stf icon indicating copy to clipboard operation
stf copied to clipboard

Question regarding stf deployment using stf commands instead of docker

Open xiaweixing opened this issue 9 years ago • 9 comments

What's the problem (or question)?

I want to deploy STF on multiple Ubuntu machines on which I could not use docker solution. I have to use stf local, but need detailed instruction on how to do it. I want to use box1 (10.241.54.10) as the master box, and connect box2(10.241.54.16) to box1, so from http://10.241.54.10:7100/#!/devices, I can see all the devices connected to both box1 and box2.

What should have happened?

My steps/cmds to run stf: on box1:

start rethinkdb
stf local --allow-remote --public-ip 10.241.54.10 --provider 10.241.54.10 --provider-min-port 15000 --provider-max-port 20000

on box2:

start rethinkdb
stf local --allow-remote --public-ip 10.241.54.16 --provider 10.241.54.16 --provider-min-port 20000 --provider-max-port 25000

Then I can successfully see devices showing up from http://10.241.54.10:7100/#!/devices and http://10.241.54.16:7100/#!/devices Above steps mean my stf env for both boxes are correct.

I tried to start stf local on box1:

stf local --allow-remote --public-ip 10.241.54.10 --provider 10.241.54.10 --provider-min-port 15000 --provider-max-port 20000

I checked the processes started by stf local and get the provider one:

test     16054 16018  0 Aug30 pts/15   00:01:22 /home/test/.linuxbrew/Cellar/node/6.1.0/bin/node /home/test/.linuxbrew/lib/node_modules/stf/lib/cli.js provider --name 10.241.54.10 --min-port 15000 --max-port 20000 --connect-sub tcp://127.0.0.1:7114 --connect-push tcp://127.0.0.1:7116 --group-timeout 900 --public-ip 10.241.54.10 --storage-url http://localhost:7100/ --adb-host 127.0.0.1 --adb-port 5037 --vnc-initial-size 600x800 --allow-remote

then tried to start stf provider on box2(with same port as above):

stf provider --allow-remote --name 10.241.54.16 --min-port 20000 --max-port 25000 --connect-sub tcp://10.241.54.10:7114 --connect-push tcp://10.241.54.10:7116 --storage-url http://10.241.54.10

but it shows: INF/provider 11938 [*] Providing all 0 of 5 device(s); ignoring "0123456789ABCDEF", "284cd8fe", "639aa439", "YLHU657DMRG6NNN7", "7TTKM78PNFS8SW7D" I cannot see devices connected to box2 from http://10.241.54.16:7100/#!/devices

Do you have an idea for a solution?

How can we reproduce the issue?

Help us understand your issue by providing context.

So far, docker solution does not seem to be possible for me to go, so I want to achieve production-like deployment with stf command manually.

Please tell us details about your environment.

  • Are you using stf local or a full deployment: stf local
  • Version used (git hash or stf -V): 2.0.1
  • Environment name and version (e.g. Chrome 39, node.js 5.4): node 6.1.0
  • Operating System and version: Ubuntu Server 14.04
  • If there was a problem with a specific device, run adb devices -l and paste the relevant row here:

xiaweixing avatar Aug 31 '16 04:08 xiaweixing

Do you solve the problem?The same as me.

zangtian2 avatar Oct 26 '16 03:10 zangtian2

No, I finally switched to docker solution. Suggest you do the same :)

xiaweixing avatar Oct 26 '16 03:10 xiaweixing

yeah,i had done it. However,a new problem.I can see device ,but the device always disapeared and disconnected when i click the device icon.Then it shows "the devices is disconnected". Form adb devices,the device is inline and it's ok. So,could you tell me how you deploy stf on multiple Ubuntu machines? Thank you so much. I had tried many smartphones,make no difference.

zangtian2 avatar Oct 27 '16 12:10 zangtian2

just follow the deployment guide. First download docker images on multiple machines, then select one as the main host, create almost all .service file accordingly on it; on other hosts, only need to start provider service.

xiaweixing avatar Oct 28 '16 02:10 xiaweixing

Guys, can you share your experience under my post, i need help.

https://github.com/openstf/stf/issues/493

cemakkok avatar Dec 01 '16 09:12 cemakkok

Just stop wasting your time and go with the docker deployment. It's definitely possible to run the commands by yourself without docker, but it would be extremely time consuming to help you with that, so we will not. You can try to take a look at the deployment guide and steal/modify the commands from there.

sorccu avatar Dec 01 '16 09:12 sorccu

I use "--connect-sub tcp://127.0.0.1:7114 --connect-push tcp://127.0.0.1:7116" instead of "--connect-sub tcp://publicip:7114 --connect-push tcp://publicip:7116" in the "stf provider" command and the problem is solved.

shuoyoung avatar Feb 28 '18 07:02 shuoyoung

> 我使用“ --connect-sub tcp://127.0.0.1:7114 --connect-push tcp://127.0.0.1:7116”,而不是“ --connect-sub tcp:// publicip:7114 --connect在“ stf provider”命令中输入-push tcp:// publicip:7116即可解决问题。

我使用命令:stf provider --name localhost.localdomain --min-port 7400 --max-port 7700 --connect-sub tcp://172.50.1.159:7114 --connect-push tcp://172.50.1.159:7116 --group-timeout 900 --public-ip 172.50.1.159 --storage-url http://172.50.1.159:7100/ --adb-host 172.17.244.11 --adb-port 5037 --vnc-initial-size 600x800 --mute-master never --allow-remote

然后就一直停在Receiving input from "tcp://172.50.1.159:7114",过了几分钟就报错了,这个是什么原因 2021-05-26_105530

yuyangcq avatar May 26 '21 04:05 yuyangcq

我使用命令:stf provider --name localhost.localdomain --min-port 7400 --max-port 7700 --connect-sub tcp://172.50.1.159:7114 --connect-push tcp://172.50.1.159 :7116 --group-timeout 900 --public-ip 172.50.1.159 --storage-url http://172.50.1.159:7100/ --adb-host 172.17.244.11 --adb-port 5037 --vnc-initial尺寸600x800-静音大师-不允许远程

然后就一直停在从“ tcp://172.50.1.159:7114”接收输入,过了几分钟就报错了,这个是什么原因 2021-05-26_104729

yuyangcq avatar May 26 '21 04:05 yuyangcq