swift $file was corrupted during upload even with retries

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
5 messages Options
Reply | Threaded
Open this post in threaded view
|

swift $file was corrupted during upload even with retries

duplicity-talk mailing list
Howdy,

Wanting to raise this issue:

https://bugs.launchpad.net/duplicity/+bug/1734144

There was discussion already it seems earlier: https://lists.nongnu.org/archive/html/duplicity-talk/2011-11/msg00037.html - but no outcome.

It would help me if this error would not be fatal but instead followed the --numretries parameter. I have no control over the backend, but most of the objects do get saved correctly.

Cheers,

_______________________________________________
Duplicity-talk mailing list
[hidden email]
https://lists.nongnu.org/mailman/listinfo/duplicity-talk
Reply | Threaded
Open this post in threaded view
|

Re: swift $file was corrupted during upload even with retries

duplicity-talk mailing list
On 03.01.2018 14:49, Remy van Elst via Duplicity-talk wrote:

> Howdy,
>
> Wanting to raise this issue:
>
> https://bugs.launchpad.net/duplicity/+bug/1734144
>
> There was discussion already it seems earlier: https://lists.nongnu.org/archive/html/duplicity-talk/2011-11/msg00037.html - but no outcome.
>
> It would help me if this error would not be fatal but instead followed the --numretries parameter. I have no control over the backend, but most of the objects do get saved correctly.
>

can you supply a proper stack trace to see where the fatal error originates?

any reason you didn't post on the launchpad bug ticket?

..ede/duply.net

_______________________________________________
Duplicity-talk mailing list
[hidden email]
https://lists.nongnu.org/mailman/listinfo/duplicity-talk
Reply | Threaded
Open this post in threaded view
|

Re: swift $file was corrupted during upload even with retries

duplicity-talk mailing list
No particular reason except for the stacktrace has to be removed of all privacy related data, and the error is quite specific and clear. I therefore didn't suspect it to be required.


On Wed, Jan 3, 2018 at 3:16 PM, edgar.soldin--- via Duplicity-talk <[hidden email]> wrote:
On 03.01.2018 14:49, Remy van Elst via Duplicity-talk wrote:
> Howdy,
>
> Wanting to raise this issue:
>
> https://bugs.launchpad.net/duplicity/+bug/1734144
>
> There was discussion already it seems earlier: https://lists.nongnu.org/archive/html/duplicity-talk/2011-11/msg00037.html - but no outcome.
>
> It would help me if this error would not be fatal but instead followed the --numretries parameter. I have no control over the backend, but most of the objects do get saved correctly.
>

can you supply a proper stack trace to see where the fatal error originates?

any reason you didn't post on the launchpad bug ticket?

..ede/duply.net

_______________________________________________
Duplicity-talk mailing list
[hidden email]
https://lists.nongnu.org/mailman/listinfo/duplicity-talk


_______________________________________________
Duplicity-talk mailing list
[hidden email]
https://lists.nongnu.org/mailman/listinfo/duplicity-talk
Reply | Threaded
Open this post in threaded view
|

Re: swift $file was corrupted during upload even with retries

duplicity-talk mailing list
In reply to this post by duplicity-talk mailing list
On 03.01.2018 15:16, edgar.soldin--- via Duplicity-talk wrote:

> On 03.01.2018 14:49, Remy van Elst via Duplicity-talk wrote:
>> Howdy,
>>
>> Wanting to raise this issue:
>>
>> https://bugs.launchpad.net/duplicity/+bug/1734144
>>
>> There was discussion already it seems earlier: https://lists.nongnu.org/archive/html/duplicity-talk/2011-11/msg00037.html - but no outcome.
>>
>> It would help me if this error would not be fatal but instead followed the --numretries parameter. I have no control over the backend, but most of the objects do get saved correctly.
>>
>
> can you supply a proper stack trace to see where the fatal error originates?
>
> any reason you didn't post on the launchpad bug ticket?
>

ok, seem to have found it.

  http://bazaar.launchpad.net/~duplicity-team/duplicity/0.8-series/view/head:/bin/duplicity#L332

@Ken: how about moving this check into the backend wrapper, where it will be retryable? Mike's reasoning in
  https://lists.nongnu.org/archive/html/duplicity-talk/2011-11/msg00037.html
is flawed anyway, because only if the upload does not fail persistently we can assume that we have a bug in "our" code base.
  also the error message is somewhat misleading and should simply state "File size of %s on the backend does not match the local size." as this is what is tested only.

..ede/duply.net

_______________________________________________
Duplicity-talk mailing list
[hidden email]
https://lists.nongnu.org/mailman/listinfo/duplicity-talk
Reply | Threaded
Open this post in threaded view
|

Re: swift $file was corrupted during upload even with retries

duplicity-talk mailing list
In reply to this post by duplicity-talk mailing list
Remy,

well it generally helps to pinpoint the error's origin. would have saved me 5 minutes to find it ;) maybe next time.

my guess is that this actually is not an upload issue, but the backend returning a wrong file size attribute. would you be willing to patch 'duplicity/backends/swiftbackend.py' on your box with a debug print out to verify this?

..ede/duply.net

On 03.01.2018 15:25, Remy van Elst via Duplicity-talk wrote:

> No particular reason except for the stacktrace has to be removed of all privacy related data, and the error is quite specific and clear. I therefore didn't suspect it to be required.
>
>
>
> https://raymii.org
>
> On Wed, Jan 3, 2018 at 3:16 PM, edgar.soldin--- via Duplicity-talk <[hidden email] <mailto:[hidden email]>> wrote:
>
>     On 03.01.2018 14:49, Remy van Elst via Duplicity-talk wrote:
>     > Howdy,
>     >
>     > Wanting to raise this issue:
>     >
>     > https://bugs.launchpad.net/duplicity/+bug/1734144 <https://bugs.launchpad.net/duplicity/+bug/1734144>
>     >
>     > There was discussion already it seems earlier: https://lists.nongnu.org/archive/html/duplicity-talk/2011-11/msg00037.html <https://lists.nongnu.org/archive/html/duplicity-talk/2011-11/msg00037.html> - but no outcome.
>     >
>     > It would help me if this error would not be fatal but instead followed the --numretries parameter. I have no control over the backend, but most of the objects do get saved correctly.
>     >
>
>     can you supply a proper stack trace to see where the fatal error originates?
>
>     any reason you didn't post on the launchpad bug ticket?
>
>     ..ede/duply.net <http://duply.net>
>
>     _______________________________________________
>     Duplicity-talk mailing list
>     [hidden email] <mailto:[hidden email]>
>     https://lists.nongnu.org/mailman/listinfo/duplicity-talk <https://lists.nongnu.org/mailman/listinfo/duplicity-talk>
>
>
>
>
> _______________________________________________
> Duplicity-talk mailing list
> [hidden email]
> https://lists.nongnu.org/mailman/listinfo/duplicity-talk
>


_______________________________________________
Duplicity-talk mailing list
[hidden email]
https://lists.nongnu.org/mailman/listinfo/duplicity-talk