- lava-devel - op-lists.linaro.org

Re: [Lava-devel] Integration of a new device

by koti koti

++ Lava-users list. Regards, Koti On Fri, 4 Dec 2020 at 17:30, <lava-devel-request(a)lists.lavasoftware.org> wrote: > Send Lava-devel mailing list submissions to > lava-devel(a)lists.lavasoftware.org > > To subscribe or unsubscribe via the World Wide Web, visit > https://lists.lavasoftware.org/mailman/listinfo/lava-devel > or, via email, send a message with subject or body 'help' to > lava-devel-request(a)lists.lavasoftware.org > > You can reach the person managing the list at > lava-devel-owner(a)lists.lavasoftware.org > > When replying, please edit your Subject line so it is more specific > than "Re: Contents of Lava-devel digest..." > > > Today's Topics: > > 1. Integration of a new device (Marcel Trattner) > > > ---------------------------------------------------------------------- > > Message: 1 > Date: Fri, 4 Dec 2020 10:50:08 +0000 > From: Marcel Trattner <marcel.trattner(a)expleogroup.com> > To: "Lava-devel(a)lists.lavasoftware.org" > <Lava-devel(a)lists.lavasoftware.org> > Cc: Alexander Wyron Wachtberger > <alexander-wyron.wachtberger(a)expleogroup.com>, Vladimir Schmidt > <vladimir.schmidt(a)expleogroup.com> > Subject: [Lava-devel] Integration of a new device > Message-ID: > < > AM9PR10MB4295B7516795BD7B514C2DF7E5F10(a)AM9PR10MB4295.EURPRD10.PROD.OUTLOOK.COM > > > > Content-Type: text/plain; charset="us-ascii" > > Dear LAVA-Team, > > We have only recently started to use LAVA in a new project and we are now > at the point of integrating a new device. As I have read in the > documentation that this is the hardest part I would like to ask for some > advice where to start and what to especially look out for. > > The current state is that LAVA, as a master and one slave is running in > two Docker containers by using the example of the following Github > repository: > > https://github.com/kernelci/lava-docker > > After reading the documentary extensively and trying things out with qemu > I'd like to try and integrate the following device: > > https://www.lantronix.com/products/sa8155p-automotive-development-platform/ > > Our plan so far would be: > Since it has a Snapdragon processor, it's running Android and it is using > fastboot for flashing, we figured we could use one of the dragonboard > Device-Type templates and duplicate it for our device. But as it comes now > to configuring the device dictionary and specifying the test jobs we are > not so sure as how to proceed and we also don't want to damage the > hardware, so we would be grateful for some help. > > Here some more information: > The DUT is connected via USB and its power supply can be turned on and off > by using Python scripts (When defining the power_on and power_off commands, > should the scripts then be on the master or on the dispatcher? And what are > all the necessary parameters for the power on/off and also the values > connection_list and connection_commands?). > It is running Android with the underlying Linux configured especially for > the hardware, so we would like to avoid deploying for the beginning and > only run some simple shell tests to see if the connection stands. > A very simple health check has been set up here< > https://paste.debian.net/1175549>, maybe you could give some feedback if > that would be sufficient and how future test jobs should be altered. > > Hope to hear from you soon > > Best Regards, > > > _________________________________________________________ > > EMAIL LEGAL MENTION / DISCLAIMER > > This message contains information that may be privileged or confidential > and is the property of the Expleo Services SAS, RCS Versailles 831 178 348, > located, 3 avenue des Pr?s, 78180 Montigny Le Bretonneux - France. It is > intended only for the person to whom it is addressed. If you are not the > intended recipient, you are not authorized to read, print, retain, copy, > disseminate, distribute, or use this message or any part thereof. If you > receive this message in error, please notify the sender immediately and > delete all copies of this message. > > Ce message contient des informations qui peuvent ?tre privil?gi?es ou > confidentielles et elles sont de la propri?t? d'Expleo Services SAS, RCS > Versailles 831 178 348, situ?, 3 avenue des Pr?s, 78180 Montigny le > Bretonneux-France. Il est destin? uniquement ? la personne ? qui est > adress?. Si vous n'?tes pas le destinataire vis?, vous n'?tes pas autoris? > ? lire, imprimer, conserver, copier, diffuser, distribuer ou utiliser ce > message ou toute partie de celui-ci. Si vous recevez ce message par erreur, > veuillez en avertir imm?diatement l'exp?diteur et supprimer toutes les > copies de ce message. >

