LAVA Validation Lab

Merge lp://qastaging/~liuyq0307/lava-lab/fast-model-network into lp://qastaging/lava-lab

fast-model-network
Merge into salt-states

Proposed by Yongqin Liu on 2013-06-25

Status:	Merged
Merged at revision:	192
Proposed branch:	lp://qastaging/~liuyq0307/lava-lab/fast-model-network
Merge into:	lp://qastaging/lava-lab
Diff against target:	312 lines (+158/-54) 13 files modified adb/adb-wrapper (+0/-30) adb/init.sls (+0/-11) lava/devices/fastmodels01/production/rtsm_ve-armv8_03.conf (+1/-1) lava/devices/fastmodels01/production/rtsm_ve-armv8_04.conf (+1/-1) lava/devices/fastmodels02/production/rtsm_ve-armv8_01.conf (+1/-1) lava/devices/fastmodels02/production/rtsm_ve-armv8_02.conf (+1/-1) lava/devices/fastmodels03/production/rtsm_ve-armv8_05.conf (+1/-1) lava/devices/fastmodels04/production/rtsm_ve-armv8_06.conf (+0/-2) lava/devices/fastmodels05/production/rtsm_ve-armv8_07.conf (+0/-2) lava/devices/fastmodels06/production/rtsm_ve-armv8_08.conf (+0/-2) lava/devices/fastmodels07/production/rtsm_ve-armv8_09.conf (+0/-2) lava/fastmodels/FMNetwork (+125/-0) lava/fastmodels/init.sls (+28/-0)
To merge this branch:	bzr merge lp://qastaging/~liuyq0307/lava-lab/fast-model-network
Related bugs:	Link a bug report

Reviewer	Date Requested	Status
Yongqin Liu		Approve on 2013-06-26
Ryan Harkin	2013-06-25	Pending
Review via email: mp+171271@code.qastaging.launchpad.net

This proposal supersedes a proposal from 2013-06-20.

Description of the change

Add the support of network for fast model instance.

This needs the RTSM_AEMv8_VE to be updated to RTSM_AEMv8_VE_0.8_4702.tgz version first.

tested the lava/fastmodels/FMNetwork script manually.
also tested running "update-rc.d FMNetwork defaults && /etc/init.d/FMNetwork start" from gateway manully.

but not tested the lava/fastmodels/init.sls script. so it would be appreciated if anyone could help to test it.
Also any comment would be appreciated.

Revision history for this message

Yongqin Liu (liuyq0307) wrote on 2013-06-20: Posted in a previous version of this proposal

when run /etc/init.d/FMNetwork start, it will cause the network loss temporarily,
will that cause the execution of salt to fail?

Revision history for this message

Dave Pigott (dpigott) wrote on 2013-06-20: Posted in a previous version of this proposal

If you were very unlucky on the timing, but otherwise it should be ok/

Dave

On 20 Jun 2013, at 10:43, Yongqin Liu <email address hidden> wrote:

> when run /etc/init.d/FMNetwork start, it will cause the network loss temporarily,
> will that cause the execution of salt to fail?
> --
> https://code.launchpad.net/~liuyq0307/lava-lab/fast-model-network/+merge/170497
> Your team Linaro Validation Team is requested to review the proposed merge of lp:~liuyq0307/lava-lab/fast-model-network into lp:lava-lab.

Revision history for this message

Yongqin Liu (liuyq0307) wrote on 2013-06-25:

Hi, Ryan

Could you help to check if there is any problem with the lava/fastmodels/FMNetwork script please?
I have tested it on LAVA manually, and it worked.
created two interfaces, and the fast model instance can use them to ping locally and internet when set the dns correctly.
But since I am not sure I have the correct understanding about the bridge usage,
so please tell me if there is anything not good.

Thanks,
Yongqin Liu

Revision history for this message

Ryan Harkin (ryanharkin) wrote on 2013-06-25:

On 25 June 2013 12:03, Yongqin Liu <email address hidden> wrote:
> Hi, Ryan
>
> Could you help to check if there is any problem with the lava/fastmodels/FMNetwork script please?
> I have tested it on LAVA manually, and it worked.
> created two interfaces, and the fast model instance can use them to ping locally and internet when set the dns correctly.
> But since I am not sure I have the correct understanding about the bridge usage,
> so please tell me if there is anything not good.
>

OK, I've had a look and it seems generally OK. But without testing it,
I can't be sure.

I doubt it's important for these virtual machines, but I added a check
to my FMNetwork script so that it doesn't start if it's already
started and doesn't stop if not running.

I also added a Status() function, but that's not important.

> Thanks,
> Yongqin Liu
> --
> https://code.launchpad.net/~liuyq0307/lava-lab/fast-model-network/+merge/171271
> You are requested to review the proposed merge of lp:~liuyq0307/lava-lab/fast-model-network into lp:lava-lab.

lp://qastaging/~liuyq0307/lava-lab/fast-model-network updated on 2013-06-26

190. By Yongqin Liu on 2013-06-26: merge with trunk

Revision history for this message

Yongqin Liu (liuyq0307) wrote on 2013-06-26:

Hi, Ryan

Thanks for your review.

