Dear maintainer(s),

Your package has an autopkgtest, great. However, it fails since early
2024 on amd64 and some time later on other architectures. I highly
suspect that the test fails when the file system lives on tmpfs, as is
the case on all architectures nowadays.

The release team has announced [1] that failing autopkgtest on amd64 and
arm64 are considered RC in testing.

More information about this bug and the reason for filing it can be
found on
https://wiki.debian.org/ContinuousIntegration/RegressionEmailInformation

With my Release Team member hat on I have tagged this as trixie-ignore
as it's so late in the freeze and it's not worth removing the package
from trixie because of this at this moment. Having said that, if a fix
is possible without fully removing the autopkgtest and without making
the test superficial, it's still welcome, but it would need to happen soon.

Paul

[1] https://lists.debian.org/debian-devel-announce/2019/07/msg00002.html

https://ci.debian.net/data/autopkgtest/testing/amd64/t/toybox/61462994/log.gz

75s PASS: cut -d newline
  75s SKIP: demo_number disabled
  75s PASS: dirname /-only
  75s PASS: dirname trailing /
  75s PASS: dirname combined
  75s PASS: dirname /a/
  75s PASS: dirname multiple
  75s FAIL: du (no options)
  75s echo -ne '' | du -k du_test
  75s --- expected	2025-06-18 19:13:21.103913606 +0000
  75s +++ actual	2025-06-18 19:13:21.107913692 +0000
  75s @@ -1,2 +1,2 @@
  75s -4	du_test/test
  75s -8	du_test
  75s +0	du_test/test
  75s +0	du_test
  75s make: *** [Makefile:77: tests] Error 1

Hi,

I've already reported this problem upstream in oct 2021.

 | https://github.com/landley/toybox/issues/301

What do you think about aplying the suggested patch on the test?
It makes the test superficial on tmpfs-like filesystem \o/

 | if [ "$(stat --format %b . 2>/dev/null)" != "0" ]; then
 |   echo "$SHOWSKIP: du (tmpfs-like filesystem)"
 |   return 2>/dev/null
 |   exit
 | fi


I think it's the best we can do for trixie.

Thanks for your time Paul!

Hi,

I've already reported this problem upstream in oct 2021.

 | https://github.com/landley/toybox/issues/301

What do you think about aplying the suggested patch on the test?
It makes the test superficial on tmpfs-like filesystem \o/

 | if [ "$(stat --format %b . 2>/dev/null)" != "0" ]; then
 |   echo "$SHOWSKIP: du (tmpfs-like filesystem)"
 |   return 2>/dev/null
 |   exit
 | fi


I think it's the best we can do for trixie.

Thanks for your time Paul!

Hi Antoni,


Ack.


Do I understand correctly that you'll only skip this particular test?
And would the test still pass in the regular sense if you run the test
on a non-tmpfs filesystem? Than it sound good to me with the
understanding that this is a test problem and not a fundamental problem
with the package. As said in the original report on ci.d.n all hosts run
the tests on a tmpfs filesystem.

Paul