5 years, 5 months

1
0
0 0

Re: [Lava-devel] [Lava-users] LAVA tests execution on already provisioned (boot) begalbone board

by Milosz Wasilewski

On Tue, 17 Mar 2020 at 12:30, koti koti <kotisoftwaretest(a)gmail.com> wrote: > > Hi, > > I have successfully booted the Beagelbone board from "https://github.com/danrue/lava.therub.org" . (corresponding ymal is https://github.com/danrue/lava.therub.org/blob/master/server-overlay/etc/la…) > > But, now I am trying to run one more scenario (may be new scenario and not sure is it supported by LAVA lab?) i.e > > 1. Run the tests on already provisioned (boot) beagelbone board. (Basically I am skipping the booting mentioned and trying to run on the already provisioned/boot board) > a) boot the target > b) Connect Board to LAVA lab > c) Just check the login prompt ("#") is available or not? > c) Run the using below test definition file (Basically this test definition file runs "ls"/"ifconfig" commands in the already provisioned(boot) board). > > " > device_type: beaglebone-black > job_name: beaglebone-black healthcheck > timeouts: > job: > minutes: 10 > action: > minutes: 5 > priority: high > visibility: public > actions: > - test: > interactive: > - name: ls test > prompts: ["#"] > echo: discard > script: > - name: ls > command: ls > successes: > - message: "ls simple test successes" > failures: > - message: "TIMEOUT" > exception: InfrastructureError > error: "ls command failed" > - name: ifconfig > command: ifconfig > - name: wait for the prompt > command: > " > > I have tried to run this. But, my test failed with the error ""Connection closed"" (Attached screenshot) Connection closed is usually not sth wrong with LAVA but with serial. Are you able to connect to the board by telneting to your ser2net port? I assume you use ser2net as in Dan's example. If not, could you explain how do you connect to the BBB's UART? milosz > > Can some one let me know solution to fix this error? > > Regards, > koti > _______________________________________________ > Lava-users mailing list > Lava-users(a)lists.lavasoftware.org > https://lists.lavasoftware.org/mailman/listinfo/lava-users

6 years, 2 months

2
1
0 0

Design meeting minutes from 2019-12-18

by Remi Duraffort

Hi folks, I forget to send the last design meeting notes that you can find on https://git.lavasoftware.org/lava/lava/wikis/design-meetings We held our regular design meeting today via Hangout. Summary of brief discussion: # Connect presentations? [Rémi] * Docker deploy/boot/test for adb? * [Antonio] will make this presentation * LAVA rest api? * Some slides in the LAVA USers forum * Remote labs setup? * [milosz] will make this presentation * LAVA users forum * [Rémi] will submit the talk # documentation layout [Rémi] * http://people.linaro.org/~remi.duraffort/site/ Comments? * [Antonio] issues with the menu on smaller screens * Some CSS to change? # read the doc [Milosz] Will ping the admins to get back the lava project # port of lava-coordinator to python3 [Antonio] * [Rémi] Already working under python3 * Antonio will update the packaging to provide the binary from src:lava and drop src:lava-coordinator # Docker support coming up soon [Antonio] * With this patch all three methods works * Bare metal * lxc * Docker * So this is safe to merge and test on staging * Won’t be able to test fastboot_via_uboot locally ============================================================================ The LAVA design meeting is held weekly, every Wednesday at 13:00 to 14:00 UTC using Google Hangouts Meet: https://meet.google.com/usu-aatj-fht Feel free to comment here or join us directly in the meeting. Minutes from this and previous meetings are also stored in the LAVA wiki: https://git.lavasoftware.org/lava/lava/wikis/design-meetings/index Cheers, -- Rémi Duraffort LAVA Architect Linaro

