I have problem printing script output when using rich https://github.com/willmcgugan/rich for some reason, i can't print script output when script is called from script-server in terminal everything is fine

there is no visible output in script server

if i use python3 print function i am getting this output <rich.table.Table object at 0x7f3b2b488400>

This is log file (logs/processes) [email protected]:/home/srdjan/script-server/logs/processes# cat GPON__ONT_RFO_LOSS_Report_192.168.128.132_200724_143245.log id:2460 user_name:192.168.128.132 user_id:192.168.128.141-04f1204895df4625 script:GPON _ONT RFO LOSS Report start_time:1595601165629 command:/home/srdjan/script-server/conf/runners/rich_ma4k_ltp_high_loss_detailed.py 24 exit_code:0

OUTPUT STARTED <<<<<

From what I can tell, this is near the perfect solution for script interfacing for nontechnical users.

The ability to schedule scripts is the only feature I feel is missing. Ideally, being able to assign specific dates, times, and/or cadences would round out this as a solution.

Sorry if this is already feature, but no mention of it in the wiki/docs.

Hello, so I set this up in docker today but its not working. So if I go to scripts.mydomain.com I can see the webui, but no scripts. If I go to scripts.mydomain.com/admin.html it says 403 no authorization.

Docker Config - https://hastebin.com/focevemota.bash - ignore first five Traefik Config - https://hastebin.com/epoyugekoc.bash Contents of Local Folder - https://hastebin.com/hakiwomafo.rb Script-Server Config - https://hastebin.com/koxigezihu.pl Runner Config - https://hastebin.com/fobiliweyi.json Script Contents - https://hastebin.com/afamitexej.coffeescript Script-Server Logs - https://hastebin.com/pezakeyaqi.cs

because when there are too many it becomes pretty complicated to remember numbers

may be there is such an option?

(i can live without it... just a thought) thanks

I have auth-proxy configuration, everything works fine but in the logs the variable USERNAME is None instead of the value in my user_header_name

conf.json

{
    "port": 5000,
    "address": "0.0.0.0",
    "title": "My Script Server",
    "max_request_size": 100,
    "access": {
	"allowed_users": "*",
	"admin_users": [ "USERA", "USERB", "USERC" ],
	"user_header_name": "X-Forwarded-User",
	"trusted_ips": "*"
    },
    "logging": {
	"execution_file": "${USERNAME}.log",
	"execution_date_format": "%y-%m-%d-%H:%M"
    }
}

There should be a possibility to add an "include" section to a script configuration, which would add/fill parameters configuration from another file. This "include" section can depend on current parameter values, so it will be loaded dynamically while user selects values on UI

There are 2 use cases behind:

1. Sharing parameters in different configurations
2. Depending on some parameter, we could include different corresponding additional parameters

For more details and reasoning please read comments in #119

Dockerfile for docker image build. Tested and working as expected. Users should mount config files and scripts into docker image. conf.json file ,runners folder , scripts folder.

Example usage:

docker run -d -p 5000:5000 \
           -v /host/path/to/conf.json:/app/conf/conf.json \
           -v /host/path/to/runners/:/app/conf/runners \
           -v /host/path/to/scripts/:/app/scripts/ \
           --name script-server \
           scriptserver0/script-server:latest

Example runner:

$ cat /host/path/to/runners/sample.json
{
  "name": "sample script",
  "description": "sample script description",
  "script_path": "/app/scripts/sample.sh"
}

Example script:

$ cat /host/path/to/scripts/sample.sh
#!/bin/bash
echo "Hello sample script"

Notes: Script-server users should know it cannot access host system processes as it's nature of docker container. If script-server must has access to host system processes then docker image is not suited for this kind of task. (But not saying it's impossible)

It would be nice to have docker image ready.

I'll try to build docker image for the project and make it build automatically for every commit that passes test (do we have one?).

We've production ready drone setup that we use daily. If you don't mind I'd use that.

In my docker setup, each container get an random IP when it starts. It would require me to do a lot of modifications to be able to give my nginx container a static IP so the trusted ip work.

I suggest 2 options

1. Allow to use * as the trusted ip, meaning any IP will be ok
2. Allow to use a subnet instead of a specific IP like 178.18.0.0/24

Currently, every time my nginx container get a new IP I have to edit the configuration file to add the new ip...

My goal is to have a multi-user server where users can login with their own username and passwords, and only have access to the scripts I allow them.

I have setup script-server behind a nginx docker, the only way to reach the script-server is via the nginx proxy, and the only public ports in my server are 80 and 443.

I created a htpasswd for my user, up to here everything is fine. But there are 2 issues:

I need to put my internal docker IP in trusted IPs, and it's not really static, while this could be fixed, it would be better if trusted IPs allowed to set a range... like 172.18.0.0/24

Biggest issue is all I see as identified as user is this:

2019-08-29 09:06:37,481 [identification.INFO] Assigned user_id=IP-46a12c6f2ff7496e to {'proxied_ip': 'IP', 'proxied_hostname': 'c95000b4.redacted.com.br', 'ip': '172.18.0.2', 'hostname': 'nginx-proxy.cloudbox'}

Obviously none of my users will have static IPs so this won't work! Am I missing something here?

I have this simple script that pull new changes from repo and build docker image & push.

