Re: [Duplicity-talk] S3 ECONNRESET during restore results in SHA1 hash mismatch

[edgar . soldin]

On 04.05.2016 22:37, Aphyr wrote:
> Hi Edgar! Thanks for the response. :)
> 
> On 05/04/2016 03:54 AM, address@hidden wrote:
>> hey Aphyr, please state your duply, duplicity, boto, python versions
> 
> duply version 1.9.1
> 
> Using installed duplicity version 0.6.24, python 2.7.9, gpg 2.0.26 (Home: 
> ~/.gnupg), awk 'GNU Awk 4.1.1, API: 1.1 (GNU MPFR 3.1.2-p3, GNU MP 6.0.0)', 
> bash '4.3.30(1)-release (x86_64-pc-linux-gnu)'.
> 
> Looks like boto is 2.34.0--that's the only egg I have in 
> /usr/lib/python2.7/dist-packages/.

there seems to be a boto 2.40 and even a boto3.. often newer version resolve 
quirks like the one you are experiencing

> 
>> --num-retries should work over all backends

didn't check the 0.6 source, but 0.7 definitely uses retry in boto backend

>> run duply w/ '--preview' and check that --num-retries is propagated to the 
>> command line.
> 
> duplicity --name duply_xxx --encrypt-key xxx --sign-key xxx --verbosity '4' 
> --num-retries=100 -t now 's3://s3-us-west-2.amazonaws.com/xxx/b' 'test/
> 
> I've tried both --num-retries 100 and --num-retries=100. Neither seems to 
> retry ECONNRESET.

hmm, maybe an issue w/ the "old" duplicity 0.6 .. try the latest 0.7.07.1 ;)

> 
>> probably should, but nobody contributed that so far ;).. ede/duply.net
> 
> Heh, I'll have to learn python and see what I can do. Just figured someone 
> else might have hit this before. :)

have a look, it's not as difficult as it seems.

good luck.. ede/duply.net