You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
We have been experiencing deterministic failure on multipart upload against Riak CS. We are not sure how to fix this as don't have solid golang experience, and just reporting it here. See [1] for complete stack trace, and [2] for what we are doing in detail. Also it seems that the commit 7138fa9 [3] introduced this issue, as we don't see this issue against v0.4.3 but later than [3].
We found a workaround increasing partsize by one byte, to -s 5242881 to avoid this issue, hope it helps.
This is resolved on master now and there is now a test for this scenario. The buffer size was not being reset in the correct place. Could you try testing and update if the issue persists in your tests?
We have been experiencing deterministic failure on multipart upload against Riak CS. We are not sure how to fix this as don't have solid golang experience, and just reporting it here. See [1] for complete stack trace, and [2] for what we are doing in detail. Also it seems that the commit 7138fa9 [3] introduced this issue, as we don't see this issue against v0.4.3 but later than [3].
We found a workaround increasing partsize by one byte, to
-s 5242881
to avoid this issue, hope it helps.The text was updated successfully, but these errors were encountered: