On Sun, Nov 08, 2020 at 10:10:32PM +0100, John Paul Adrian Glaubitz wrote: > On 11/8/20 9:57 PM, Dominic Hargreaves wrote: > > Right, it looks like the issue described at #931641 has returned. > > This was apparently[1] fixed in qemu-user but I don't know any more > > details about the fix (or indeed the deployment status of the buildds) > > to check how it regressed. > > It must be a slightly different issue and most likely this bug #970460 [1]. > > A heads-up to us porters would have been nice prior the upload, now I'm > very much in a hurry to get perl_5.32 to be built not to fall back in the > transition *sigh*.
Hi, I'm sorry about that. I don't think either of us noticed the failure in the experimental upload in September. It's easy to miss failures of non-release architectures, given how often transient issues arise. If I understand that bug correctly, there's a workaround that you would be able to apply on the buildds - is that correct? I filed #974004 to track this on the perl side. Thanks Dominic
