Project

General

Profile

Issue #486

Stream tests failing

Added by Jonathan Davies over 11 years ago. Updated over 10 years ago.

Status:
Closed
Priority:
Normal
Assignee:
Martin Willi
Category:
libstrongswan
Affected version:
dr|rc|master
Resolution:
No change required

Description

As seen in https://launchpadlibrarian.net/162349161/buildlog_ubuntu-precise-i386.strongswan_5.1.2~dr2-0~10374%2B18~ubuntu12.04.1_FAILEDTOBUILD.txt.gz

The stream tests are failing:

...
  Running suite 'stream':
    Running case 'sync': ++-
      Failure in 'test_sync': service != NULL (suites/test_stream.c:62, i = 2)
    Running case 'async': ++-
      Failure in 'test_async': service != NULL (suites/test_stream.c:128, i = 2)
    Running case 'all': ++-
      Failure in 'test_all': service != NULL (suites/test_stream.c:180, i = 2)
    Running case 'concurrency': ++-
      Failure in 'test_concurrency': service != NULL (suites/test_stream.c:221, i = 2)
  Passed 0/4 'stream' test cases
...

History

#1 Updated by Martin Willi over 11 years ago

  • Category set to libstrongswan
  • Status changed from New to Feedback
  • Assignee set to Martin Willi

Hi Jonathan,

The first test in each case uses a Unix socket, the second IPv4 and the third IPv6. Given that IPv4 works, but IPv6 doesn't, it is likely that is related to the IPv6 configuration of your host.

Does your build host support IPv6, and does it have a loopback address? Does ping6 ::1 work?

#2 Updated by Jonathan Davies over 11 years ago

Martin Willi wrote:

The first test in each case uses a Unix socket, the second IPv4 and the third IPv6. Given that IPv4 works, but IPv6 doesn't, it is likely that is related to the IPv6 configuration of your host.

Does your build host support IPv6, and does it have a loopback address? Does ping6 ::1 work?

I noticed after failing this that this only failed on the 12.04 build on i386. Everything else appears to have gone through fine. So this may have been a temporaily glitch.

The buildd should be running on standard Ubuntu stuff, so there should be a ::1 address by default.

I'll keep an eye on the builds for future problems like this.

#3 Updated by Martin Willi over 10 years ago

  • Status changed from Feedback to Closed
  • Resolution set to No change required

So this may have been a temporaily glitch.

Closing the issue for now, feel free to reopen should it re-appear.