Control: clone -1 -2 Control: reassign -2 libio-socket-ip-perl 0.39-1 Control: tag -2 bullseye sid Control: retitle -2 libio-socket-ip-perl: fails to build on IPv6-only buildds Control: found -1 5.30.2-1 Control: block -1 with -2
On Tue, Jun 02, 2020 at 11:04:28AM +0200, Julien Cristau wrote: > Control: tag -1 patch > > On Tue, Jun 02, 2020 at 08:09:43AM +0200, Salvatore Bonaccorso wrote: > > On Tue, Jun 02, 2020 at 06:40:41AM +0200, Salvatore Bonaccorso wrote: > > > Source: perl > > > Version: 5.30.3-1 > > > Severity: serious > > > Justification: FTBFS > > > > > > Hi Dom an Niko > > > > > > Guess you have seen it but filling a bug for tracking. arch:all build > > > failed on buildd: > > > > > > https://buildd.debian.org/status/fetch.php?pkg=perl&arch=all&ver=5.30.3-1&stamp=1591057198&raw=0 > > > > FTR, the issue is that the build got picked up by a IPv6-only enabled > > buildd. > > FWIW, it seems IO::Socket::IP passes AI_ADDRCONFIG to getaddrinfo, which > then doesn't return ipv4 addresses because the host doesn't have > non-local ones, even though the address we're trying to resolve is > "127.0.0.1". > > Passing either GetAddrInfoFlags => AI_NUMERICHOST or GetAddrInfoFlags => > 0 to both IO::Socket::IP->new calls in the test file lets it pass, by > turning off the smarts in getaddrinfo. Thanks. The immediate concern has been solved by rescheduling the failed builds on different buildd hosts. As this wasn't broken by a change in perl, I'm marking it found in the current version in testing too so as not to block testing migration of 5.30.3-1. It still needs to be fixed, of course. I'm also cloning a separate bug for libio-socket-ip-perl, which has the same tests. I've verified that it also fails to build in the same way. FWIW, an easy way to test this is # unshare -n # ip li set lo up # ip dev add dummy0 type dummy # ip li set dummy0 up It's probably best to fix libio-socket-ip-perl first as it's self contained, and then copy the fix over to src:perl. -- Niko Tyni nt...@debian.org
