I'm trying to run a script from Teamcity Agent on other environment (mac).
I am using script that contains 'ssh administrator#33.333.33.33..'
when running the script as administrator on the Teamcity agent environment it works (after I used the publicKey). when running the Teamcity project I'm getting
[Step 1/1] Permission denied (publickey,keyboard-interactive).
[00:18:15][Step 1/1] Process exited with code 255
[00:18:15][Step 1/1] Step run protractor test (Command Line) failed
I've tried also adding the ssh key file to the project as seen here
https://confluence.jetbrains.com/display/TCD8/SSH+Keys+Management
I'm still getting the error. I assume that maybe Teamcity Agent uses other user (than Administrator) to run the script.
any ideas?
This may be of use, but is potentially not a "TeamCity" issue
Permission denied (publickey,keyboard-interactive)
Script is run using the account which is running Teamcity agent service. You need to make sure this account has enough rights to run the script.
Related
Trying to run my postman collection in azure devops inside a self-hosted agent. When I try to run the command inside the agent "newman run postman_collection.json -e postman_environment.json -r cli,htmlextra" it's running fine. But when I run the same through a a command line script task in release pipeline it's throwing the error "newman is not recognized..". I also tried to have a npm task for newman installation i.e. "npm install -g newman" it's also throwing the erro "##[error]Unable to locate executable file: 'newman'. Please verify either the file path exists or the file can be found within a d...."
azure devop selfhosted agent, newman command not recognized
According to the error message "##[error]Unable to locate executable file: 'newman" when you using the npm install -g newman, you could try to add C:\Users\[BUILDSERVER-USERNAME]\AppData\Roaming\npm to the PATH variable for the [BUILDSERVER-USERNAME] user.
You could refer to this document How to fix the Newman task for Team Foundation Server silently failing for some more details.
Besides, when we use command line to install the newman, it will take a few minutes to install it, so we need to wait for a few minutes before we using the command line:
"newman run postman_collection.json -e postman_environment.json -r cli,htmlextra"
You could add powershell task to sleep a few minutes：
echo "Sleeping for 10 mins..."
Start-Sleep -s 600
I have a build in Gitlab CI that takes a long time (10mins+) to run, and it's very annoying to wait for the entire process every time I need to experiment / make changes. It seems like surely there's a way to access some sort of shell during the build process and run commands interactively instead of placing them all in a deploy script.
I know it's possible to run Gitlab CI tests locally, but I can't seem to find a way to access the deploy running in process, even after scouring the docs.
Am I out of luck or is there a way to manually control this lengthy build?
I have not found a clean way for now, but here is how I do it
I start building locally gitlab-runner exec docker your_build_name
I kill gitlab-runner using control + c right after the docker image to be built. You can still add the command sleep 1m as the first script line just to have time enough to kill gitlab-runner
Note: gitlab-runner will create a docker and then delete it once the job is done… killing it will ensure the docker is still there - no other alternative I know for now….
Manually log into the container docker exec -i -t <instance-id/tag-name> bash
Run your script commands manually…
I'm attempting to use vsts-npm-auth to get the authentication token for our VSTS package repository. On my development machine I can run the commands
npm install -g vsts-npm-auth
vsts-npm-auth -config path-to-my\.npmrc
and it succeeds in providing me with an authentication token. I'm now trying to recreate this as a build step on VSTS, so I create the powershell script auth-vsts.ps1
$npmrcFile = "$PSScriptRoot\path-to-my\.npmrc";
npm install -g vsts-npm-auth;
vsts-npm-auth -config $npmrcFile;
and add it as a powershell task. However, the task fails as follows
2017-05-30T09:37:41.1082686Z ##[section]Starting: auth-vsts
2017-05-30T09:37:41.1092712Z ==============================================================================
2017-05-30T09:37:41.1092712Z Task : PowerShell
2017-05-30T09:37:41.1092712Z Description : Run a PowerShell script
2017-05-30T09:37:41.1092712Z Version : 1.2.3
2017-05-30T09:37:41.1092712Z Author : Microsoft Corporation
2017-05-30T09:37:41.1092712Z Help : [More Information](https://go.microsoft.com/fwlink/?LinkID=613736)
2017-05-30T09:37:41.1092712Z ==============================================================================
2017-05-30T09:37:41.1112679Z ##[command]. 'd:\a\1\s\auth-vsts.ps1'
2017-05-30T09:37:47.3792461Z C:\NPM\Modules\vsts-npm-auth -> C:\NPM\Modules\node_modules\vsts-npm-auth\bin\vsts-npm-auth.exe
2017-05-30T09:37:47.3792461Z C:\NPM\Modules
2017-05-30T09:37:47.3802239Z `-- vsts-npm-auth#0.25.0
2017-05-30T09:37:47.3802239Z
2017-05-30T09:37:47.3802239Z
2017-05-30T09:37:47.3802239Z vsts-npm-auth v0.25.0.0
2017-05-30T09:37:47.3802239Z -----------------------
2017-05-30T09:37:47.3802239Z Creating npmrcFile. Path: D:\a\1\s\.npmrc
2017-05-30T09:37:47.3802239Z Getting new credentials for source:https://our-domain/_packaging/SharedLib/npm/registry/, scope:vso.packaging_write vso.drop_write
2017-05-30T09:37:49.8729702Z Caught exception: The prompt option is invalid because the process is not interactive.
2017-05-30T09:37:49.8729702Z Parameter name: PromptType
2017-05-30T09:37:49.8729702Z Caught exception: The prompt option is invalid because the process is not interactive.
