Navigation

    Inedo Community Forums

    Forums

    • Login
    • Search
    • Categories
    • Recent
    • Tags
    • Popular
    • Users
    • Groups
    1. Home
    2. dwynn_6489
    D
    • Profile
    • Following
    • Followers
    • Topics
    • Posts
    • Best
    • Groups

    dwynn_6489

    @dwynn_6489

    1
    Reputation
    46
    Posts
    15
    Profile views
    0
    Followers
    0
    Following
    Joined Last Online

    dwynn_6489 Follow

    Best posts made by dwynn_6489

    • RE: Halting a pipeline stage without using fail

      Warn will still automatically advance.

      I was able to utilize the Set-ReleaseVariable as you suggested. Thanks!

      posted in Support
      D
      dwynn_6489

    Latest posts made by dwynn_6489

    • RE: Halting a pipeline stage without using fail

      Warn will still automatically advance.

      I was able to utilize the Set-ReleaseVariable as you suggested. Thanks!

      posted in Support
      D
      dwynn_6489
    • RE: Halting a pipeline stage without using fail

      Hi John - thanks for the response. Unfortunately, forcing normal will allow the pipeline to advance to the Deploy stage, which defeats the validation of build criteria (I don't think I made that clear in my original post).

      More clearly stated goal: halting a pipeline without using "fail"

      more verbose-ly stated goal: we want to be able to differentiate between known issues that should halt a pipeline (i.e. validation of build criteria) and unknown errors that halt a pipeline (i.e. coding errors). Event listeners only know that a pipeline stage failed, not why it failed, so they catch every failure.

      Maybe what we are looking for is the ability to control the advancement to the next pipeline stage?

      posted in Support
      D
      dwynn_6489
    • Halting a pipeline stage without using fail

      Is there a way to halt a pipeline stage without using the "fail" keyword?

      We have a 2 stage pipeline (Build and Deploy). During the all stages, we monitor for programming errors using Event Listeners that trigger on failed executions and send email to operations to review the error (things like variable resolution or syntax errors).

      There are also several criteria we look for in the Build stage that should cause it to halt - things like missing build values, other concurrent executions, etc. If any of these criteria are met, we alert the necessary parties and use "fail" to halt the stage. However, using fail also triggers the event listener and sends an email to operations that they don't need.

      The desired logic in the plan would look something like

      call CheckBuildCriteria
      (
          validBuild => $IsValidBuild
      );
      
      if $IsValidBuild != true
      {
          Halt-Stage; # currently we have to use "fail" here.
      }
      

      Product: BuildMaster
      Version: 6.1.5

      posted in Support
      D
      dwynn_6489
    • Comments in plans that don't display in the execution log

      Is there a notation for comments in plans that will prevent them from showing up in the execution log?

      Product: BuildMaster
      Version: 6.1.5

      posted in Support
      D
      dwynn_6489
    • RE: SVN Repository Monitor won't trigger

      What other information can I provide? Right now the feature is not functional.

      Side note - the "Plan" option when configuring a monitor only shows global plans, no application scoped plans.

      Thanks.

      posted in Support
      D
      dwynn_6489
    • SVN Repository Monitor won't trigger

      I have created an SVN repository monitor, but I can't tell that it is doing anything. I want this monitor to trigger a build/deployment for a release in the application to which I associated the monitor, so when I created the monitor I entered the name of the desired plan and selected the application. Where would I see the results of the execution?

      I see this in the Service Log, so it appears monitor is triggering.

      Running Repository Monitor...
      Fetching repository monitors and last recorded states...
      Found 1 active repository monitors.
      Repository Monitor completed.

      When viewing the Monitor on screen, in the "States" column it says "no branches recorded".

      I'm using a Resource Credential. Is it possible it isn't configured correctly, and is there a way to tell?

      This is what I see in the RepositoryMonitors table

      RepositoryMonitor_Id = 1
      RepositoryMonitor_Name = sq_test
      Application_Id = 1175
      ApplicationGroup_Id = NULL
      RepositoryMonitor_Configuration (shown below)
      CronExpression_Text = 0 0/1 * * * ?
      Active_Indicator = Y
      ExecuteOn_Server_Id = NULL
      Plan_Name = sq_dummy2
      BranchFilter_Text = integration

      <Inedo.Extensions.Subversion.RepositoryMonitors.SvnRepositoryMonitor Assembly="Subversion">
      <Properties CredentialName="sq_test" />
      </Inedo.Extensions.Subversion.RepositoryMonitors.SvnRepositoryMonitor>

      Thanks.

      Product: BuildMaster
      Version: 6.1.5

      posted in Support
      D
      dwynn_6489
    • Powershell array parameter with default value blocks subsequent parameters

      In the powershell code shown below, the value for param3 is not received from the plan when the release is executed. Is this a powershell limitation, or something to do with how BuildMaster is parsing the script? If I remove $arr_param, param3 is properly handled.

      When the plan is viewed in visual mode, param3 doesn't show up when I edit the test_PS_script Operation. Including a value for arr_param in the plan doesn't change the behavior (although the script does receive the passed in value for arr_param).

      SCRIPTASSET - test_PS_script

      Param (
      $param1,
      $param2,
      $arr_param = @(),
      $param3
      )

      Write-Output "param1: ${param1}"
      Write-Output "param2: ${param2}"
      Write-Output "arr_param.Count: $($arr_param.count)"
      Write-Output "param3: ${param3}"

      PLAN

      PSCall test_PS_script
      (
          param1: val1,
          param2: val2,
          param3: val3
      );
      

      OUTPUT

      Found script test_PS_script.ps1 in Default raft.
      Using LocalAgent agent on TESTSERVER01
      Importing param3...
      Assigning parameter param1...
      Assigning parameter param2...
      param1: val1
      param2: val2
      arr_param.Count: 0
      param3:

      posted in Support
      D
      dwynn_6489
    • Repository Monitor for Subversion?

      BuildMaster has an extension for Subversion, but I can only find repository monitors for Git. Am I missing something, or is this a design decision? If so, do you have plans for Subversion repository monitors?

      Thanks.

      Product: BuildMaster
      Version: 6.1.4

      posted in Support
      D
      dwynn_6489
    • RE: Legacy Build Triggers

      We use Schedule Triggers for a variety of jobs - some system level (cleanup, etc) but also to run recurring deployments for a variety of development and testing purposes. They are an important part of our day to day operations.

      I'd be happy to provide more information about our use-cases if it would help.

      posted in Support
      D
      dwynn_6489
    • RE: Where are release package triggers in 6.1.0?

      Thanks Tod. That fixes the visibility.

      Our triggers are Schedule Triggers, not SCM based. What are Schedule Triggers replaced with?

      posted in Support
      D
      dwynn_6489