On 01/07/2017 09:19 AM, Adam D. Barratt wrote: > On Fri, 2017-01-06 at 23:23 +0000, Adam D. Barratt wrote: >> Control: tags -1 + pending >> >> On Thu, 2017-01-05 at 21:27 +0100, Sebastiaan Couwenberg wrote: >>> On 01/05/2017 09:04 PM, Adam D. Barratt wrote: >>>> On Tue, 2016-12-06 at 22:00 +0100, Sebastiaan Couwenberg wrote: >>>>> Sorry for the outdated debdiff, for p-u the distribution has been >>>>> changed to stable. >>>> >>>> Please go ahead. >>> >>> Thanks! >> >> Flagged for acceptance. > > The upload subsequently FTBFS on all architectures on the buildds. > > The relevant portion of the logs appears to be: > > [...] > > I'm not entirely sure how you managed to build the amd64 packages > without this issue occurring, but please look into this ASAP. As things > stand, we won't be including mapserver in 8.7.
I simply built the package in an up-to-date jessie cowbuilder chroot. The buildds used a newer PHP from p-u, my build was with 5.6.24+dfsg-0+deb8u1 in jessie. I can reproduce the build failure with an i386 chroot on barriere, which also uses php 5.6.29+dfsg-0+deb8u1 from jessie-security. Disabling the PHP support in mapserver may be an option, it has been disabled in testing/unstable too because it lacks support for PHP 7. It will force users to migrate to other mapscript implementations sooner than their upgrade to stretch. Another option it cherry-pick this change from upstream for this issue: https://github.com/mapserver/mapserver/commit/d5cc5840683bded229dbc2cc7d534db6d1e1f386#diff-f8894221d4e5ead75cf65d8aa506ae91 If that change to mapscript/php/error.c is acceptable for jessie, I'll include it in the package and prepare mapserver/6.4.1-5+deb8u2. Kind Regards, Bas -- GPG Key ID: 4096R/6750F10AE88D4AF1 Fingerprint: 8182 DE41 7056 408D 6146 50D1 6750 F10A E88D 4AF1
