Hi, there,
Frequently, I found when a job becomes incomplete there will immediately a health check job on that device happen automatically to check the status of the device. But strange, some times I won't see that healthy check job.
So, my question is: when I see that healthy check job, is it just by chance? Or it really designed that there will be a healthy check after incomplete job? Thanks.
Regards, Larry
On Wed, 15 Apr 2020 at 08:24, Larry Shen larry.shen@nxp.com wrote:
Hi, there,
Frequently, I found when a job becomes incomplete there will immediately a health check job on that device happen automatically to check the status of the device.
But strange, some times I won’t see that healthy check job.
So, my question is: when I see that healthy check job, is it just by chance? Or it really designed that there will be a healthy check after incomplete job? Thanks.
IIRC health check is triggered always after 'infrastructure error'. So if your test job goes incomplete due to timeout in a test shell there won't be a health check after it. But if there is an error caused by infrastructure (in LAVA terms), for example 'connection closed', a health check will be triggered. You can also configure LAVA to run health check after N ordinary jobs. I can imagine N=1. This has a practical application. For example health check might bring the board to the desirable condition.
milosz
Regards,
Larry
Lava-users mailing list Lava-users@lists.lavasoftware.org https://lists.lavasoftware.org/mailman/listinfo/lava-users
Thanks Milosz for the helpful information which clear my concern.
Although I don't need it currently, but still I'm interested to know the detail about "You can also configure LAVA to run health check after N ordinary jobs".
Where can I do this configure?
-----Original Message----- From: Milosz Wasilewski milosz.wasilewski@linaro.org Sent: Wednesday, April 15, 2020 3:40 PM To: Larry Shen larry.shen@nxp.com Cc: lava-users@lists.lavasoftware.org Subject: [EXT] Re: [Lava-users] About health check behavior?
Caution: EXT Email
On Wed, 15 Apr 2020 at 08:24, Larry Shen larry.shen@nxp.com wrote:
Hi, there,
Frequently, I found when a job becomes incomplete there will immediately a health check job on that device happen automatically to check the status of the device.
But strange, some times I won’t see that healthy check job.
So, my question is: when I see that healthy check job, is it just by chance? Or it really designed that there will be a healthy check after incomplete job? Thanks.
IIRC health check is triggered always after 'infrastructure error'. So if your test job goes incomplete due to timeout in a test shell there won't be a health check after it. But if there is an error caused by infrastructure (in LAVA terms), for example 'connection closed', a health check will be triggered. You can also configure LAVA to run health check after N ordinary jobs. I can imagine N=1. This has a practical application. For example health check might bring the board to the desirable condition.
milosz
Regards,
Larry
Lava-users mailing list Lava-users@lists.lavasoftware.org https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Flist s.lavasoftware.org%2Fmailman%2Flistinfo%2Flava-users&data=02%7C01% 7Clarry.shen%40nxp.com%7C689acd2a533443d7ac7008d7e1103cbf%7C686ea1d3bc 2b4c6fa92cd99c5c301635%7C0%7C0%7C637225332166930621&sdata=80rgmQVq S%2BDb9bkJOIrlaIPKiAvWVT8aLuAkjYfIjX8%3D&reserved=0
On Wed, 15 Apr 2020 at 08:48, Larry Shen larry.shen@nxp.com wrote:
Thanks Milosz for the helpful information which clear my concern.
Although I don't need it currently, but still I'm interested to know the detail about "You can also configure LAVA to run health check after N ordinary jobs".
Where can I do this configure?
In admin UI for device type or through REST API. For example: https://validation.linaro.org/api/v0.2/devicetypes/panda/ You can see:
"health_frequency": 24, "health_denominator": "hours",
Frequency is pretty obvious, denomination can be 'hours' or 'jobs'. I'm not 100% sure if health_denomination can be changed through REST API. Default form doesn't have this option.
milosz
-----Original Message----- From: Milosz Wasilewski milosz.wasilewski@linaro.org Sent: Wednesday, April 15, 2020 3:40 PM To: Larry Shen larry.shen@nxp.com Cc: lava-users@lists.lavasoftware.org Subject: [EXT] Re: [Lava-users] About health check behavior?
