Checkbox is skipping the monitor/vga tests in some Latitude systems with VGA port

[Ara Pulido]

Some Dell Latitude systems with VGA ports are returning some how unsupported for the VGA resource, causing the monitor VGA tests to be skipped.

One example of the systems that are skipping the test:

   https://certification.canonical.com/hardware/201307-13905/submission/92540/test-results/skipped/

[Brendan Donegan (brendan-donegan)]

It seems to be the case that xrandr thinks it *doesn't* have a VGA port. It's on the submissions attachments list: https://certification.canonical.com/hardware/201307-13905/submission/92540/attachments/636749/

We see:

eDP1 connected (laptops display)

and

HDMI1 disconnected
DP1 disconnected

for the external displays. So, unless there's been some miscommunication somewhere and this system doesn't in fact have a VGA port, then there must be a kernel/X bug here.

[Daniel Manrique (roadmr)]

The Latitude 3440 has a VGA port, it's listed in the data sheet and if you roam through the gallery of pictures you can see it on the left side:

http://www.dell.com/ed/business/p/latitude-3440-laptop/pd

guess it's time to ask the good X folk about this system...

[Brendan Donegan (brendan-donegan)]

Yes, I checked with Yung as well who has this system in lab, and it does have a VGA *port*. However as shown by the output of xrandr, this is not shown as such by X. It appears to be the case that a VGA connector has been hacked on to a DP line. This is proven by the fact that - as confirmed by Yung, the DP1 port shows as 'connected' when the display is plugged in to the 'VGA' port.

[Daniel Manrique (roadmr)]

It's probably worth asking the X folk about this system. It may be a bug they need to fix (in which case we win), or it may just be a hackish vga connector with no way to be actually detected (in which case we need to find a workaround for this).

[Brendan Donegan (brendan-donegan)]

I asked them and they concur that it's a VGA port hacked on to a Displayport output. I'm not sure there really is a workaround for this. Any suggestions what to do next?

[Daniel Manrique (roadmr)]

Offering only relevant tests based on resources is a convenience, if detection of a particular port (and one as essential as VGA, at that) is unreliable and is impacting certification, I think the best solution is to remove the requires on VGA port. The user will always be prompted to test VGA and will have to manually skip the test if the device is not available.

This is slightly more cumbersome but to be fair, most systems do include a VGA port so defaulting to assuming we have one will not mean a large amount of extra work.

Optionally, we can require either VGA or DP to display the VGA test. This will be slightly confusing on systems with DP but no VGA, but again, hopefully users will read the instructions and skip the test where appropriate.

I'd do this only for VGA as we haven't seen similar detection reliability issues with other ports.

[Zygmunt Krynicki (zyga)]

I'm marking this as WONT FIX but please reopen if this is actively hampering anyone's work. My goal is to limit the number of open bugs to get a better idea as to what is really important.

Remember that you can always escalate bugs by contacting us in #checkbox on freenode (or #cert-infra in the internal IRC) or by responding in bugs directly.