This is the mail archive of the glibc-bugs@sourceware.org mailing list for the glibc project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

[Bug network/20826] New: Tests fail on hosts without network access


https://sourceware.org/bugzilla/show_bug.cgi?id=20826

            Bug ID: 20826
           Summary: Tests fail on hosts without network access
           Product: glibc
           Version: 2.25
            Status: UNCONFIRMED
          Severity: normal
          Priority: P2
         Component: network
          Assignee: unassigned at sourceware dot org
          Reporter: infinity0 at pwned dot gg
  Target Milestone: ---

Created attachment 9637
  --> https://sourceware.org/bugzilla/attachment.cgi?id=9637&action=edit
Don't condition failure of double-dot-FQDN on single-dot-FQDN

Hi, our build machines on tests.reproducible-builds.org completely block
network access during the build. As a result, posix/tst-getaddrinfo.c fails the
build:

https://tests.reproducible-builds.org/debian/rbuild/unstable/amd64/glibc_2.24-5.rbuild.log.gz

The attached patch fixes the problem by having the test unconditionally check
that names with two trailing dots fail to resolve.

By contrast, the existing code conditionally does this only if the
one-trailing-dot version successfully resolves, and requires there be at least
one successful resolution. I don't understand the point of this, and it's not
clear from #16469 (the bug the test was introduced for) why this should be the
case.

If I understand correctly, normally systems "resolve" localhost via a local
alias in nsswitch/getent, bypassing getaddrinfo completely. So it should not be
required to successfully resolve "localhost." anyways.

-- 
You are receiving this mail because:
You are on the CC list for the bug.

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]