[MIR] indicator-network

Bug #1612619 reported by Pete Woods
8
This bug affects 1 person
Affects Status Importance Assigned to Milestone
indicator-network (Ubuntu)
Fix Released
Undecided
Unassigned

Bug Description

[Availability]
 * Available in universe

[Rationale]
 * This package is required by unity8

[Security]
 * No known security issues at this time.

[Quality assurance]
 * This package has both unit tests and integration tests.

[Dependencies]
 Most dependencies are already in main with the exception of the following:
 * libqofono (needs bug)
 * network-manager-openvpn (bug #1574576)
 * urfkill (needs bug)

[Standards compliance]
 * This package uses cmake and is properly translated.

[Maintenance]
 * This package is maintained by Canonical and actively in use on the phone images.

Related branches

Revision history for this message
Michael Terry (mterry) wrote :

- Can you please subscribe ~unity-api-bugs to the ubuntu package in LP?
- The tests are run with || true. It looks like the intent is to only do that on s390x, but I'm seeing the "|| true" bit on amd64 as well. I think the ifneq maybe should be an ifeq? Didn't dig into that.

- Lots of bugs, but that's not a surprise. I know the team is working on them.

Changed in indicator-network (Ubuntu):
status: New → Incomplete
Revision history for this message
Pete Woods (pete-woods) wrote :

Good spot with the || true for the tests. I've put a MR together now that should fix it.

Revision history for this message
Pete Woods (pete-woods) wrote :

I don't think unity-api-bugs is the correct team for the subscription (https://launchpad.net/~unity-api-bugs/+members#active). It looks like only Steve Langasek and Brian Murray and members. We already have unity-api-team subscribed, and we're the people who actually look after the package.

Revision history for this message
Pete Woods (pete-woods) wrote :

Okay, the package from that MR is now in -proposed.

Revision history for this message
Michael Terry (mterry) wrote :

So the changelog entry says "re-enable tests on all architectures", but the diff actually skips the tests on "arm64 s390x powerpc". Why are those tests broken? The MP says "slow build servers" -- can we increase a timeout or something?

Revision history for this message
Launchpad Janitor (janitor) wrote :

This bug was fixed in the package indicator-network - 0.8.0+16.10.20160817.2-0ubuntu1

---------------
indicator-network (0.8.0+16.10.20160817.2-0ubuntu1) yakkety; urgency=medium

  [ Antti Kaijanmäki ]
  * Use a timer to set the sim for mobile data if only one sim exists.
    (LP: #1598010)

  [ Pete Woods ]
  * Re-enable tests on all architectures (LP: #1612619)

 -- Pete Woods <email address hidden> Wed, 17 Aug 2016 11:50:15 +0000

Changed in indicator-network (Ubuntu):
status: Incomplete → Fix Released
Revision history for this message
Pete Woods (pete-woods) wrote :

Whoops. Should have updated that message. The current state is that all the tests work on xenial and yakkety. But arm64 is broken on vivid, we've removed the build for s390x (due to dependencies being removed from distro), and the powerpc build servers just kept being unbelievably slow when trying to build the silo (timing out after several hours).

Michael Terry (mterry)
Changed in indicator-network (Ubuntu):
status: Fix Released → Incomplete
Revision history for this message
Michael Terry (mterry) wrote :

You can do conditional debian/rules code based on distro version (i.e. run tests on yakkety/arm64 but not vivid/arm64).

I'll approve this for now, but filed bug 1616475 to track re-enabling arm64.

Changed in indicator-network (Ubuntu):
status: Incomplete → Fix Committed
Revision history for this message
Pete Woods (pete-woods) wrote :

Any tips on where to look for for this? I could see any clearly labelled debian build variables that would tell me which distro we were on.

Revision history for this message
Michael Terry (mterry) wrote :

Good question, I was vague before because I couldn't remember the variables for it.

Looks like one thing you can do is:

include /usr/share/dpkg/default.mk

ifeq (Ubuntu,$(DEB_VENDOR))
  ifeq (vivid,$(DEB_DISTRIBUTION))
    XXX
  endif
endif

(stolen from php7.0)

Revision history for this message
Steve Langasek (vorlon) wrote :

Override component to main
indicator-network 0.8.0+16.10.20160930.5-0ubuntu1 in yakkety amd64: universe/gnome/optional/100% -> main
indicator-network 0.8.0+16.10.20160930.5-0ubuntu1 in yakkety arm64: universe/gnome/optional/100% -> main
indicator-network 0.8.0+16.10.20160930.5-0ubuntu1 in yakkety armhf: universe/gnome/optional/100% -> main
indicator-network 0.8.0+16.10.20160930.5-0ubuntu1 in yakkety i386: universe/gnome/optional/100% -> main
indicator-network 0.8.0+16.10.20160930.5-0ubuntu1 in yakkety powerpc: universe/gnome/optional/100% -> main
indicator-network 0.8.0+16.10.20160930.5-0ubuntu1 in yakkety ppc64el: universe/gnome/optional/100% -> main
Override [y|N]? y
6 publications overridden.

Changed in indicator-network (Ubuntu):
status: Fix Committed → Fix Released
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.