l-m-c is failing in the lab

[Michael Hudson-Doyle]

I haven't really looked into this, but it seems that all tests that run linaro-media-create in the lab (which isn't all the many any more, but includes the kernel ci tests) fail to do so correctly.

[Michael Hudson-Doyle (mwhudson)]

Actually I think this is a problem with the hwpacks ci.linaro.org is generating...

[Michael Hudson-Doyle (mwhudson)]

A failing build is here: http://validation.linaro.org/lava-server/scheduler/job/20698/log_file#entry2

[Deepti B. Kalakeri (deeptik)]

The test hwpacks that come out of ci.linaro.org fail to boot as the kernel deb package is being built for armel, while including it in the precise armhf hwpack. Hence the lmc is not able to boot the kernel on lava.

Building the armhf kernel deb package solved the problem.

Here is the link http://validation.linaro.org/lava-server/dashboard/streams/private/team/linaro/ci-linux-linaro-tracking/bundles/9d692d8ba69754e9c8787f0d367021946cefec21/8948bce8-ab18-11e1-9729-68b599be548c/ to the successful boot of the kernel being built on ci* after applying the above fix.

Thanks!!!
Deepti.

[Deepti B. Kalakeri (deeptik)]

The hwpack boot problem for panda, vexpress, snowball boards is fixed now.
Beagle board is failing to boot completely with a kernel panic for which I have already sent an email to linaro-kernel.
Origen hwpack boot failure is tracked in the bug 928203. The origen builds require additional configs to be set apart from the ones which are available in the default configs. The investigation on this is in process.

I feel this bug can be closed now as the majority of the boot failures have been fixed and the ones which are still failing are being tracked already in different bugs.

Thanks!!!
Deepti.

[Michael Hudson-Doyle (mwhudson)]

Awesome, thanks for digging into this.