2017-05-30T09:37:49.8729702Z Parameter name: PromptType
2017-05-30T09:37:49.8729702Z Couldn't get an authentication token for //our-domain/_packaging/SharedLib/npm/registry/:_authToken.
2017-05-30T09:37:50.1769711Z ##[error]Process completed with exit code 1.
2017-05-30T09:37:50.1809715Z ##[section]Finishing: auth-vsts
The error gives no indication as to why it can't obtain the credentials. Any ideas why this might be?
I faced this issue while trying to execute via Visual Studio Code`s powershell terminal
vsts-npm-auth -config .npmrc
But running the same command via simple console solved this issue and I was redirected to authentication window.
Can suggest that due to internal limitations powershell disabled to open another windows.
The error did indicate why it cannot obtain the credentials:
The prompt option is invalid because the process is not interactive.
This could be caused by the build agent does not run in interactive mode which make the credential dialog cannot be prompted. If you are using Hosted Build Agent, the build agent is run as service and there isn't any way to change to interactive mode.
However, the issue here is that if you want to use the feed in a build step, it does not make sense to prompt a credential dialog during the build process since the build step cannot enter the required credential automatically. Not sure if there is any specific requirement in your environment, but the general workflow should be uploading the .npmrc file generated in your local machine to the Source Control so that npm can use the auth token in the file to install/publish packages to VSTS Feed.
Inside your project, you can open a terminal and run
vsts-npm-auth -F -C .npmrc
This script refreshes the npm token. Here I set two parameters: -F forces the refresh (if not set, the token is refreshed only if it is already expired), while -C fileName defines the configuration file.
The vsts authentication system sometimes authenticates the use by popping up a browser window. If the terminal you're running the command from is not interactive (e.g., ssh terminal, vscode terminal) it won't be able to pop up that window, and the authentication will fail.
This worked for me
npx vsts-npm-auth -config .npmrc
I have a chef recipe that runs a chocolatey install for Microsoft SQL server. From my workbench when I run
knife winrm [IP] 'chef-client -o "recipe[NetDevMachine::default]"' -m -x 'domain\myuser'
over a node it fails with the error:
532459699 or sometimes 2022834173 or 2057043966
However if I log onto the VM as the same user and locally run
chef-client -o "recipe[NetDevMachine::default]"
It works. Does anyone know what the difference is between running chef-client locally and running it remotely from the workbench? What does chef-client do differently here? Both are pulling the recipe from the same chef-server repo.
Additional Details
I am using the same user for both
I have successfully run other recipes from the workstation, it's just this MicrosoftSQLServer install that's not working
running "knife winrm [IP] 'choco install MicrosoftSQLServer' -m -x 'domain\myuser' also does not work
Recipe contents:
powershell_script "choco install MicrosoftSQLServer2014" do
code <<-EOH
choco install MicrosoftSQLServer2014
EOH
end
Error:
[ERROR] Running C:\Users\myuser\AppData\Local\Temp\MicrosoftSQLServer2014\setup.exe with
/QUIET /IACCEPTSQLSERVERLICENSETERMS /ACTION=INSTALL /INSTANCENAME=MSSQLSERVER /FEATURES=SQL,TOOLS,LOCALDB
/SQLSVCACCOUNT="NT AUTHORITY\Network Service" /SQLSYSADMINACCOUNTS=BUILTIN\ADMINISTRATORS
/SKIPRULES=REBOOTREQUIREDCHECK UIMODE=AUTOADVANCE was not successful.
Exit code was '-532459699'.
Edit - the only difference I can see is that one runs in the foreground and the other doesn't. So the location of the compile chef-script is /temp/2 for the failing run. I don't know if there is a way to force chef to run in the foreground or if that will even help.
Thanks
I never got to the bottom of this and still don't understand how running the chef client remotely using knife was different from running chef client directly on the VM. Perhaps it was something to do with the handling of credentials or permissions when remotely sending commands.
But I did find that removing chocolatey and replacing it a command line silent install inside a chef powershell resource allowed me to install MSSQL.
I met this problem when I was trying to config Jenkins to fetch the repository on the server of a fellow(university course).
Started by user anonymous
Building in workspace /var/lib/jenkins/jobs/TEST1/workspace
$ hg clone --rev default --noupdate ssh://XXX#partch.anu.edu.au/XXX/XXX /var/lib/jenkins/jobs/TEST1/workspace
remote: Host key verification failed.
abort: no suitable response from remote hg!
ERROR: Failed to clone ssh://XXX#partch.anu.edu.au/XXX/XXX
ERROR: Failed to clone ssh://XXX#partch.anu.edu.au/XXX/XXX
Finished: FAILURE
Then I tried to do the same command in terminal, with the different error
abort: Permission denied: /var/lib/jenkins/jobs/TEST1/workspace
I tried to change the location for the command:
hg clone --rev default --noupdate ssh://XXX#partch.anu.edu.au/XXX/XXX /home/administrator/TEST
And it works
According to all these, and the fact that this is no workspace(Error: no workspace)--I know it's because of no successful build ever.
Is it a problem with the workspace configuration of Jenkins? but don't know what to do.
Thank you very much if anyone gives me some advice!
Your error doesn't mention anything about a workspace. There error clearly states that it comes from remote hg and it is an authentication failure.
Also, are you using the Mercurial Plugin to perform the SCM checkout or trying to replicate it with a Build Shell build step?