Comments
Sort by recent activity
Thank you, @Monday. It took me a while to find time to dig into this. I had never looked into the actual files within the nupkg file. Now, thanks to you, I do see the db\output\PackageScript.sql. I also see how it has ALL of the migrations including baseline and will intelligently apply them to whatever DB I set in the sqlcmd var DatabaseName. There are a couple of issues with this file (in its raw form) that I will need to resolve to be able to pass it to the ops DBAs so they can execute in SSMS via SQLCMD mode: 1) I'll need to uncomment the sqlcmd vars section 2) I'll need to manually set the ReleaseVersion and PackageVersion. Not sure why the PackageVersion isn't set by New-DatabaseBuildArtifact cmdlet since we are passing that in. Probably a lil powershell will take care of the above concerns, but it would be nice to have options during build to extract this file with these taken care of. UPDATE: I am also able to get the PackageScript.sq content from the PackageScript property of the BuildArtifact object returned from Invoke-DatabaseBuild. This way I can skip the Export-DatabaseBuild step, and just save this to a file and process it. / comments
Thank you, @Monday. It took me a while to find time to dig into this. I had never looked into the actual files within the nupkg file. Now, thanks to you, I do see the db\output\PackageScript.sql...
@Mac_F, thank you the reply. However that does not absans my question. Please see my previous post for clarification. / comments
@Mac_F, thank you the reply. However that does not absans my question. Please see my previous post for clarification.
I agree with what @Eddie D said. To add onto that, I would recommend a production db as the "target" when creating the project of you already have a production db. I had some cases where we already had a shared dev db, but not a prod db. For these, I believe the best technique is to not set the "target db" during project creation. Set the development db to your own local, empty db. Then, connect to the shared dev db and import the existing objects as your first migration (rather than baseline). If you do set the shared dev db as the target during project creation, you can run into some deployment challenges down the line related to the "baseline" script if the deployment target is not completely empty (e.g. If you have any users or std objects in your model db). In that case, I have been able to resolve by either removing those objects or by: 1) unchecking "mark first folder as baseline" in the project settings. 2) adding <SkipBaselineCheck>True</SkipBaselineCheck> to the project file xml property group. / comments
I agree with what @Eddie D said. To add onto that, I would recommend a production db as the "target" when creating the project of you already have a production db.I had some cases where we already ...
@Monday - I love how your'e thinking here. I was considering adding that functionality to make it more robust, but decided on a quick dev first. / comments
@Monday - I love how your'e thinking here. I was considering adding that functionality to make it more robust, but decided on a quick dev first.
Bummer. I'd like to create coherent concept of "version" from SCA to Bamboo to Octo as well as in the __MigrationLog. Frankly, I want it to integrate with DLM Dashboard, too. But...I digress. I think I'll have to add a custom build step to recurse through the migrations folder to look for the latest sem ver folder name.... / comments
Bummer. I'd like to create coherent concept of "version" from SCA to Bamboo to Octo as well as in the __MigrationLog. Frankly, I want it to integrate with DLM Dashboard, too. But...I digress.I t...
Thank you, @Mike U! I think this falls under the umbrella of needing to update the docs. I'd be willing to help you and the team with this process. Cheers. -Peter / comments
Thank you, @Mike U! I think this falls under the umbrella of needing to update the docs. I'd be willing to help you and the team with this process. Cheers.-Peter
Thanks, team. Monday, I saw that
part of the docs and assumed it meant that ReleaseVersion would be
automatically populated, but that was not happening. I got it
working by manually creating a variable in Octopus - ReleaseVersion, and setting it
to #{Octopus.Release.Number}
Now, I'm seeing it populated in the dbo.__MigrationLog
table (e.g. "1.1.0-build22")
Which came from Bamboo via the Package Version: 1.1.0-build${bamboo.BuildNumber}. Making some
progress...
Now, I just want to find the right way to populate
the "1.1.0" in Bamboo from the SCA project...probably another post,
tho... / comments
Thanks, team. Monday, I saw that
part of the docs and assumed it meant that ReleaseVersion would be
automatically populated, but that was not happening. I got it
working by manually creating a v...
Thank you, Mike. So, I need to create an Octopus variable called ReleaseVersion and set it? Also wondering how I get the version info from the existing variables in Octopus in order to set ReleaseVersion? Edit: I created an Octoput variable, ReleaseVersion and set its value to: #{Ocopus.Release.Id} This just set it to Release.26, even though the name of the release was 1.1.0-build22...looking for another var... / comments
Thank you, Mike. So, I need to create an Octopus variable called ReleaseVersion and set it? Also wondering how I get the version info from the existing variables in Octopus in order to set Releas...