6 years, 5 months

1
0
0 0

Design meeting minutes from 2019-09-11

by Remi Duraffort

Hi folks, We held our regular design meeting today via Hangout. Summary of brief discussion: # 11th September 2019 # Reducing the privilege of lava-slave [Antonio] It’s more or less ok that lava-run requires being run as root. But can we run lava-slave as non-root? Lava-slave needs to cleanup after lava-run if something goes wrong, so there might be stuff owned by root left over and this needs to be handled. Otherwise it’s `go for it`. # Replacing LXC with docker [Antonio] How exactly do we want to do that: See https://git.lavasoftware.org/lava/lava/issues/305 and https://git.lavasoftware.org/lava/lava/issues/286 The goal is to: * simplify the job definition (see the job definition in https://git.lavasoftware.org/lava/lava/issues/305) * allow to use user provided docker container to run adb # Job level privileges [milosz] How does that work compared to device and device-type level privileges? See https://git.lavasoftware.org/lava/lava/merge_requests/693 for more documentation The job level permission are still applied, but the device-type and device permissions are also applied (was not the case before). # LAVA package uploads to Debian? [Steve] I don't have time to do it anymore, and we have RC bugs: * some django2 issues too? patches from Antonio * One more patch https://git.lavasoftware.org/lava/lava/merge_requests/696 * Gunicorn problem? Need to update the dependency * VLANd python3 conversion underway * lava-coordinator debian package depends on python2 # New maintainer for vland [milosz] Since Steve is moving to a new project we’ll need a new maintainer for vland. # SAN19 - all stuff up to date and slides done? [Steve] Any meetings to plan? * lavafed # Moving docker base images to buster? [Rémi] When should we use buster as the base image? Will be the base image for 2019.10 # Hangout meetings links changed [Steve] The documentation should be updated. ============================================================================ The LAVA design meeting is held weekly, every Wednesday at 13:00 to 14:00 UTC using Google Hangouts Meet: https://meet.google.com/usu-aatj-fht Feel free to comment here or join us directly in the meeting. Minutes from this and previous meetings are also stored in the LAVA wiki: https://git.lavasoftware.org/lava/lava/wikis/design-meetings/index Cheers, -- Rémi Duraffort LAVA Team, Linaro

6 years, 8 months

1
0
0 0

LAVA 2019.07 dependencies

by Dave Pigott

Hi Remi, When I was doing the lkft upgrade this morning I noticed that there is a Buster dependency for LAVA. One of the workers had somehow been missed out on the Buster upgrade and was still on Stretch. I discovered the following: lava-dispatcher : Depends: python3-requests (>= 2.20.0) but 2.12.4-1 is to be installed Does this mean we really have to check for Debian version 10.0 or later for remote, bare metal, workers? Thanks Dave ---------------- Dave Pigott LAVA Lab Lead Linaro Ltd t: (+44) (0) 1223 400063