#!/bin/bash
# simple docker build script

now=$(date +'%H:%M:%S')
echo -e "\e[1mStep1 $now: Changing directory to dependencies\e[0m"
cd /home/shinee/web/service/dependencies-base

now=$(date +'%H:%M:%S')
echo -e "\e[1mStep2 $now: Pulling update from master\e[0m"
git pull origin master

now=$(date +'%H:%M:%S')
echo -e "\e[1mStep3 $now: Updating image node:8-slim\e[0m"
docker pull node:8-slim

now=$(date +'%H:%M:%S')
echo -e "\e[1mStep4 $now: Updating image nginx:stable\e[0m"
docker pull nginx:stable

now=$(date +'%H:%M:%S')
echo -e "\e[1mStep5 $now: Building image myrepo/image:tag\e[0m"
#docker build --force-rm -t myrepo/image:tag -f Dockerfile . || { echo "couldn't build docker image" ; exit 1; }

now=$(date +'%H:%M:%S')
echo -e "\e[1mStep6 $now: Pushing image myrepo/image:tag\e[0m"
#docker push myrepo/image:tag

I intentionally commented out docker build & push part for the sake of testing. But bug is still reproducable.

To reproduce: If you run this script from command line. ./test_build.sh only line that contains echo -e bash formatted lines will be bold, and commands outputs will be normal text.

But this same script's output doesn't look as expected in script-server window. As you see

echo -e "\e[1mStep2 $now : Pulling update from master\e[0m"    # this part should be formatted only
git pull origin master     # but this commands output are also getting formatted, sometimes. (most of the time)

I have three versions of script-server latest 1.17.1, dev and master.

First it would be cool if user/admin page would display the version.

But reason for raising this issue is to ask how to use --lws parameter to make "list with search" ... if I use it in -master as in the "Very parameterized" example it works fine, but when I try to add this to 1.17.1 the --lws parameter becomes a command line argument to the script.

All other options are identical.

Many thx, R.

Hi, I would like to create the menu items at multi hierarchy level. Currently it supports one level only. Since I have multiple scripts, facing difficulties in grouping them.

Hi, Currently, On UI, the search has the ability to search within the group names only. It is unable to search the scripts within a group.

I have created a lot of scripts in each group, and finding the specific script became a challenge to me.

Hi,

Not 100% sure if my configuration is valid, however it seems to work find as long as I don't select anything on the multiselect :)

The main configuration file is:

{
  "name": "test",
  "script_path": "ls",
  "output_format": "terminal",
  "include": "test.inc",
  "parameters": [
    {
      "name": "dir",
      "required": true,
      "default": "/tmp",
      "constant": true
    }
  ]
}

The included file is:

{
  "hidden": true,
  "parameters": [
    {
      "name": "Files",
      "type": "multiselect",
      "values": {
        "script": "ls ${dir}"
      },
      "multiselect_argument_type": "argument_per_value"
    }
  ]
}

The multiselect is populated, and works as expected. However, selecting an option results in an error. For example, selecting the item call results in:

Calling the script...

Invalid parameter Files value: has value 'call', but should be in []

And the same in the logs:

2022-12-16 11:46:55,295 [web_server.ERROR] Invalid parameter Files value: has value 'call', but should be in []
2022-12-16 11:46:55,296 [tornado.access.WARNING] 400 POST /executions/start (192.168.1.55) 15.99ms

Is this configuration invalid, or am I running into a bug?

Regards,

Mark

Hi,

I am currently experiencing issues downloading output files. Sometimes the download button appears, but most of the time it is not visible.

My config file looks like this;

Everytime I execute the script, the correct file is placed in temporary directory (resultFiles) on our server running script-server. It seems to detect the correct file(s), but the download button is not shown at the bottom of the script output.

The last ten lines in the server log are as follows; 2022-12-01 15:52:56,907 [script_server.execution_service.INFO] Calling script #498: bash /mnt/shared/tools/script-server/GS/run_ffpe_qc.sh 105285 Human forward 8 2022-12-01 15:52:56,917 [script_server.file_download_feature.INFO] Created download folder for s.wielhouwer: temp/resultFiles/18e344481b61/1669906376917 2022-12-01 15:52:56,919 [tornado.access.INFO] 200 POST /executions/start (10.159.1.224) 33.86ms 2022-12-01 15:52:56,928 [tornado.access.INFO] 101 GET /executions/io/498 (10.159.1.224) 1.25ms 2022-12-01 15:52:56,930 [script_server.audit_utils.WARNING] Could not get hostname for 10.159.1.224 2022-12-01 15:52:56,931 [script_server.audit_utils.WARNING] Could not get hostname for 10.159.1.224 2022-12-01 15:52:57,031 [script_server.audit_utils.WARNING] Could not get hostname for 10.159.1.224 2022-12-01 15:52:57,032 [web_server.INFO] s.wielhouwer disconnected 2022-12-01 15:52:57,036 [script_server.audit_utils.WARNING] Could not get hostname for 10.159.1.224 2022-12-01 15:52:57,037 [tornado.access.INFO] 200 POST /executions/cleanup/498 (10.159.1.224) 2.13ms

Any ideas what could possibly cause this? Is this related to the hostname error?

Best regards,

Serge

Hello there!

My name is Ana. I noted that you use the mutation testing tool in the project. I am a postdoctoral researcher at the University of Seville (Spain), and my colleagues and I are studying how mutation testing tools are used in practice. With this aim in mind, we have analysed over 3,500 public GitHub repositories using mutation testing tools, including yours! This work has recently been published in a journal paper available at https://link.springer.com/content/pdf/10.1007/s10664-022-10177-8.pdf.

To complete this study, we are asking for your help to understand better how mutation testing is used in practice, please! We would be extremely grateful if you could contribute to this study by answering a brief survey of 21 simple questions (no more than 6 minutes). This is the link to the questionnaire https://forms.gle/FvXNrimWAsJYC1zB9.

Drop me an e-mail if you have any questions or comments ([email protected]). Thank you very much in advance!!