Jump to content

THE testing thread


lanefu

Recommended Posts

22 hours ago, Tido said:

Ubuntu's Icon set


URLs to a new icon in the code is preferred way.

 

1 hour ago, Tido said:

So, my PR only contains the readme.md


That's o.k. I'll check it later. Currently working on code fixes and I don't have more ports to attach devices. My home office has only 8 x 10GB and 42 x 1GB ports :P

 

 

 

Link to comment
Share on other sites

IS ANYBODY OUT THERE ???

I have improved the How To, Igor improved the Tool even more.  Will you please give it a chance and test your Single-Board-Computer (SBC)  - who doesn't like to test and tinker ;).  If you depend on your installation do a backup of your SDcard first   and then, please give it a try:  https://github.com/armbian/autotests

Report back here or on github if bad or good. If good even more, who doesn't like to hear it works well.

 

Thank you :beer:

 

Link to comment
Share on other sites

When people will realise, when we will know how to tell them that, that stability of their systems depends from the things like this, then perhaps someone will also follow and wire stuff together, perhaps find it interesting to deal with and do it instead of me ... :)

 

This is essentially a (cool) tool for us to quickly see how things looks from the users perspective. Now and for the future. With most boards at once. If anyone wants to have this look or help to creates that looks, one shell join. But I don't expect people staying in the line to help out. Even its is in their, ours interest.

Link to comment
Share on other sites

On 3/22/2020 at 8:53 PM, Igor said:

URLs to a new icon in the code is preferred way.

Which is making an internet connection necessary. Your current update including the icon is my preferred method. As mentioned before I would look for SVG as this scales smoothly.

I stopped looking for icons of SVG because of your comment before.  -   any changes on your side?

 

You have edited my suggestion (PR)  for the ReadME.md and removed the installation procedure I put there.

To simulate my initial situation on Ubuntu Mate 18.04.4 I did remove:

sudo apt-get remove sshpass iperf3 jq

The test is running twice, my root user credentials are given.

My phone is laying around bluetooth enabled, WiFi on the router is on  -  Rock Pi 4B.

I got  2x Syntax error   and 1x [ err. ] Wrong sources were installed  plus the missing software  `dpkg -l sshpass git iperf3 jq` 