6 years, 9 months

2
1
0 0

Design meeting minutes from 2019-07-31

by Remi Duraffort

Hi folks, We held our regular design meeting today via Hangout. Summary of brief discussion: # 31st July 2019 # Using celery inside LAVA [Rémi] Could be used for: * Parsing description.yaml * Replace crontabs * Compressing logs * Removing old jobs * Sending logs to ES * Splitting scheduling * notifications # August release ? [Steve] Yes - aim for 28th The release will have database changes Tag the day before to let some time for lavafed to test the tag # LAVA sessions for Connect [Steve] Three sessions has been accepeted: * LAVA Users' Forum * Hacking and contributing to LAVA * Advanced testing in python # Support packages in Debian removed/being removed [Steve] * lava-tool removed (https://tracker.debian.org/pkg/lava-tool) * django-compat marked for removal ( https://tracker.debian.org/pkg/django-compat) * django-hijack depends on django-compat, marked for removal too ( https://tracker.debian.org/pkg/django-hijack) * Rémi will remove the dependency/mention * VLANd needs some rework to go to [python 3]( https://git.lavasoftware.org/lava/vland/issues/5) # Migrating to django 2.2 (next LTS release) [Milosz] * lava source code itself is compatible with django 2.2 * Dependencies on filters and rest framework * Still waiting for a release compatible with django 2.0 * Trying to contact the maintainer to give some help (if needed) # gitlab-runner on aarch64 [Steve] The current package in [Debian]( https://tracker.debian.org/pkg/gitlab-ci-multi-runner) was a bit too old. GitLab don't provide anything for arm64 but the Debian maintainer uploaded a new version that fixe our issues. ============================================================================ The LAVA design meeting is held weekly, every Wednesday at 13:00 to 14:00 UTC using Google Hangouts Meet: https://meet.google.com/qre-rgen-zwc Feel free to comment here or join us directly in the meeting. Minutes from this and previous meetings are also stored in the LAVA wiki: https://git.lavasoftware.org/lava/lava/wikis/design-meetings/index Cheers, -- Rémi Duraffort LAVA Team, Linaro

6 years, 10 months

1
0
0 0

Design meeting minutes from 2019-07-24

by Remi Duraffort

Hi folks, We held our regular design meeting today via Hangout. Summary of brief discussion: # 24th July 2019 # install.git-deps [milosz] This feature works nicely: https://lkft.validation.linaro.org/scheduler/job/834097 Proposal: keep `install` option but restrict it so it’s not trying to install system packages. [Rémi] Will submit a patch to remove the “deprecation” warning in the documentation. # Authentication refactoring [milosz] Under review by Remi. Looks good. # Connect sessions where accepted [Rémi] * LAVA users forum * Hacking and contributing to LAVA * Advanced testing in python # Playing with Sentry error reporting [Rémi] * Will create a ticket to have it installed in the linaro lab. * Will create sentry.lavasoftware.org * No debian package available for python3-sentry-sdk * Should be installed from pip (sentry-sdk) * Will send a patch to install sentry-sdk from pip in lava-server docker container. * Activate it for lavafed instances. ============================================================================ The LAVA design meeting is held weekly, every Wednesday at 13:00 to 14:00 UTC using Google Hangouts Meet: https://meet.google.com/qre-rgen-zwc Feel free to comment here or join us directly in the meeting. Minutes from this and previous meetings are also stored in the LAVA wiki: https://git.lavasoftware.org/lava/lava/wikis/design-meetings/index Cheers, -- Rémi Duraffort LAVA Team, Linaro

