Grooming Policy Too Simplistic

[DavidHertzberg]

x509 and everybody else,

My new Support Case, described as a "precise re-statement of the last two sentences of the Problem Description" in x509's Support Case (he PM'ed me his Support Case number, so I could point to it in my Support Case—I can't look at his Support Case) is #71425.  To show the multi-user interest in this enhancement, I even mentioned NoelC although—from his last post in this thread—he has quit using Retrospect because of lack of this enhancement.

My Support Case is basically a cut-down copy of three posts in this thread: this post, this post (which I had to split into two Additional Notes—one of which split itself because the last sentence made it exceed the Support Case software length limit), and the third substantive paragraph of this post as another Additional Note.

IMHO x509's failure to get Tech Support to "recognize a problem or issue as something that they should submit as a feature request" illustrates the pitfalls of a lack of dialogue between the application user and the systems analyst that is inherent in Retrospect "Inc"'s Support Case system of enhancement requests.  IME the systems analyst must understand 3 separate items: what the user says he/she wants, what the user really wants, and what the user really needs.  That understanding can only be gained by a back-and-forth process, which is what we've had to some extent in this thread.  The second sentence of my Support Case specifically links to the OP of this thread, so that T.S. can see that back-and-forth process at second hand.

 

 

[x509]

David,

In my experience in my "day job" as a software product manager, some companies recognize the value of customer issues as a source of product improvement ideas, others not.  As with everything else, it comes down to the attitudes of top management.  In my case, I responded to tech support by suggesting this issue be submitted as a feature request.

[DavidHertzberg]

x509,

I've expressed myself as far as I dare to on what's going on vis-a-vis the new StorCentric top management; I have had Forums posts deleted in the past because they were deemed too critical of Retrospect or Forums-participant personnel.

However IMHO the problem in this case is that NoelC "poisoned the waterhole" with his Support Case, which was submitted prior to yours—much less mine.  I don't know what he said in it, but it seems he complained about "transient" files generated by Windows—particularly ones generated by VSS.  That explains why Retrospect Tech Support responded to your Support Case with "Unfortunately, there is no way to groom these kind of files as with every windows update, we do not know what windows will change exactly. There is no way to modify or to groom windows files."

In an Additional Note to my Support Case, I suggested that the proposed "transient" Groom feature could have the capability of grooming out previous versions of VSS-generated "snapshot"  files.  These are always created in a specific directory, and have names that differ from VSS "snapshot" to "snapshot" only in a numeric prefix in front of a leading open-curly-bracket character.  I suggested that my 3-file-merge procedure for VSS "snapshots" could sort those files by numeric-prefix-eliminated name so that they merged together, and then "groom"-out the older versions of the same file.  I didn't yet say so in my Support Case (I will), but an alternative would be to do what AOMEI Backerupper (see P.S. in this up-thread post) apparently does—which is to access some kind of an index that Microsoft maintains for VSS "snapshot" files.

[x509]

David,

I'm going to guess here that Support doesn't have enough in-depth knowledge of Windows.

 

[x509]

On 11/16/2019 at 7:47 PM, x509 said:

True enough, but there is no report that identifies those files that have been backed up N times in the last N days/weeks, as in a list that includes file name, directory, file size and dates of recent backups. And it would be even nicer if I could just check a box and Retrospect would automagically groom out these file and even add an exclude item to my backup script.  Yeah, I'm smoking the good stuff now.

I know that when I back up my PROGRAMS drive, which is Windows and installed programs, I'm backing up some daily or weekly updates for my security software and probably other utilities, and also a HUGE number of Windows files with really cryptic names.  But it's too much work to track down these files to determine which are transient.  It's "easier" to spend a few bucks on a larger backup drive.

 

Adding to an older post of mine in this thread, I think I found the source of many of those weird, transient files.  A post on Tenforms.com suggested that I look at C:\Windows\servicing\LCU.  There was one subfolder there, with a strange title that seemed to be a Windows KB article number.  Inside were over 700 folders, with those weird names, with a total of 110 thousand individual files, for a total of almost 32 GB.  All the files on my entire C drive, which is Windows and programs only, are about 127 GB, so this one maintenance patch added an additional third to the total storage used.  Microsoft Update can't be turned off for Windows Professional.  WU has weekly and sometimes more often updates, between Windows, Office, Windows Defender security and driver updates.  I don't even use Windows Defender, but I get almost daily "definitions" updates.

From a Tenforums.com thread, I found this link: https://itknowledgeexchange.techtarget.com/windows-enterprise-desktop/when-disk-cleanup-fails-use-search/. 

This link includes a table of a bunch of different Windows and Windows Server folders that can safely be deleted, and probably can be excluded from backups.  The folder I mentioned above is next-to-last in this table.

 

x509

[Nigel Smith]

Interesting...

Most "transient" files are "here today, gone tomorrow" -- think cache files etc. But, for whatever reason, Windows doesn't seem to delete these update packages after they have been used. All I can think of (aside from clumsiness by MS!) is that they are also used when you uninstall System updates.

So, to be safe, I'd probably exclude them from backups but would only delete them from disk once I was happy that I wouldn't need to uninstall.