Spoiler

$ ./go.sh
[ o.k. ] Host ###.###.###.7 found [ Run 1 out of 2 ]
[ o.k. ]   [ ###.###.###.7 ]
[ o.k. ] 0002-update-and-upgrade.bash [ 10:21:09 ]
[ o.k. ] Updating and upgrading [  packages ]
[ o.k. ] 0008-connect-wireless-devices-on-2.4Ghz.bash [ 10:21:09 ]
[ o.k. ] 0013-iperf-on-all-wired-interfaces.bash [ 10:21:12 ]
[ o.k. ] 0014-iperf-on-all-wireless-interfaces.bash [ 10:21:13 ]
[ o.k. ] 0015-connect-wireless-devices-on-5.0Ghz.bash [ 10:21:14 ]
[ o.k. ] 0017-iperf-on-all-wireless-interfaces.bash [ 10:21:14 ]
[ o.k. ] 0018-io-tests-memory.bash [ 10:21:15 ]
(standard_in) 1: syntax error
(standard_in) 1: syntax error
[ o.k. ] Max random memory throughput on 16Mb files [ Read:  MBits/s - Write:  MBits/s ]
[ o.k. ] 0019-io-tests-drive.bash [ 10:21:15 ]
(standard_in) 1: syntax error
(standard_in) 1: syntax error
[ o.k. ] Max random roofs throughput on 16Mb files [ Read:  MBits/s - Write:  MBits/s ]
[ o.k. ] 0111-bluetoth.bash 
[ o.k. ] 0115-strong-stressing.bash [ 10:21:15 + 18s ]
/media/Daten_1/01_Computer/01.05_GIT/autotests/tests/0115-strong-stressing.bash: Zeile 9: sshpass: Befehl nicht gefunden
[ o.k. ] 0116-dvfs.bash [ 10:21:15 ]
[ o.k. ] 9500-install-kernel-sources.bash 
[ err. ] Wrong sources were installed 
[ o.k. ] 9999-reboot.bash [ 10:21:15 ]
[ o.k. ] Rebooting in 3 seconds 
[ o.k. ] Host ###.###.###.7 found [ Run 2 out of 2 ]
[ o.k. ]   [ ###.###.###.7 ]
[ o.k. ] 0002-update-and-upgrade.bash [ 10:21:38 ]
[ o.k. ] Updating and upgrading [  packages ]
[ o.k. ] 0008-connect-wireless-devices-on-2.4Ghz.bash [ 10:21:38 ]
[ o.k. ] 0013-iperf-on-all-wired-interfaces.bash [ 10:21:41 ]
[ o.k. ] 0014-iperf-on-all-wireless-interfaces.bash [ 10:21:42 ]
[ o.k. ] 0015-connect-wireless-devices-on-5.0Ghz.bash [ 10:21:43 ]
[ o.k. ] 0017-iperf-on-all-wireless-interfaces.bash [ 10:21:43 ]
[ o.k. ] 0018-io-tests-memory.bash [ 10:21:44 ]
(standard_in) 1: syntax error
(standard_in) 1: syntax error
[ o.k. ] Max random memory throughput on 16Mb files [ Read:  MBits/s - Write:  MBits/s ]
[ o.k. ] 0019-io-tests-drive.bash [ 10:21:44 ]
(standard_in) 1: syntax error
(standard_in) 1: syntax error
[ o.k. ] Max random roofs throughput on 16Mb files [ Read:  MBits/s - Write:  MBits/s ]
[ o.k. ] 0111-bluetoth.bash 
[ o.k. ] 0115-strong-stressing.bash [ 10:21:44 + 18s ]
/media/Daten_1/01_Computer/01.05_GIT/autotests/tests/0115-strong-stressing.bash: Zeile 9: sshpass: Befehl nicht gefunden
[ o.k. ] 0116-dvfs.bash [ 10:21:44 ]
[ o.k. ] 9500-install-kernel-sources.bash 
[ err. ] Wrong sources were installed 
[ o.k. ] 9999-reboot.bash [ 10:21:44 ]
[ o.k. ] Rebooting in 3 seconds 
This whole procedure took 0 minutes.

 

 

Link to comment
Share on other sites

34 minutes ago, Tido said:

You have edited my suggestion (PR)  for the ReadME.md and removed the installation procedure I put there.

To simulate my initial situation on Ubuntu Mate 18.04.4 I did remove:

 

Yes, since its not needed. Script should take care of that. Well, a sudo apt install has to be added ... 

 

34 minutes ago, Tido said:

The test is running twice, my root user credentials are given.

My phone is laying around bluetooth enabled, WiFi on the router is on  -  Rock Pi 4B.

I got  2x Syntax error   and 1x [ err. ] Wrong sources were installed  plus the missing software  `dpkg -l sshpass git iperf3 jq` 


- syntax error has to be ironed out, working on it

- wrong sources is a correct response of a test case.

My tests are now automatically upload to https://dl.armbian.com/_test-reports/

Link to comment
Share on other sites

28 minutes ago, Igor said:

installing dependencies

there is no information given, that my system is downloading software. I hate these kind of installer like Firefox, TEAMS and so on.

autotests$ ./go.sh 
[sudo] Password for tido: 
Setup finished. Edit userconfig/configuration.sh and run ./go.sh again!
reto@X470-ULTRA:/media/Daten_1/01_Computer/01.05_GIT/autotests$ 

 

 

First run,  on line 127 it couldn't find something:

autotests$ ./go.sh 
./go.sh: Zeile 127: /media/Daten_1/01_Computer/01.05_GIT/autotests/reports/data.in: Datei oder Verzeichnis nicht gefunden
[ o.k. ] 0. RockPi-4B Linux 5.4.28-rockchip64 stable [ 11.13 ]
[ o.k. ] Host 11.13 found [ Run 1 out of 2 ]

 

I have configured 2 passes and it took 17 minutes - is this normal?

