Comments
Sort by recent activity
The code block did not format well. Let's try an image: [image] [image] / comments
The code block did not format well. Let's try an image:
Excellent. Thank you, Kendra. I also did some testing this morning - back and forth between plug-ins on the same project. Looks like it's working more smoothly than my previous tests last year. Could have been user funk or even a senior moment of course. Mostly just wanted to confirm that it wasn't considered a no-no. Could be helpful to see a note in the docs on this. On a side note, these tools are evolving nicely as I knew they would. So fun to be a part of this process. Cheers, -Peter / comments
Excellent. Thank you, Kendra. I also did some testing this morning - back and forth between plug-ins on the same project. Looks like it's working more smoothly than my previous tests last year. ...
Thank you, Kendra and Sebastian. I'm rolling with plan B at this point, but happy to continue the conversation. [image] The initial error in the "Redgate SQL Change Automation Build" extension tasks execution was: " Windows NT user or group '<myDomain>\<MyAADGroup>' not found. Check the name again." (Note that the build extension config is using SQL LocalDB) I do not see a place to change the auth method in the Build task...which agrees with Kendra's "...but currently this option isn't available in the graphical plugin in Azure DevOps". To answer your question, Sebastian, yes - is was within the release extension that I hit the "Principal '<myAADgroup>' could not be created. Only connections established with Active Directory accounts can create other Active Directory users." error. I do agree that switching the release extension task to AAD Password Auth would probably fix that issue, Sebastian. Getting back to Kendra's comments...I used to do a lot of SCA CI (Bamboo and Jenkins) using the SCA powershell functions, but haven't done that in Azure DevOps yet. I may do a POC on a demo/POC project top see if I can make some progress in that direction. Thanks for all your help, team! Cheers! -Peter / comments
Thank you, Kendra and Sebastian. I'm rolling with plan B at this point, but happy to continue the conversation. The initial error in the "Redgate SQL Change Automation Build" extension tasks execut...
I'm going to learn YAML, aren't I? / comments
I'm going to learn YAML, aren't I?
Hey, @Kendra_Little - Quick follow-up. I started making some progress by using dynamic SQL that tested whether the SQL was executing in Azure SQL DB vs SQL Server by using: IF SERVERPROPERTY('Edition') = 'Sql Azure' BEGINI tested it (with rollback) on my local dev SQL Server 2019 DB and on an Azure SQL DB. However, when I committed my code, the CI build process (Azure DevOps Pipeline + Redgate SCA extensions) barfed on the fact that it couldn't recognize the AAD group. I hacked the script even more by just skipping the CREATE USER code if this was just build validation on the CI server: IF DB_NAME() NOT LIKE 'sql_change_automation_%' BEGIN (...) That got me past the build, only to fail in the release deployment: Principal '<mySecurityPrincipal>' could not be created. Only connections established with Active Directory accounts can create other Active Directory users. So now I'm faced with the task of figuring out how to adjust my deployment agent service account...or just giving up and getting this migration out of the project and handling it manually. Leaning towards "B". [image] Thanks again for your help! Cheers, -Peter / comments
Hey, @Kendra_Little - Quick follow-up. I started making some progress by using dynamic SQL that tested whether the SQL was executing in Azure SQL DB vs SQL Server by using:IF SERVERPROPERTY('Editio...
Thank you, Kendra. I appreciate your reply and jiggling my noggin' by suggesting dynamic SQL. I'm a little embarrassed that I didn't think of that. I'm going to roll with that method for now. Thank you! / comments
Thank you, Kendra. I appreciate your reply and jiggling my noggin' by suggesting dynamic SQL. I'm a little embarrassed that I didn't think of that. I'm going to roll with that method for now. Th...
Thank you, Sujay. That's helpful. I was able to select the Top 10 Waits, then choose "all of history", then select a wait type. For some, it showed some related queries. For others it did not. I'm trying to correlate with SQL Agent jobs, so what would be really cool would be if I could see this info in the analysis page. That way I would be able to see patterns of wait "bubbles". / comments
Thank you, Sujay. That's helpful. I was able to select the Top 10 Waits, then choose "all of history", then select a wait type. For some, it showed some related queries. For others it did not. ...
I will admit to a little disappointment. Would have loved to get a discount on "Teams" flyway...will follow-up with our sales folks. Meanwhile, I'm using the dbops powershell module - it's free and provides a decent amount of similar functionality - super-simple, migrations-based, db change deployment. / comments
I will admit to a little disappointment. Would have loved to get a discount on "Teams" flyway...will follow-up with our sales folks. Meanwhile, I'm using the dbops powershell module - it's free an...
Thank you, @Tianjiao_Li. Since the community version is free and (I assume) "unsupported", it seems odd that it would restrict the SQL version based on "5 years of guaranteed database support timeline". Is there any way to turn off the warning/error in the Community edition or am I forced to use the Teams edition if I require a SQL 2014 target? Cheers, -Peter / comments
Thank you, @Tianjiao_Li. Since the community version is free and (I assume) "unsupported", it seems odd that it would restrict the SQL version based on "5 years of guaranteed database support time...
We are back. [image] I did have some disjointed visions of schema version and dependency tracking capabilities during my hike in the Sangres. The visions included using some blend of git commit messages and hashes along with the built in dependency metadata in SQL Server...along with stamping version paths for schema objects - tables n column - possibly in SCA metadata tables. Did I mention disjointed? / comments
We are back. I did have some disjointed visions of schema version and dependency tracking capabilities during my hike in the Sangres. The visions included using some blend of git commit messages ...