6 years, 10 months

1
0
0 0

Design meeting minutes from 2019-07-17

by Remi Duraffort

Hi folks, We held our regular design meeting today via Hangout. Summary of brief discussion: # 17th July 2019 # Large job definitions causing outages [deanb] * Issue: https://git.lavasoftware.org/lava/lava/issues/299 * Wondering if for large jobs (configurable limit) simply not making ActionData objects is a sensible approach. * Tried this: https://git.lavasoftware.org/dean-birch/lava/commit/dd220c0bd82bf092e35e643… * In my test instance this reduced outage to 30 seconds (from hours). * If not, what else can we do? * Anything extra needs to be added? * Documentation? * [deanb] will send a patch with the first improvements (CLoader) * [deanb] will look at using bulk save to save all objects in one call * [stevan] investigate ActionData: is it possible to create them later on or even maybe not creating them? # Test from inline with git [milosz] The idea is to source test-definition YAML from inline but use git repository to prepare overlay. Example: https://github.com/andersson/bootrr Bjorn doesn’t want to have YAML file in this repository [Rémi] Using install.git-repos might work * https://git.lavasoftware.org/lava/lava/blob/master/lava_dispatcher/tests/te… * https://docs.lavasoftware.org/lava/lava_test_shell.html#adding-git-bzr-repo… [milosz] will try install.git-repos If that’s working Rémi will add some tests in lavafed or meta-lava # Switching between serial connections on device with multiple UARTs [Malcolm Brooks] * Issue: We have devices which use separate serial outputs for MCC, AP and SCP UARTs. * Workaround: Use the `new_connection` boot method to switch between UART1 and UART2 in order to catch the kernel booting once MCC flash stage is complete. * Idea: Allow all connections (or possibly a subset defined in the “connection_tags” for example) to be established and followed from the beginning of the job, and allow each action/stage to select which they are actually listening/interacting with the `connection` option (example below). ```yaml - boot: namespace: target connection: uart1 method: minimal ``` [Rémi] Sounds like a good idea. * Using feedback LAVA can already use one connection and listen/print the other ones * Malcom will create an issue on gitlab. ============================================================================ The LAVA design meeting is held weekly, every Wednesday at 13:00 to 14:00 UTC using Google Hangouts Meet: https://meet.google.com/qre-rgen-zwc Feel free to comment here or join us directly in the meeting. Minutes from this and previous meetings are also stored in the LAVA wiki: https://git.lavasoftware.org/lava/lava/wikis/design-meetings/index Cheers, -- Rémi Duraffort LAVA Team, Linaro

6 years, 10 months

1
0
0 0

Large job definitions in LAVA

by Dean Birch

Hi All, We've previously had an issue on our LAVA instance where it stopped responding to workers and stopped dispatching jobs when it finished running large job definition (around 25000 lines in the definition, around 1000 deploy/boot/test actions). I've been looking into reproducing this safely in a development environment, and I've got a few observations and questions about how the situation could be improved. The lava-master process appears be stuck processing the job results, and takes a painstakingly long time to finish this and send an ACK for END_OK. During this processing, the master doesn't respond to worker pings, and doesn't schedule other jobs. Tracking a bit deeper, it seems that the vast majority of time (I've never seen it finish as I have always restarted the lava services before it finishes) in the walk_actions and build_action functions of the lava_results_app/dbutils.py file: https://git.lavasoftware.org/lava/lava/blob/2019.05.post1/lava_results_app/… https://git.lavasoftware.org/lava/lava/blob/2019.05.post1/lava_results_app/… What options is there to mitigate this issue? Some ideas below: - Could we optimize the build_action function? There are a few Django model/db queries in build_action, could some results be queried once and cached? With an obscenely large job, would this even give us enough savings to make the time invested in safely optimizing this worth it? - What are the implications of not having created ActionData objects for a job? Does this mean that no options will be available in the "Pipeline ↓" drop-down on the job page for quick navigation? Could we optionally abort after a certain amount of these (and make it configurable per LAVA instance)? - Should/could the handling of the results be forked off, so lava-master can continue to schedule more jobs and respond to worker pings, but slowly the ActionData objects can be populated? I'm unsure if you have to be on a special thread to write to Django models. Even if this could be done, would any weird behaviours occur on the slave side as it will still be waiting for the ACK for END_OK from the master? Any guidance on how to proceed with this would be appreciated! I'm happy to place this and some more details in as a LAVA issue on git.lavasoftware.org if this is easier to track and discuss. Thanks, Dean

6 years, 10 months

2
2
0 0

No design meeting for 2019-07-10

by Steve McIntyre

Hi folks, As Rémi and Stevan are both out and we don't have any items listed for discussion in advance, I'm cancelling today's design meeting. Cheers, -- Steve McIntyre steve.mcintyre(a)linaro.org <http://www.linaro.org/> Linaro.org | Open source software for ARM SoCs

6 years, 10 months

1
0
0 0

2026

2025

2024

2023

2022

2021

2020

2019

2018

lava-devel