and a waste of space in the long-run. The liklihood of ever needing them is probably not worth the time, effort, and space. JMHO.

Daffy

That's great Daffy, exactly what I originally asked for but I am now convinced that the method Ollie originally suggested in this thread would offer better protection because the PAR files would be stored on a different medium it's less likely they will also become corrupt. Also you only ever have to extract & recover 1 file which seems more convenient to me.

I think Liekloo, geopoul, & myself can all agree that if you want to create PAR files based on an ISO you should go thorugh the steps listed below:

• Burn the data to be backed-up on DVD-R
• Extract the data with DVD Decrypter in ISO mode
• Create PAR recovery files based on the ISO
• Burn the PAR files with a copy of DVD Decrypter on CD-R
• If corruption occurs extract the corrupt ISO with DVD Decrypter
• Copy PAR files to same directory & begin recovery

can somebody explain in easy words, why an ISO (or the work for it) is necessary for the process of creatring par2-recevery data at the begin ?

What is exactly the advantage of creating isos ?


Let's go through the process, if a dvd should fail:

you need to extract as much files & data as possible from the damaged dvd, even damaged files partly.

DVDdecryptor was mentioned,
can that program extract directories and (damaged) files
(and fill up damaged data with zeros, if that should be necessary ?) ?

So, if u have the iso, or all files extracted, the re-creating of damaged files should be the same ?


1 thing is unclear to me about quickpar-gui proggie for creating par2 recovery data:

imagine, you make the effort, and add each single sub-directory content of the dvd to the quickpar-gui (instead of the iso).

are the pathes of the files of any meaning ?
or would the result be the same, like you would copy all single files (assuming all files have different name) to 1 huge directory, and add this 1 directory to quickpar-gui ?

I also begin to agree, that this is the best method to archive.

I've re-read that thread what Daffy suggested:
http://www.hydrogenaudio.org/forums/index....topic=18456&hl=

What is interesting: PeterBClements created a utility, which could extract several PAR2 sets from an ISO! If that is true, it would be possible to store several PAR2 sets on one disc, perhaps on DVD (not CD). And if that PAR2 disc get damaged, we could extract not only one PAR2 set, but all!

As far as I know, that utility is not published yet. I think it's time to ask Peter at Quickpar Forum. That posts was a year ago.

Well, there are some advantages:
- If there is only one par2 set for a DVD, the effect of the par2 set is distributed on the disk.
- If the Disc Layout (same as TOC in CDs) get damaged, you can only extract to ISO, not folders and files.
- Currently Quickpar doesn't support directories, but when using ISO method, it doesn't matter.


No, DVD Decrypter is not able to extract directories and files (if not DVD-Video). It only can extract to ISO. And it can't fill data with zeros, but is not necessary.


If you can recover the whole ISO, you get everything back.


It's good question, it's also unclear to me.

thanks your your reply, it helps to sort things

1.
of course, it is very clear, that we all want to create only 1 set of par2 data for the whole medium, eg. DVD 4.35 GB.

2.1 easy method, if all files in various subdirectories have different names:
extract all files by search for *.* and move them to a temp directory, which you could add to Quickpar gui, and you are done.

2.1.2 eg., in each subdirectory, there are files with same file name.
Possibility: add each subdirectory in quickpar gui manually.
well, for ca. 10-15 Loslsess albums on 1 DVD, this could be still be possible, but not very convenient, imagine, you don't want to backup audio, but hundred small directories with various data files.


2.2: alternative: make an iso file, burn it, make the pa2 data for the iso file



3. bad case, you need the par2 data, as dvd is corrupted, maybe even the TOC.

3.1 extract the iso of the dvd, eg. by DVDdecryptor.

3.2: Can you then copy all directories and files from the extracted and mounted (Daemon tool) iso to your HD eg., so that you could start par2-recovery,
even if you created the par2 data for all *.* files (2.1) of the dvd or for all files by subdirectories of the entire dvd (2.1.2) ?


I would want to avoid the iso creation at the beginning, as that seems an extra step (even with a lot HD stress/time for copying/moving 4.35 GB data), which could be unnecessary, if 2.1/2.1.2 work.

I tried it and it works! A bit unexpected, because I expected the paths would make recovery impossible, but that worry appears to be unjustified). Thanks for pointing this out, geopaul.
This makes it a good alternative to Westgroveg's method above. Pro: no ISO extraction needed (initially). Con: during recovery, only files are recovered (no directories).


The main advantage is being able to extract partially damaged files. Windows explorer refuses to copy damaged files, so that there is a lot more data missing (you need much more recovery data).


yes the results would be the same (the current quickpar doesn't use paths).

No need to move, just search, rightclick, run Quickpar. See for yourself

yes, this sounds very good, no iso creation needed for creation par2 creation.




Liekloo:
The main advantage is being able to extract partially damaged files. Windows explorer refuses to copy damaged files, so that there is a lot more data missing (you need much more recovery data).

sorry, misunderstanding,
I asked about iso creation at the very begin before par2 data creation.
That it is better in worst case, to extract an iso of the damaged DVD, that should be clear now.
But now, it seems to be clear also, that for the creation of the par2 data set for the entire DVD, no iso is needed.
(either you make 2.1, or 2.1.2)

Wait a minute! Are you saying, that when par2 set is created from files themself (from all files on DVD), and then you extract DVD to an ISO, and feed Quickpar with that ISO, does Quickpar recover files?

This seems like a lot of work and use of media just to store data/audio on DVD-R. I would think that technology changes would render this process obsolete before you would ever need to use the PAR recovery files. That assumes, however, you use high quality DVD-R and protect your media from the elements of danger (dogs, kids, harsh weather conditions, excessive handling, etc.).

However, contrary to my opinion, if this level of protection is necessary and desired I like this approach. What settings would you use for the creation of the PAR file? I would think the block size would be a very large multiple of 2048 since you plan on using the entire CD-R to burn the PAR file and your source file is a ~ 4.5mb *.iso file. Also, if you use 700mb CD-R a 15% redundancy level can be used (700/4500). Finally, for extra protection you can buy dual DVD cases (or dual CD cases if you need the space) and keep both the DVD-R back-up and CD-R recovery disc together. All you would have to do then is wait for some type of corruption to set in to see if it was worth the effort.....oops, there I go again.