Re: Survey: compatibility vs speed of rdiff-backup development

On 3/4/23 20:17, Frank Crawford wrote:
> On Sat, 2023-03-04 at 12:58 -0700, Leland Best wrote:
>> If by "non-interactive" you mean something that happens automatically
>> as, say,
>> part of the distribution's package upgrade/update process, I vote for
>> a big
>> "No".
> 
> No, I mean specifically it doesn't ask questions, such as where the
> repo is, or even if you do or don't want to run it.
> 
> 
>>
>> First, I don't see how the package manager stuff (i.e. scripts, etc.)
>> could
>> possibly figure out reliably where all one's repositories are.
>> 'rdiff-backup'
>> let's one create one _anywhere_.  There might be "system" backup
>> repositories,
>> repositories created by individual users for their own use, testing
>> repositories, who knows what?  Some may not even be available at the
>> time of
>> upgrade (removable drives, non- "auto mounted" network drives, etc.).
> 
> This is a good point, so probably does indicate it needs to be run out
> of rdiff-backup itself, when it is accessing the repo for some reason.
> 
> It could be part of rdiff-backup, or just some extra script that rdiff-
> backup calls as needed.

I can assure you that _none_ of my backup repositories are mounted while package updates are being done -- not the one on the internal disk, not the copies on the two external disks, and certainly not the copy that is rotated offsite.

The conversion should be an extra script for which rdiff-backup _prompts_ before executing when an unconverted repository is encountered. I keep backups for several different systems on the same media, and blindly converting one of those just because I happen to reference it with the wrong version of rdiff-backup could be a disaster. Really, I think the best solution would be for rdiff-backup to fail and show an error message that conversion is required. Unexpected prompts occurring during the execution of my backup scripts would be quite annoying**, and implementing such prompts for a client calling an rdiff-backup server might not be simple.

** Think: a script that takes quite some time to perform a backup of most of the system, but later invokes rdiff-backup two more times for portions of the system that are kept in separate repositories (for different retention requirements).

-- 
Bob Nichols     "NOSPAM" is really part of my email address.
                 Do NOT delete it.