Hi,
I'm trying to use a situation where a large bulk of similar jobs are being sent to several devices of the same device type. These devices require the 'transfer_overlay' option in order for the tests to work due to their setup. We currently are using a single master and worker on the same server.
It seems that the same transfer overlay file location (i.e., /var/lib/lava/dispatcher/tmp/overlay-1.3.2.5.tar.gz) is used regardless of the job id and that some of our jobs are experiencing a collision trying to use different copies of that file at the same time. Thus, one of our jobs is deleting the transfer overlay file before another job tries to delete it and the second gets errors.
This is the error LAVA produces when viewing the job in the GUI:
File "/usr/bin/lava", line 11, in <module>
load_entry_point('lava-tool==0.25', 'console_scripts', 'lava')()
File "/usr/lib/python2.7/dist-packages/lava/tool/dispatcher.py", line 150, in run
raise SystemExit(cls().dispatch(args))
File "/usr/lib/python2.7/dist-packages/lava/tool/dispatcher.py", line 140, in dispatch
return command.invoke()
File "/usr/lib/python2.7/dist-packages/lava/dispatcher/commands.py", line 247, in invoke
self.args.validate)
File "/usr/lib/python2.7/dist-packages/lava/dispatcher/commands.py", line 91, in run_pipeline_job
exitcode = job.run()
File "/usr/lib/python2.7/dist-packages/lava_dispatcher/pipeline/job.py", line 340, in run
self.cleanup(self.connection)
File "/usr/lib/python2.7/dist-packages/lava_dispatcher/pipeline/job.py", line 375, in cleanup
self.pipeline.cleanup(connection)
File "/usr/lib/python2.7/dist-packages/lava_dispatcher/pipeline/action.py", line 251, in cleanup
child.cleanup(connection)
File "/usr/lib/python2.7/dist-packages/lava_dispatcher/pipeline/action.py", line 649, in cleanup
self.internal_pipeline.cleanup(connection)
File "/usr/lib/python2.7/dist-packages/lava_dispatcher/pipeline/action.py", line 251, in cleanup
child.cleanup(connection)
File "/usr/lib/python2.7/dist-packages/lava_dispatcher/pipeline/action.py", line 649, in cleanup
self.internal_pipeline.cleanup(connection)
File "/usr/lib/python2.7/dist-packages/lava_dispatcher/pipeline/action.py", line 251, in cleanup
child.cleanup(connection)
File "/usr/lib/python2.7/dist-packages/lava_dispatcher/pipeline/actions/boot/__init__.py", line 458, in cleanup
os.unlink(self.url)
OSError: [Errno 2] No such file or directory: '/var/lib/lava/dispatcher/tmp/overlay-1.3.2.5.tar.gz'
We are working on the 2017.7 build of LAVA, so if this has been changed in a later build, please let me know. If this has not been updated, is there another way to change things so that each job uses a different name and/or location for the overlay file?
Thanks,
Jorgen Peddersen
Jorgen Peddersen
Researcher
t +61 2 6103 4700
m
e jorgen.peddersen(a)seeingmachines.com
a 80 Mildura Street, Fyshwick ACT 2609, Australia
SEEINGMACHINES.COM<http://seeingmachines.com>
[SEEINGMACHINES]
This email may contain confidential information. If you are not the intended recipient, destroy all copies and do not disclose or use the information within. Any content that is not concerned with the business of the Seeing Machines Group reflects the views of the sender only, not those of Seeing Machines. No warranties are given that this email does not contain viruses or harmful code. [http://www.seeingmachines.com/wp-content/uploads/2018/04/sm-email-twitter-i…] <https://twitter.com/seeingmachines> [http://www.seeingmachines.com/wp-content/uploads/2018/04/sm-email-linkedin-…] <https://www.linkedin.com/company/258949/>
Hello
I have queued jobs that I want to cancel but I get instead:
500 Internal Server Error
[Errno 2] No such file or directory: '/var/lib/lava-server/default/media/job-output/2018/07/10/7181/job_data.gz'
when I click on cancel.
Note that thoses queued jobs are only queued and not assigned to any device.
