49 results found
-
Linux Compatility
Given that no GUI is required with Release and dotnet core is out, has this option been considered? Most deployment tools are Linux based and Octopus is limited in usefulness, or at least our application has outgrown it.
31 votes -
Ability to exclude static table columns when deploying static data
Currently I have option to include or exclude data compare using -IgnoreStaticData switch (New-DatabaseRelease cmdlet).
But since part of my static data tables columns are environment based (in each environment dev/qa/prod I got different url)
I have to exclude columns.Via Data Compare commad line I am using /Project switch which contains the exclude columns definitions.
(Another method that exists is IncludeColumns and /ExcludeColumns switches)I need this options in order to start using SQL Release, otherwise static data deployment will make undesired changes.
27 votesWe have a couple of options for this which may help some folks who have voted or commented.
For SQL Change Automation projects, the SSMS plugin now allows users to select only specific columns to add to the project for static data tables.
For both SQL Change Automation projects and for SQL Source Control projects, you also have the option to use a post-deployment script to dynamically control how static data is deployed to a given environment.
An example post-deployment script for this is here: https://documentation.red-gate.com/soc7/common-tasks/working-with-pre-post-deployment-scripts/static-data
-
As an deployment engineer I'd like to run environment specific scripts during deployment
Say we have scenarios where we have client machine configuration stored against machine names. When we deploy to DEV environment, these machines are different to TEST and we need a way to reseed the configuration data for the specific environment.
Having first class support for PreDeploy / PostDeploy scripts which could then be named using Environments (A.DEV.sql, A.TEST.sql) would be a nice feature - Especially if coupled with SQL Source Control.
We've worked around this at the moment but would love this OOTB
21 votesKendra respondedThanks for this feedback.
We have quite strong support for this request in SQL Change Automation projects, which support multiple pre- and post- deployment scripts and which also have an option to “seed data” for larger tables which you would like to populate: https://documentation.red-gate.com/sca/developing-databases/concepts/data-population/strategies-for-data-population
For SQL Source Control, we now support pre- and post- deployment scripts, however you are limited to only one of each type. Additionally, the seed data methodology is not available in SQL Source Control.
-
Rollback deployments
I would like to be able to rollback deployments done with SQL Release.
20 votesOur current guidance for rollback support on SQL Change Automation is here: https://documentation.red-gate.com/sca/developing-databases/concepts/advanced-concepts/rollbacks
Our roadmap for 2021 does include further researching and evolving rollback procedures: https://www.red-gate.com/products/redgate-deploy/roadmap
-
Flag to Force a deployment
Occasionally the only changes in a specific build is the addition to the PostScript post-deployment script. Since there are no schema changes, the PostScript change is not executed.
Faking a schema change by adding a comment or something is a poor workaround in an automated CI process.
A simple flag -AlwaysRunUpgrade to always run the PreScript and PostScript as part of the deployment would be ideal.
17 votesFor clarification for readers, this request is specific to SQL Source Control projects which are deployed by SQL Change Automation. Comments in the item give more detail.
-
Support semantic versioning
NuGet 3.5 will support semantic versioning and it would be nice if DLM Automation could implement it too.
14 votes -
Build Plugins - Ignore Static Data
When performing a database build for Continuous Integration Plugins can we have a setting to ignore static data?
13 votes -
Deploy to multiple databases concurrently
I want to be able to release database changes across several databases. Once confident that the targets are all ready to go, and that the deployment script is correct, I want the deployment to be concurrent (so that it takes less time).
12 votesWhile we haven’t implemented this exact feature, I want to share some information with similar patterns which may help some readers, or inspire comments or feedback in others.
One common approach is to manage multi-database deployments via an orchestrator, such as Azure DevOps, Octopus Deploy, or similar. This may be managed concurrently or in a simple loop depending on the needs. One example of this with Azure DevOps is demonstrated here: https://www.youtube.com/watch?v=-rZxLCRrgmI
When using a single release artifact to deploy to many databases, it’s important that you control for database drift on a regular basis outside of the deployment process. One way to do that is to set up monitoring that alerts you to schema changes that occur. An example of monitoring for this in SQL Monitor is in this custom metric: https://sqlmonitormetrics.red-gate.com/unauthorized-object-changes/
-
new-databaseDocumentation: add filetype parameter
The sqldoc executable has a filetype parameter that allows you to change the generated documentation's file type. Could you add the filetype parameter to the new-databaseDocumentation cmdlet to enable flexibility on the documentation type that is inside the build artifact? I realize that I could probably build the documentation on my own via the sqldoc command line, but I feel like if I'm going to just use the individual command lines it defeats the purpose of the module! Based on the initial pass I took through the schemaDocumentation object structure - I am optimistic that this is a reasonable ask.…
10 votes -
Support publishing custom schema names to DLM Dashboard
Please make it possible to specify a custom schema name when publishing a build to DLM Dashboard. The current implementation only supports using the nuget package version which has restrictive formatting rules, so for example it does not work for apps which use a four-part version number.
