Jump to content

Can anyone confirm the general theory here?


Humptydank

Recommended Posts

Hi folks --

 

Sorry, as I mentioned in a previous post I'm a newbie at this stuff, but still have the responsibility for evaluating a backup solution for our small company. So I apologize for my learning curve here.

 

I'd just like to make sure I understand the general theory at work here:

 

-- Am I correct in describing the file portion of a Retrospect backup set as an unduplicated collection of files on all the machines backed up to that set? So files that are on every machine but are exactly the same (like system files would be) are only added once to the set?

 

-- But if I backup the same machines to a new backup set (the next night, for example), all those files are copied again to the new set. So if I'm creating a new backup set every night for a week, then I have to wait a week to come back around to the first set for a "progressive" backup to take place?

 

-- Am I correct in describing a Snapshot as just an index, indicating which of the many files in a Backup Set's file collection existed on a particular machine at a particular point in time?

 

-- So if a file that is already in the backup set is revised, it sounds like it is simply added to the backup set again as a unique instance, not trying to be smart about whether it's a new file or not, Retrospect simply says "this doesn't match anything in the Backup Set's collection" and adds it. Conversely, if it finds a file anywhere, on any machine, that matches exactly a file in its collection it doesn't add it. It then lets the Snapshots sort out what goes where if it needs to.

 

-- Are Retrospect Catalog Files then essentially the Table of Contents to a Backup Set, listing all the unique files and Snapshots contained in a set? That seems to be the case, since it can be re-created from the Backup Set itself.

 

-- If the above is correct, then it sounds like a Progressive Backup looks at a machine's files, adds any new/changed files to the Backup Set in their entirety as new instances, and then stores a Snapshot and moves on. Is that accurate?

 

Just quick answers are all I'm asking for here, thanks so much for any and all help.

 

-- Dave

Link to comment
Share on other sites

Hi

 

You are on the right track.

 

A backup set is just a "bucket 0' files". A big container with a bunch of files in it and no folders or folder structure.

 

The catalog file is an index of which files exist in the backup set and thier corresponding attributes.

 

Snapshots are an image of each backup volume that records the folder structure of the disk at the time of backup, the files located on said disk, the permmisions of said files and the system state information. The snapshot is the "map" Retrospect uses to pull files out of the backup set and restore them to thier previous locations.

 

Thanks

Nate

Link to comment
Share on other sites

  • 5 months later...

I am having some trouble understanding how to audit Retrospect's operation. I am a brand-spanking new user.

 

I have a Volume with 24359 Files and 1574 Folders (25933 "entities").

 

I did a backup to a file backup set of that volume. Due to a couple of power outages, that backup required 3 sessions. Each of the sessions backed up 1926, 11207, and 9939 files respectively (total of 23072 files).

 

Summary says I backed up 23072 files. Sounds okay.

 

Restore (method that restores all files/folder, including empty ones) puts back 24359 files and 1574 files. Cool.

 

So folders aren't really backed up - got that. But what about files? Duplicates are not backed up twice, right? But I've catalogged (via DOS commands and other means) all the duplicate files in the Volume and I can't for the life of me come with 23072 files (usually I come up with less when subtracting duplicates, even if I match dates, sizes and names).

 

What represents the difference between the 23072 files backed up and the 24359 files that get restored? (I know, why do I care? It works... I'm just curious).

 

- Sean

Link to comment
Share on other sites

You might try:

 

1. At a Command prompt, for each volume

 

CD to the volume's root

Do a dir/s/-p > filelist.txt

 

2. Run a recycle backup to see how many files/bytes Retrospect reports as having been backed up for each volume.

 

3. do needed restores and see what gets reported.

 

Note: I've been thinking of making available a program that will create a list of the file name extensions on each drive, hiving the following info for each file name extension:

 

a. Number of files

b. Number of bytes

c. % of bytes on the volume

 

The list is sorted by the cumulative percentages,

 

Columns in the list are tab delimited so one can copy/open with Word/Excel and format as desired.

 

I'll post the program if folkes are interested.

Link to comment
Share on other sites

Quote:

 

 

Note: I've been thinking of making available a program that will create a list of the file name extensions on each drive, hiving the following info for each file name extension:

 

 

 

a. Number of files

 

b. Number of bytes

 

c. % of bytes on the volume

 

 

 

The list is sorted by the cumulative percentages,

 

 

 

Columns in the list are tab delimited so one can copy/open with Word/Excel and format as desired.

 

 

 

I'll post the program if folkes are interested.

 


 

 

 

I've posted the program at http://www.standards.com/index.html?GetFileTypeDistribution

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

×
×
  • Create New...