[ o.k. ] Rebooting in 3 seconds [ RockPi-4B ]
diff: /media/Daten_1/01_Computer/01.05_GIT/autotests/reports/data.in: Datei oder Verzeichnis nicht gefunden
This whole procedure took 17 minutes.

 

Link to comment
Share on other sites

1 hour ago, Tido said:

First run,  on line 127 it couldn't find something:


Will be fixed.

 

1 hour ago, Tido said:

I have configured 2 passes and it took 17 minutes - is this normal?

 

In my case, 21 devices, 2 passes = 357 minutes.

Link to comment
Share on other sites

22 hours ago, Igor said:

357 minutes.

Is it the download that makes it so long and if so, why does it need a download in it?
If it is only for you, fair enough - but if testers just want to test locally their devices   and  apart from that this generates quite some load on the servers and network traffic.

How many user switch the kernels or need the sources - I guess most don't.  So, if it is for a minorty only, why should everybody doing it.

Last but not least, you want many to use it and to help/report  -  it should be kept like a Mini (car)  "As little as possible, as much as required".

 

Just my 2 cents, for me it is obviously too much.   I like the idea to report back online to collect the data, but the  nightly/stable & sources  is just a waste of resources IMHO.

 

Link to comment
Share on other sites

20 minutes ago, Tido said:

Is it the download that makes it so long and if so, why does it need a download in it?


Sources install is the slowest process. Not just download, but also unpacking, lots of small files.

 

20 minutes ago, Tido said:

If it is only for you, fair enough - but if testers just want to test locally their devices   and  apart from that this generates quite some load on the servers and network traffic.

 

I plan to group tests later ... now just trying to make usable ones and create an engine. Since this is not designed for end users in first place, things like this does not matter.

 

20 minutes ago, Tido said:

How many user switch the kernels or need the sources - I guess most don't. 


This tools is for analysing the situation.

 

20 minutes ago, Tido said:

for me it is obviously too much.


For me, it tells very little at this point. Current tests are inside only to build a system around ... while several hundreds of testes should be the correct number. I doubt we will ever get there, but it has to be made possible.

 

https://www.toolsqa.com/software-testing-tutorial/

Link to comment
Share on other sites

On 4/1/2020 at 10:08 PM, Igor said:

to make usable ones

How long or how many times does it try to get the WiFi connection?

I am asking, because mine was not successful in both passes.

 

Link to comment
Share on other sites

44 minutes ago, Tido said:

How long or how many times does it try to get the WiFi connection?


Currently it does not retries. Only once. Will do two retries, otherwise error.

Link to comment
Share on other sites

3 hours ago, lanefu said:

That's pretty sweet!.  Are you using NFS images, or just swappign sdcards?

 

SD card swapping at this point. For that it would be ideally to use @Hijax invention. But already without, I see where things needs repair.

Link to comment
Share on other sites

@Tido et all,  here comes the files of the project I started but had to suspend.

Firstly,  as I by mistake did a short circut using the only board I have assembled.

Secondly - covid syndrome (the same number of tasks at work, twice the number of meetings and no workshop entry)

 

Anyway, latest files I have on my github. Zipped.

3 folders - 3 KiCad projects: source files, bom, other docs needed and so on.

 

Have a fun and stay healthy!

armbian-testing.zip

Link to comment
Share on other sites

8 hours ago, Hijax said:

KiCad projects: source files, bom, other docs needed

@TonyMac32 & @martinayotte 

I thought, there are some PCB manufacturer that do also the assembly, as long as your board uses parts they 'have'.

Do you already have experience with such an approach?

Could you support to replace/find another part if one isn't in the assemblers list?  (my knowledge is too little)

 

PS: I think the  __MACOSX   in the .zip is empty - it contains only some 212 bytes files.

