[dpdk-dev] [Bug 826] red_autotest random failures

[Danilewicz, MarcinX]

Hi Brandon,

that’s good news.

Regards,
/Marcin

-----Original Message-----
From: Brandon Lo <blo at iol.unh.edu> 
Sent: Wednesday, March 9, 2022 3:48 PM
To: Danilewicz, MarcinX <marcinx.danilewicz at intel.com>
Cc: Lincoln Lavoie <lylavoie at iol.unh.edu>; Dumitrescu, Cristian <cristian.dumitrescu at intel.com>; Ajmera, Megha <megha.ajmera at intel.com>; Singh, Jasvinder <jasvinder.singh at intel.com>; Zegota, AnnaX <annax.zegota at intel.com>; Yigit, Ferruh <ferruh.yigit at intel.com>; thomas at monjalon.net; david.marchand at redhat.com; ci at dpdk.org
Subject: Re: [dpdk-dev] [Bug 826] red_autotest random failures

On Wed, Mar 9, 2022 at 5:01 AM Danilewicz, MarcinX <marcinx.danilewicz at intel.com> wrote:
>
> Hi Brandon,
>
> Sorry for late response, but I was busy in mean time. But after your mail well .. I've tried to run test dpdk from docker image. Few instances of images in parallel, enough to get machine fully loaded. But in turn, red_autotest never failed.
>
> Is it possible for you to share some additional details? About hardware used for testing, memory sizes, etc. To get some hint how to get these failures. I've seen other test failing constantly, depending on machine I was running other autotests. Maybe tests before red_autotest are changing hardware to the state where red_autotest is failing. Anyone tried to change autotests execution order?

Hi Marcin,

Unfortunately, I don't have any more details other than the ones we talked about before. It is possible that the issue is not as common now that we limited the number of compile jobs that can happen on each machine at the same time.
We limited the amount of RAM that each job can use again so that the systems are not getting a high load.

> Also, I've don’t almost all to reproduce error and perhaps it is better to ignore that random error for now. It looks like you are able to successfully pass that test all the time, even when is failing from time to time. Right? If that is the true error, it will come out elsewhere.

Yes, I think it's ok if you think the test is good for now. I haven't seen it fail in a while, so it might just be due to the load we put on the systems.
If the issue comes up, I can contact you again.

Thanks,
Brandon



--
Brandon Lo
UNH InterOperability Laboratory
21 Madbury Rd, Suite 100, Durham, NH 03824 blo at iol.unh.edu www.iol.unh.edu
--------------------------------------------------------------
Intel Research and Development Ireland Limited
Registered in Ireland
Registered Office: Collinstown Industrial Park, Leixlip, County Kildare
Registered Number: 308263


This e-mail and any attachments may contain confidential material for the sole
use of the intended recipient(s). Any review or distribution by others is
strictly prohibited. If you are not the intended recipient, please contact the
sender and delete all copies.