9 votes -
Produce code coverage from Azure Devops pipeline
When using the Redgate Change automation plugin for Azure DevOps pipelines, I wish to be able to run the tests AND produce the code coverage.
We can create the code coverage from SSMS, and it would seem to be available from Powershell, but it's not available from the Plugin.
We wish to expose this so that we can add a code coverage quality gate on Pull requests.
8 votesKendra respondedThe sqlcodeguard.exe command line may help with this request. The command line is in ‘preview’ mode and we value your further/continued feedback if you like the feature.
-
Improve Performance - Invoke-DatabaseBuild - By removing powershell progressbar
I have discovered that too much information is being processed to simply display a percentage completed with the Invoke-DatabaseBuild command. I suspect other SQL Change Automation powershell commands are also utilizating the same behavior.
By turning the powershell progressbar off via the ps setting.
$ProgressPreference = "SilentlyContinue"
I improved a single database build performance:
From: 4 minutes -> To: 37 seconds.I suspect there's alot of data serialization and deserialization taking place within SQL Change Automation update progressbar code to do a simple progressbar reporting update.
This code within sql change automation needs to be looked at more closely to…
7 votes -
Add invalid objects test
Like the test in SQL Prompt.
7 votes -
Incorporate new/improved SQL Compare differences views into DLM Automation Changes.html
The latest version of SQL Compare introduced improvements to the SQL View as well as a new Summary View, as detailed here: https://www.red-gate.com/hub/product-learning/sql-compare/sql-compare-summary-view
It would be good if the Changes.html file generated by DLM Automation could also display the results in the same format so that it is easier and clearer to see the changes that are being deployed.
It would also be better if changes to static data were broken down so that for each static data table modified it explicitly states which records have been added/removed/modified/remain identical.
6 votes -
Add SqlCmdVariable parameter to Build procedure
We would like to add SQLCMD variables to our SQL Change Automation projects to denote whether a deployment is applying to a Local environment or to our production pipeline (isLocalDeployment). Our thought for how to do this with the least amount of changes required was to default the variable to be isLocalDeployment = FALSE. This would mean that if we are doing a deployment to a local developer database we would run additional scripts to configure that environment, but once the project is deployed into our normal production deployment pipeline those scripts would no longer run and we wouldn't have…
6 votes -
Disable nuget package version normalization in TeamCity plugin
In TeamCity plugin step "Redgate SQL Change Automation Build" build.version is being normalized to SemVer (for example 2.1 -> 2.1.0 or 2.1.0.0 -> 2.1.0).
Because we use these nuget packages internally only we don't need to follow SemVer. It seems that Nuget team doesn't want to remove that normalization (https://github.com/NuGet/Home/issues/3050) so solution would be to create and use custom Nuget build. Thats how Octopu Deploy solved this issue: https://github.com/OctopusDeploy/Issues/issues/28986 votes -
Add an option to exclude stub check scripts from the package script
We have to manually remove our stub check scripts from our package scripts on every build. Can you please add an option to automatically exclude stub check scripts?
6 votes -
Include the name of the object that failed to deploy in SQL CI error messages
I want to know which object failed to deploy when a SQL Server error is thrown during validation.
6 votes -
As an build engineer I'd like to be able to override the package version number in the TeamCity plugin
We are currently using GitFlow for a small POC and have hit an issue as we are unable to use a NuGet safe version number for our package name without going through some hoops.
Adding support for an optional version number field in the SQL CI TeamCity plugin would be a nice addition and allow us to drive this from another variable.
5 votes -
Team City Plugin - Timeout Field
During the build portion of our pipeline, builds occasionally fail due to timeout while inserting static data. There is a work around where you open up the powershell scripts underneath DLM to specify the timeout, but it would be much more useful to have this as a field in the Team City plugin.
I'm sure other plugins could use this as well.
4 votes
- Don't see your idea?