> On 25 June 2013 12:03, Yongqin Liu <email address hidden> wrote:
> > Hi, Ryan
> >
> > Could you help to check if there is any problem with the
> lava/fastmodels/FMNetwork script please?
> > I have tested it on LAVA manually, and it worked.
> > created two interfaces, and the fast model instance can use them to ping
> locally and internet when set the dns correctly.
> > But since I am not sure I have the correct understanding about the bridge
> usage,
> > so please tell me if there is anything not good.
> >
>
> OK, I've had a look and it seems generally OK. But without testing it,
> I can't be sure.

I have tested it on LAVA, and it worked.

>
> I doubt it's important for these virtual machines, but I added a check
> to my FMNetwork script so that it doesn't start if it's already
> started and doesn't stop if not running.
On LAVA, we use the static ip for the bridge interface directly, not use the dhclient command.
so I think we can't check the status by checking the existence of the dhclient process.

And on LAVA, this script will be run only one time when the node boots up,
we will not run it many times. that means we will not run the start again after the node booted up,
and will not run the stop until the node shutdown.

So I think it's OK to not have the status check.

Thanks,
Yongqin Liu
> I also added a Status() function, but that's not important.
>
>
> > Thanks,
> > Yongqin Liu
> > --
> > https://code.launchpad.net/~liuyq0307/lava-lab/fast-model-
> network/+merge/171271
> > You are requested to review the proposed merge of lp:~liuyq0307/lava-lab
> /fast-model-network into lp:lava-lab.

Revision history for this message

Ryan Harkin (ryanharkin) wrote on 2013-06-26:

On 26 June 2013 03:13, Yongqin Liu <email address hidden> wrote:
> Hi, Ryan
>
> Thanks for your review.
>
>> On 25 June 2013 12:03, Yongqin Liu <email address hidden> wrote:
>> > Hi, Ryan
>> >
>> > Could you help to check if there is any problem with the
>> lava/fastmodels/FMNetwork script please?
>> > I have tested it on LAVA manually, and it worked.
>> > created two interfaces, and the fast model instance can use them to ping
>> locally and internet when set the dns correctly.
>> > But since I am not sure I have the correct understanding about the bridge
>> usage,
>> > so please tell me if there is anything not good.
>> >
>>
>> OK, I've had a look and it seems generally OK. But without testing it,
>> I can't be sure.
>
> I have tested it on LAVA, and it worked.

Great :-) That's what counts.

>
>>
>> I doubt it's important for these virtual machines, but I added a check
>> to my FMNetwork script so that it doesn't start if it's already
>> started and doesn't stop if not running.
> On LAVA, we use the static ip for the bridge interface directly, not use the dhclient command.
> so I think we can't check the status by checking the existence of the dhclient process.
>
> And on LAVA, this script will be run only one time when the node boots up,
> we will not run it many times. that means we will not run the start again after the node booted up,
> and will not run the stop until the node shutdown.
>
> So I think it's OK to not have the status check.

Agreed. I thought this was how it worked, so that's OK with me.

>
> Thanks,
> Yongqin Liu
>> I also added a Status() function, but that's not important.
>>
>>
>> > Thanks,
>> > Yongqin Liu
>> > --
>> > https://code.launchpad.net/~liuyq0307/lava-lab/fast-model-
>> network/+merge/171271
>> > You are requested to review the proposed merge of lp:~liuyq0307/lava-lab
>> /fast-model-network into lp:lava-lab.
> --
> https://code.launchpad.net/~liuyq0307/lava-lab/fast-model-network/+merge/171271
> You are requested to review the proposed merge of lp:~liuyq0307/lava-lab/fast-model-network into lp:lava-lab.

On 26 June 2013 03:13, Yongqin Liu <yongqin.liu@linaro.org> wrote:
> Hi, Ryan
>
> Thanks for your review.
>
>> On 25 June 2013 12:03, Yongqin Liu <yongqin.liu@linaro.org> wrote:
>> > Hi, Ryan
>> >
>> > Could you help to check if there is any problem with the
>> lava/fastmodels/FMNetwork script please?
>> > I have tested it on LAVA manually, and it worked.
>> > created two interfaces, and the fast model instance can use them to ping
>> locally and internet when set the dns correctly.
>> > But since I am not sure I have the correct understanding about the bridge
>> usage,
>> > so please tell me if there is anything not good.
>> >
>>
>> OK, I've had a look and it seems generally OK. But without testing it,
>> I can't be sure.
>
> I have tested it on LAVA, and it worked.

Great :-)  That's what counts.

Agreed.  I thought this was how it worked, so that's OK with me.

Revision history for this message

Yongqin Liu (liuyq0307) wrote on 2013-06-26:

tested it on lava and from concept Ryan is OK with it too

review: Approve