[Mageia-dev] Rebuild failed on noarch for @211804:firefox-l10n-10.0.2-2.mga2.src.rpm

Mon Feb 27 12:00:09 CET 2012

Am 27.02.2012 03:30, schrieb Anssi Hannula:
+> 27.02.2012 02:22, Pascal Terjan kirjoitti:
+>> On Mon, Feb 27, 2012 at 00:09, Pascal Terjan <pterjan at gmail.com> wrote:
+>>> On Sun, Feb 26, 2012 at 23:12, Ulri the scheduler bot
+>>> <mageia-sysadm at mageia.org> wrote:
+>>>> Build of the following packages failed:
+>>>>
+>>>> - @211804:firefox-l10n-10.0.2-2.mga2.src.rpm
+>>>>
+>>>> Failure details available in http://pkgsubmit.mageia.org/uploads/failure/cauldron/core/release/20120226231059.pterjan.valstar.5570/log
+>>>> Reason:
+>>>> @211804:firefox-l10n-10.0.2-2.mga2.src.rpm: build_failure
+>>>>
+>>>> Log files generated:
+>>>> http://pkgsubmit.mageia.org/uploads/failure/cauldron/core/release/20120226231059.pterjan.valstar.5570/log/chroot/initialize-1.0.20120226231100.log
+>>>> http://pkgsubmit.mageia.org/uploads/failure/cauldron/core/release/20120226231059.pterjan.valstar.5570/log/firefox-l10n-10.0.2-2.mga2/rpm_qa.0.20120226231100.log
+>>>> http://pkgsubmit.mageia.org/uploads/failure/cauldron/core/release/20120226231059.pterjan.valstar.5570/log/firefox-l10n-10.0.2-2.mga2/build.0.20120226231100.log
+>>>> http://pkgsubmit.mageia.org/uploads/failure/cauldron/core/release/20120226231059.pterjan.valstar.5570/log/firefox-l10n-10.0.2-2.mga2/install_deps-1.0.20120226231100.log
+>>> I don't understand what's wrong here (it seems to forget about -zh-TW
+>>> and -zu packages, but it works fine if I manually build it in a iurt
+>>> chroot)
+>> Ah actually it worked fine when I build it from a src.rpm I manually
+>> generated after simplifying a bit the template (which would not change
+>> anything) but not if I try to rebuild
+>> http://pkgsubmit.mageia.org/uploads/failure/cauldron/core/release/20120226231059.pterjan.valstar.5570/@211804:firefox-l10n-10.0.2-2.mga2.src.rpm
+>>
+>> Maybe the expanded spec got too long and simplifying the template "fixed" it?
+>> Anyway, it is now built.
+>>
+> AFAIK there is/was a buffer size issue in rpm-build, causing
+> %{expand:foo} to fail if 'foo' is too long. Never got around to look at
+> it closer, though...
+>
+Yep, it's this one https://bugzilla.redhat.com/show_bug.cgi?id=431009
+Not got around to re-report it yet, and it still happens, as you can see.
+