Re[2]: [rdiff-backup-users] rdiff-backup completes without errors, but then have a corrupt archive...

[listserv . traffic]

Hello Andrew,

Monday, March 30, 2009, 6:24:47 PM, you wrote:

> Greg,

> As Jason suggested, I would try using the 1.3.3 version. That version
> has the new --use-compatible-timestamps option which should clear up  
> your trouble.

> I'll try and look into whether it makes sense to clean this up on  
> Cygwin in 1.2.x. As I've mentioned before, I don't think the Cygwin  
> version is worth focusing on since the native Windows version will do
> a better job: it simply understands more NTFS data, including ACLs,  
> which aren't properly available in Cygwin.

> Isn't it possible for you to call out the native Windows version from
> your Cygwin scripts??


> regards,
> Andrew


> On Mar 30, 2009, at 2:16 AM, address@hidden wrote:

>> A few notes after some testing...
>>
>> The Native windows version doesn't exhibit the same symptoms.
>> After backing up the same tree to the same location, a "-l" lists
>> "Found 0 increments:"
>> Current Mirror ...
>>
>> Which is what I would expect.
>>
>> ---
>> This is the error I get with the cygwin version
>>
>> $ rdiff-backup -l  /cygdrive/e/s-bup/
>> Fatal Error: Bad directory /cygdrive/e/s-bup.
>> It doesn't appear to be an rdiff-backup destination dir
>>
>> ---
>> So, it appears to have something to do with cygwin. (At least it does
>> to me.)
>>
>> -Greg

Yeah, it is, and I was pondering doing so.

Are the exit codes the same, and available to cygwin event though
it's a windows binary? (I do see the output is the same unix style
output (with a unix vs dos line return).

What if I have a remote source/destination and I need to do SSH over it? (Isn't
putty quite messy to setup vs cygwin?)

Other than remote destinations, I think the windows binary will work.

As for 1.3.3, how "unstable" is it? Do I dare roll it out in a
production environment?

Thanks for the thoughts. I'll try them and see what happens.

-Greg