PPS: I think this is what it should look like completed @Hijax  all PCBs are needed,  or is one just an improvement?  https://forum.armbian.com/topic/10841-the-testing-thread/?do=findComment&comment=83460

 

Edited by Tido
PPS:
Link to comment
Share on other sites

12 minutes ago, Tido said:

 

PS: I think the  __MACOSX   in the .zip is empty - it contains only some 212 bytes files.

 

 

Well, it is Mac OS resource "garbage"  you can delete it.

 

Link to comment
Share on other sites

19 hours ago, Hijax said:

Anyway, latest files I have on my github. Zipped.

3 folders - 3 KiCad projects: source files, bom, other docs needed and so on.


Do we have someone to build this together? I can ask around, just need to know.

Link to comment
Share on other sites

5 hours ago, Igor said:

Do we have someone to build this together?

This is not the big deal, but it would cost a fortune in my country just to get an empty PCB in such little quantity.

So, it was my idea to find a company in China as mentioned in this posting: https://forum.armbian.com/topic/10841-the-testing-thread/?do=findComment&comment=98628  to manufactur these.

 

If you have some contacts in your country or China to support this - would be good to ask.

 

Link to comment
Share on other sites

@Tido no idea what country is yours ;) but I can suggest ... use google for searching china pcb prototyping. I have recently used easyeda service, or jlcpcb. They manufacture 5 pcs (of each board, hence 3 times 5 pcs) for as little as 2USD plus delivery. Check their site. They can also do some basic assembly (I was not using that option yet, they can have some parts at their stock, anyway you may discuss the SMD soldering only as THT is simpe one)

 

Happens I have 4 set of boards (as I assembled one set only) and I can send those somewhere, to @Igor for example?

Zrzut ekranu 2020-04-6 o 14.41.12.png

Link to comment
Share on other sites

On 4/5/2020 at 10:23 PM, martinayotte said:

Right ! But never tried such services .

Okay, I gave it a try.  First I tried to improve the Bill-of-Materials form Hijax. I tried to identify the SMT parts (added a column in the attached file and put some text into it what I have changed).

Then I searched for the parts of the BOM here:   https://jlcpcb.com/parts

LED and Resistor I have identified based on Hijax given size and jlcpcb categories.  However, Multiplexer and such is way above my knowledge.  There are only 9-10 parts to check in total, I have done the ground work. Can you help please??

 

If JLCPCB doesn't fit, but we have the parts list ready (checked and amended from you)  we can try others like https://www.pcbway.com/pcb-assembly.html

 

 

Edit:  just thinking, would a Google doc be better to edit together?

Edited by Tido
removed the file, created a Google Doc instead.
Link to comment
Share on other sites

@Hijax, By second source I mean the "same" product from another manufacturer that has the same footprint like the one we already have, for example:

 

47219-2001 - Hinge Memory Socket, 47219 Series, Micro SD, 8 contacts, 5000 cycles https://sg.element14.com/molex/47219-2001/connector-micro-sd-8pos/dp/2614940
DM3CS-SF - Hinge Memory Socket, DM3, 8 contacts, https://ch.farnell.com/hirose-hrs/dm3cs-sf/steckverbinder-micro-sd-hinge/dp/1764376

460DE08C3 - Hinge Memory Socket, 460DE, 8 contacts, https://ch.farnell.com/multicomp/460de08c3/stecker-micro-sd-hinged/dp/1686452

502774-0891 - Push - Push Memory Socket, 502774, Micro SD, 8 contacts, https://ch.farnell.com/molex/502774-0891/steckverbinder-microsd-8pos/dp/2064063

DM3D-SF -  Push - Push Memory Socket, DM3, 8 contacts, https://ch.farnell.com/hirose-hrs/dm3d-sf/steckverbinder-micro-sd-push-pull/dp/1764377

 

Why did you choose  the one with a hinge   and not:

Hirose DM3CS (Hinge, Push-Pull, manual, without ejection mechanism)

Hirose DM3AT and DM3BT (Push - Push, with ejection mechanism)

Hirose DM3D (Push -Pull, manual, without ejection mechanism)

 

Link to comment
Share on other sites

×
×
  • Create New...

Important Information

Terms of Use - Privacy Policy - Guidelines