Regards
Hi,
I'm sending this email to confirm if anyone has encountered the same problem as me when using LAVA...I have installed a lava single node instance in a docker container and are trying to add an Android device for testing. But when I ran the adb devices command, in addition to the Android device I connected, an "emulator-5554 connecting" appeared. I have tried many ways to remove this device but I failed. I am sure that the emulator appeared after the docker container containing LAVA instance started, so I wonder if this issue is related to LAVA.
Thanks
Jiayan Kong
> Just configure the worker to connect to your master.
> This is done in /etc/lava-dispatcher/lava-slave
> An example configuration file is available at /usr/share/lava-dispatcher/
lava-slave
>
> For a first try, you can start without encryption and only define the
following variables in the configuration file:
> MASTER_URL="tcp://localhost:5556"
> LOGGER_URL="tcp://localhost:5555"
Hello Remi,
Do you, Linaro guys, have some Open Source Lava Server on the Cloud, where
I can create lava admin account, hook to it my Lava Worker and try it?
I need some access to create device-type BeagleBone Black and device BBB0x
as well.
This will be interesting experiment, don't you agree?
Thank you,
Zoran Stojsavljevic
_______
On Tue, Aug 7, 2018 at 1:53 PM, Remi Duraffort <remi.duraffort(a)linaro.org>
wrote:
> Hello,
>
> the worker will register itself to the master automatically. You don't
> have to create the worker manually.
>
> Just configure the worker to connect to your master.
> This is done in /etc/lava-dispatcher/lava-slave
> An example configuration file is available at /usr/share/lava-dispatcher/
> lava-slave
>
> For a first try, you can start without encryption and only define the
> following variables in the configuration file:
> MASTER_URL="tcp://localhost:5556"
> LOGGER_URL="tcp://localhost:5555"
>
> You can also specify the HOSTNAME if you want. By default the slave will
> use the name of the server it's running on.
>
> This will instruct you local lava-slave (this is the name of the service)
> to connect to the local lava-master.
> Then restart the lava-slave with "service lava-slave restart"
>
>
The document about installing lava system, does "Single Master Instance installation " mean master(include lava-server) and worker(include lava-dispatcher) can be installed in the one physical computer? In this simple mode,wheather the worker(logic unit)must only have one?
I am ready to deploy a local lava single master demo system. The server and dispatcher had be installed in a single computer.
In order to enable the server to kno the dispatcher(worker),Then how to configure something.
Is doing this?:
sudo lava-server manage workers add <HOSTNAME> ,that HOSTNAME is changed in /etc/lava-dispatcher/lava-slave file?
And when did,How to see whether dispatcher is working already?
Hi,
According to lava installation document ,I had installed lava-server to debian 9.5.0.Creating super user operation is ok(sudo lava-server manage createsuperuser --username $USERNAME --email=$EMAIL) . But when using firefox browser to login in by created name , response message:
--
SRF verification failed. Request aborted.
You are seeing this message because this site requires a CSRF cookie when submitting forms. This cookie is required for security reasons, to ensure that your browser is not being hijacked by third parties.
If you have configured your browser to disable cookies, please re-enable them, at least for this site, or for 'same-origin' requests.
--
After configured browser to disable cookies,the response is still so.
Jiang Lao
Hi,If there is no “dameon” program in DUT. How to add/update the program running in DUT
for example application test program or linux kernel or uboot etc.?Jiang Lao
--------------------------------------------
On Wed, 1 Aug 2018 at 10:18, ljh_dev <ljh_dev at 126.com> wrote:
> Hi,
> I have browsed the website documentation of lava overall. I think the
> DUT(Device Under Test) should has "dameon" program to communicate with the
> Worker Dispatcher module. But I did not find where the DUT "dameon"
> program source code is,and how to build&install this DUT "dameon" program.
>
It does not need to be built or installed. The lava-dispatcher package does
all that work for you.
https://validation.linaro.org/static/docs/v2/index.html#architecturehttps:/…
/var/log/lava-dispatcher/lava-slave.log
If there is no "dameon" program in DUT. How to add/update the program running in DUT for example application test program or linux kernel or uboot etc.?
Best regards,
