Wrong oFono property used for data technology indication

Bug #1245951 reported by Tony Espy
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
Canonical System Image
Fix Released
Medium
Unassigned
indicator-network (Ubuntu)
Fix Released
Medium
Unassigned
indicator-network (Ubuntu RTM)
Fix Released
Undecided
Unassigned

Bug Description

Image: Touch #100 ( aka v1.0 ).

The current indicator-network code uses the ofono's NetworkRegistration interface's "Technology" property to determine which mobile data connection icon to display on the status bar.

The code currently compares the value of "Technology" to:

 - "gsm" -- this causes "2G" ( or a tortoise ) to be displayed

 - "umts" -- this causes "3G" to be displayed

 - "edge" -- this causes "E" to be displayed

 - "hspa" -- this causes "H" to be displayed

 - "lte" -- valid, but currently not supported in the ofono/rilmodem driver ( LTE isn't yet 100% supported )

The code instead should be examining ofono's ConnectionManager interface's "Bearer" property. This is not currently being set by the ofono/rilmodem code ( see bug #1245954 for more details ).

The valid values of the "Bearer" property ( with associated icons in parentheses ) are: "none", "gprs" ( 2G ), "edge" ( E ), "umts" ( 3G ), "hsupa"/"hsdpa"/"hspa" ( H ), and "lte" ( L ). oFono doesn't currently support reporting HSPA+, whereas RIL does, so if have the need for differentiating between HSPA and HSPA+, then we'll need to patch ofono so that the value "hspap" is valid for "Bearer".

 -

Tags: rtm14

Related branches

Revision history for this message
Tony Espy (awe) wrote :

See lp:~indicator-applet-developers/indicator-network/trunk.14.04/view/head:/network/network-action-manager.vala and checkout the functions: netreg_property() and ofono_tech_icon_to_name().

Changed in indicator-network:
status: New → Confirmed
Changed in indicator-network (Ubuntu):
status: New → Confirmed
Tony Espy (awe)
description: updated
Tony Espy (awe)
description: updated
Tony Espy (awe)
description: updated
Revision history for this message
Tony Espy (awe) wrote :

Note, bug #1245954 ( ConnectionManger 'Bearer' property not set ) needs to be fixed first.

Charles Kerr (charlesk)
Changed in indicator-network:
assignee: nobody → Ted Gould (ted)
Revision history for this message
Matthew Paul Thomas (mpt) wrote :

The set of symbols defined in <https://wiki.ubuntu.com/Networking#phone-indicator> is not sacrosanct. The goal is to convey, to someone familiar with a carrier, why their data speed on that carrier has suddenly changed. For example, my carrier+phone connect with 3G in some locations, Edge in others, and GPRS in others, and I have learned that some apps are not worth trying to use with anything less than 3G.

So if the list has connection types that are either missing or redundant, please let me know. If there exists a range chart showing the speeds achieved by the various connection types, and/or the combinations of types typically provided by a carrier, that would also help me understand how useful the distinctions are.

Revision history for this message
Tony Espy (awe) wrote :

I think the only thing we're missing currently, is the ability to distinguish between generic HSPA and HSPA+, as mentioned in the bug description.

The code to fix this bug #1245954 is currently in our upstream ofono git repository, and will be landing in the archive shortly.

That said, I did *not* modify the code to address the HSPA/HSPA+ issue. If this is thought to be important, it can be fixed, but as mentioned, it will require changes to the core ofono code ( ie. it can't be accomplished via changes to our rilmodem plugin alone ).

Regarding a chart that shows various connection speeds, unfortunately I don't have anything to point you at. The specifics aren't all that meaningful to end-users other than the fact that:

GSM < EDGE < UMTS < HSPA < HSPA+ < LTE

description: updated
Changed in indicator-network:
assignee: Ted Gould (ted) → Antti Kaijanmäki (kaijanmaki)
status: Confirmed → Triaged
Changed in indicator-network (Ubuntu):
status: Confirmed → Triaged
Changed in indicator-network:
importance: Undecided → High
Changed in indicator-network (Ubuntu):
importance: Undecided → High
assignee: nobody → Antti Kaijanmäki (kaijanmaki)
Revision history for this message
Tony Espy (awe) wrote :

I just reviewed this and discovered that although the code has been re-written from Vala to C++, the bug still exists. Added an "rtm14" tag as this really needs to get fixed.

tags: added: rtm14
Thomas Strehl (strehl-t)
tags: added: touch-2014-09-04
Thomas Strehl (strehl-t)
Changed in indicator-network:
assignee: Antti Kaijanmäki (kaijanmaki) → Jussi Pakkanen (jpakkane)
Changed in indicator-network (Ubuntu):
assignee: Antti Kaijanmäki (kaijanmaki) → Jussi Pakkanen (jpakkane)
tags: added: touch-2014-09-11
removed: touch-2014-09-04
Revision history for this message
Antti Kaijanmäki (kaijanmaki) wrote :

for now, network-registration has been giving reliable results, so I don't see an immediate reason to fix this ASAP.

