Bug#826858: ostree: FTBFS due to test suite errors

2016-06-21 Thread Simon McVittie 
Control: clone 826858 -2
Control: retitle 826858 ostree: FTBFS on i386, hppa: Opening objects/ 
directory: openat: No such file or directory
Control: tags 826858 + unreproducible
Control: retitle -2 ostree: FTBFS on sparc64: gjs: Segmentation fault
Control: notforwarded -2
Control: severity -2 normal

On Thu, 09 Jun 2016 at 10:35:48 -0400, Aaron M. Ucko wrote:
> - On i386:
...
> - On hppa (admittedly not a release architecture):
...

These share a symptom ("Opening objects/ directory: openat: No such file or
directory"), which wasn't seen again in 2016.5-4, and I couldn't reproduce it
on i386 locally. I assume this was either some transient infrastructure issue,
or an ostree bug that rarely occurs (perhaps a race condition).

> - On sparc64 (not a release architecture either):

This appears to be a separate issue, and was also visible in 2016.5-4.
Cloning a separate bug for this one.

S

Bug#826858: ostree: FTBFS due to test suite errors

2016-06-09 Thread Simon McVittie 
Control: forwarded 826858 https://github.com/ostreedev/ostree/issues/328

On Thu, 09 Jun 2016 at 10:35:48 -0400, Aaron M. Ucko wrote:
> For some reason, ostree hit test suite errors on some architectures.

I'm aware, and already looking into the i386 and hppa errors
(they have the same symptoms). The upstream maintainer has fixed
some issues where the test wasn't immediately aborting when something
went wrong, which made it harder to see what was going on.

Unfortunately, I can't reproduce the i386 failure in a local chroot.

> FWIW, signal 5 is SIGTRAP, at least on x86.

GLib often raises SIGTRAP on fatal errors, including in its test framework.
You have to read a few lines back in the verbose logs to see why.

> - On sparc64 (not a release architecture either):
> 
>   ERROR: tests/test-pull-corruption.sh - too few tests run (expected 2, got 0)
>   ERROR: tests/test-pull-corruption.sh - exited with status 1

This one is a segfault in gjs, a JavaScript environment using Mozilla
code for the interpreter and GObject-Introspection for runtime libraries.
OSTree uses it as a scripting language in some tests, to confirm that
its language bindings work.

It seems likely that this one is a sparc64 bug in either libmozjs,
gjs, GObject-Introspection, or possibly libffi (used by GObject).

S

Bug#826858: ostree: FTBFS due to test suite errors

2016-06-09 Thread Aaron M. Ucko 
Source: ostree
Version: 2016.5-3
Severity: important
Justification: fails to build from source

For some reason, ostree hit test suite errors on some architectures.

- On i386:

  ERROR: tests/test-sysroot-c - too few tests run (expected 1, got 0)
  ERROR: tests/test-sysroot-c - exited with status 133 (terminated by signal 5?)

- On hppa (admittedly not a release architecture):

  ERROR: tests/test-basic-c - missing test plan
  ERROR: tests/test-basic-c - exited with status 133 (terminated by signal 5?)

- On sparc64 (not a release architecture either):

  ERROR: tests/test-pull-corruption.sh - too few tests run (expected 2, got 0)
  ERROR: tests/test-pull-corruption.sh - exited with status 1

I don't have further details, but perhaps you can reproduce the
problem on a porterbox (or a local chroot or VM, for i386).  FWIW,
signal 5 is SIGTRAP, at least on x86.

Could you please take a look?

Thanks!