Re: [Duplicity-talk] Resume S3 restore?

[edgar . soldin]

On 07.12.2015 02:04, Mark Lewis wrote:
> I am trying to restore from an old backup from 2011 that I've had in S3, and 
> the restore keeps getting interrupted with errors like this:
> 
> Download 
> s3+http://deja-dup-auto-akiaifjuquylnba7gurq/duplicity-full.20110623T045205Z.vol2255.difftar.gz
>  failed (attempt #4, reason: SSLError: ('The read operation timed out',))
> 
> This happens at random times in the process-- sometimes it has only restored 
> a few hundred megabytes, once it almost finished and got to about 23 out of 
> 25 gigabytes.
> 
> The problem is each time it happens, it will fail repeatedly until duplicity 
> gives up on the restore.  But if I restart the process, that volume restores 
> fine, so something weird is happening that causes a single run to fail each 
> time it tries to fetch the volume, but a subsequent run can fetch it fine.
> 
> That by itself is a weird problem, but I could work around it if there were 
> some way to resume a failed restore.  I haven't seen anything in the man page 
> about that, is there any way to tell it to resume a restore?  Or does anybody 
> have any ideas about the S3 issue?
> 

nope, there is no resume for restores currently. sorry.

1. did you try to upgrade your boto already?
 https://github.com/boto/boto/issues/2409

2. you can raise the number retries via --num-retries, check the manpage

3. if all fails or you are in a hurry, download everything via some s3 client 
to a local folder and restore form there


good luck ..ede/duply.net