Changed in indicator-network (Ubuntu):
importance: High → Medium
Changed in indicator-network:
importance: High → Medium
tags: added: touch-2014-09-25
removed: touch-2014-09-11
tags: added: touch-2014-10-02
removed: touch-2014-09-25
tags: added: touch-2014-10-09
removed: touch-2014-10-02
Revision history for this message
Antti Kaijanmäki (kaijanmaki) wrote :

OK, now an actual real world case from asac where the technology and bearer don't match:
https://pastebin.canonical.com/118313/

Would still keep this as Medium, as the worst case scenario is that the user actually sees "3G" instead of "2G".

Revision history for this message
John McAleely (john.mcaleely) wrote :

the new duplicate bug #1386821 has a reliable, recent repro that might need the specific setup noted in the description

Revision history for this message
John McAleely (john.mcaleely) wrote :

the touch tag on this bug looks wrong.

tags: removed: touch-2014-10-09
Revision history for this message
John McAleely (john.mcaleely) wrote :

Just to note there is barajas bug now duped against this, so adding to canonical system image.

Revision history for this message
John McAleely (john.mcaleely) wrote :

I'm not convinced that Medium is correct for this - it's a credibility issue about data connections that cost money.

Revision history for this message
Antti Kaijanmäki (kaijanmaki) wrote :

I went through the barajas bugs, but can't find a clear dup. John, could you attach a bug number?

The only situation when the Technology and the Bearer properties differ is when the mobile broadband connection speed is limited by the operator to be _lower_ than the cellular networks normal technology; voice might go through 3G but mobile data is limited to 2G. So the user gets a lower internet connection speed than actually reported by the indicator.

Then on the other hand, even though if the user has an actual 3G or even 4G mobile broadband connection the actual data transfer speeds sometimes drop to the equivalent levels of 2G connections depending on the loads on the carrier network, so the whole icon is more or less a hint of the ideal connection speeds you _might_ get. Sadly for the customers there are no definitive requirements for the carriers on minimum throughput.

Anyway, that is a real bug, but there should be no "surprises" for the user in terms of billing. Hence Medium.

Revision history for this message
John McAleely (john.mcaleely) wrote :

Users pay for particular grades of service. This bug is that the user sees a 3G service they are paying for when in fact they are getting 2G. That might be good for operator PR, but it is bad for ubuntu phone's credibility.

If you feel the two dupes listed are not dupes, please comment there, and we can separately discuss them.

Revision history for this message
Antti Kaijanmäki (kaijanmaki) wrote :

Checked the duplicates. Both are actually duplicates of a different bug #1366014.
Updated them accordingly.

Pete Woods (pete-woods)
no longer affects: indicator-network
Changed in indicator-network (Ubuntu):
assignee: Jussi Pakkanen (jpakkane) → nobody
Revision history for this message
Launchpad Janitor (janitor) wrote :

This bug was fixed in the package indicator-network - 0.5.1+15.04.20150401~rtm-0ubuntu1

---------------
indicator-network (0.5.1+15.04.20150401~rtm-0ubuntu1) 14.09; urgency=medium

  [ Pete Woods ]
  * Port all dbus interactions to QDBus (LP: #1390136, #1427064,
    #1245951)
 -- CI Train Bot <email address hidden> Wed, 01 Apr 2015 15:31:25 +0000

Changed in indicator-network (Ubuntu RTM):
status: New → Fix Released
Changed in canonical-devices-system-image:
status: New → Fix Released
milestone: none → ww13-ota
importance: Undecided → Medium
Revision history for this message
Launchpad Janitor (janitor) wrote :

This bug was fixed in the package indicator-network - 0.5.1+15.04.20150401.2-0ubuntu1

---------------
indicator-network (0.5.1+15.04.20150401.2-0ubuntu1) vivid; urgency=medium

  [ Marcus Tomlinson ]
  * Add integration tests for indicator
  * Port all D-Bus interactions over to QDBus (LP: #1390136, #1427064,
    #1245951)

  [ Pete Woods ]
  * Add integration tests for indicator
  * Port all D-Bus interactions over to QDBus (LP: #1390136, #1427064,
    #1245951)
 -- CI Train Bot <email address hidden> Wed, 01 Apr 2015 15:31:55 +0000

Changed in indicator-network (Ubuntu):
status: Triaged → 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.