Caution: EXT Email
On Wed, 15 Apr 2020 at 08:24, Larry Shen larry.shen@nxp.com wrote:
Hi, there,
Frequently, I found when a job becomes incomplete there will immediately a health check job on that device happen automatically to check the status of the device.
But strange, some times I won’t see that healthy check job.
So, my question is: when I see that healthy check job, is it just by chance? Or it really designed that there will be a healthy check after incomplete job? Thanks.
IIRC health check is triggered always after 'infrastructure error'. So if your test job goes incomplete due to timeout in a test shell there won't be a health check after it. But if there is an error caused by infrastructure (in LAVA terms), for example 'connection closed', a health check will be triggered. You can also configure LAVA to run health check after N ordinary jobs. I can imagine N=1. This has a practical application. For example health check might bring the board to the desirable condition.
milosz
Regards,
Larry
Lava-users mailing list Lava-users@lists.lavasoftware.org https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Flist s.lavasoftware.org%2Fmailman%2Flistinfo%2Flava-users&data=02%7C01% 7Clarry.shen%40nxp.com%7C689acd2a533443d7ac7008d7e1103cbf%7C686ea1d3bc 2b4c6fa92cd99c5c301635%7C0%7C0%7C637225332166930621&sdata=80rgmQVq S%2BDb9bkJOIrlaIPKiAvWVT8aLuAkjYfIjX8%3D&reserved=0
Got it, thanks!
-----Original Message----- From: Milosz Wasilewski milosz.wasilewski@linaro.org Sent: Wednesday, April 15, 2020 3:59 PM To: Larry Shen larry.shen@nxp.com Cc: lava-users@lists.lavasoftware.org Subject: Re: [EXT] Re: [Lava-users] About health check behavior?
Caution: EXT Email
On Wed, 15 Apr 2020 at 08:48, Larry Shen larry.shen@nxp.com wrote:
Thanks Milosz for the helpful information which clear my concern.
Although I don't need it currently, but still I'm interested to know the detail about "You can also configure LAVA to run health check after N ordinary jobs".
Where can I do this configure?
In admin UI for device type or through REST API. For example: https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fvalidation... You can see:
"health_frequency": 24, "health_denominator": "hours",
Frequency is pretty obvious, denomination can be 'hours' or 'jobs'. I'm not 100% sure if health_denomination can be changed through REST API. Default form doesn't have this option.
milosz
-----Original Message----- From: Milosz Wasilewski milosz.wasilewski@linaro.org Sent: Wednesday, April 15, 2020 3:40 PM To: Larry Shen larry.shen@nxp.com Cc: lava-users@lists.lavasoftware.org Subject: [EXT] Re: [Lava-users] About health check behavior?
Caution: EXT Email
On Wed, 15 Apr 2020 at 08:24, Larry Shen larry.shen@nxp.com wrote:
Hi, there,
Frequently, I found when a job becomes incomplete there will immediately a health check job on that device happen automatically to check the status of the device.
But strange, some times I won’t see that healthy check job.
So, my question is: when I see that healthy check job, is it just by chance? Or it really designed that there will be a healthy check after incomplete job? Thanks.
IIRC health check is triggered always after 'infrastructure error'. So if your test job goes incomplete due to timeout in a test shell there won't be a health check after it. But if there is an error caused by infrastructure (in LAVA terms), for example 'connection closed', a health check will be triggered. You can also configure LAVA to run health check after N ordinary jobs. I can imagine N=1. This has a practical application. For example health check might bring the board to the desirable condition.
milosz
Regards,
Larry
Lava-users mailing list Lava-users@lists.lavasoftware.org https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fli st s.lavasoftware.org%2Fmailman%2Flistinfo%2Flava-users&data=02%7C0 1% 7Clarry.shen%40nxp.com%7C689acd2a533443d7ac7008d7e1103cbf%7C686ea1d3 bc 2b4c6fa92cd99c5c301635%7C0%7C0%7C637225332166930621&sdata=80rgmQ Vq S%2BDb9bkJOIrlaIPKiAvWVT8aLuAkjYfIjX8%3D&reserved=0
lava-users@lists.lavasoftware.org
-
Larry Shen -
Milosz Wasilewski