Comments
Sort by recent activity
Thanks, Kendar. Stoked to have you engage in this thread. To answer your questions... 1) At this point, we only have a production and dev integration environment (both Azure SQL DBs). As I dive into this, I'm considering adding some new ones to test releases/deployments - esp a stage before prod. Not sure if short-lived (even containers) might come into play. I do you hear you in regards to "resetting" the dev int env. 2) We have done a little with RG SQL Clone, but aren't using it on a reg basis at this point. I'm willing to consider options here. I also want to be cognizant of other devs with perhaps a lower tolerance for complexity and new tools. I do see value here, tho. 3) We are using VS SCA. We started with SSMS, but found it lacking in several areas. Plus our devs were actually more comfy with VS as they do a fair amount of SSIS dev. PS On a side note, I did play around some with a demo project n repo yesterday. I used a develop and main branch (git-flowy) config. Was able to get feature branches to CI build, but not release/deploy anywhere. PR into develop would CI build + release and deploy to a dev integration DB. PR into main does same for "prod". I'm heading to the mountains with my wife shortly. Catch you next week. Thanks for all you do! -Peter / comments
Thanks, Kendar. Stoked to have you engage in this thread.To answer your questions...1) At this point, we only have a production and dev integration environment (both Azure SQL DBs). As I dive int...
Also, if I wanted to complicate this already complicated discussion, I would add in the possibilities of both the SQL Clone for branch switching and the Hybrid SSC/SCA model. / comments
Also, if I wanted to complicate this already complicated discussion, I would add in the possibilities of both the SQL Clone for branch switching and the Hybrid SSC/SCA model.
Thanks, Eddie. That may suffice in a pinch. It would still be helpful if we could configure an alert to trigger one notification and then stop. / comments
Thanks, Eddie. That may suffice in a pinch. It would still be helpful if we could configure an alert to trigger one notification and then stop.
Thank you, @Sergio R. That's what I figured. I already have similar functionality in some of my powershell scripts that interact with Azure SQL DB. Cheers. / comments
Thank you, @Sergio R. That's what I figured. I already have similar functionality in some of my powershell scripts that interact with Azure SQL DB. Cheers.
Apologies for the delay. Yes, let's table this one and revisit if it rears up again. Thank you! / comments
Apologies for the delay. Yes, let's table this one and revisit if it rears up again. Thank you!
Hi @DanC Well, now it's going to be messy AND fuzzy - since I don't recall the exact details. I just recall that the pulled changes couldn't be applied since there were other changes in the dev's local DB (not yet in migrations) that conflicted. We also looked at trying to make migrations, but that, too...was messy. Apologies. I'm wishing I had a better record of the situation. In related news, we are seeing fewer issue overall. I think the devs are getting more familiar with the process and tend to "pull and apply" before making new changes. I'm also heading them towards a branching strategy. Lastly, we are moving the team to the VS SCA plugin for a number of reasons. The primary one being a more familiar source control interaction, but also a more evolved SCA experience overall. I wanted to work with the SSMS one, but it's just not "ripe" yet. / comments
Hi @DanC Well, now it's going to be messy AND fuzzy - since I don't recall the exact details. I just recall that the pulled changes couldn't be applied since there were other changes in the dev's l...
Thanks, @DanC. The actual event is now quickly becoming a blurry memory. I can't recall if the dev's VSC tab showed other files as coming through via pull. We had not mad any mods to sqlproj.user at that time, but we have since - to ensure that we use localhost (dev edition compat 150) instead of a sql localDB instance (compat 130). I'm not clearly undretanding your last question: "Assuring that you have both users using the same development on the local machine". Assuming you mean "Assuring that you have both users using the same development DB on the local machine". No - each dev uses their own sandbox dev DB on their laptops. And yes, gitignore has .sqlproj.user: # ----------------------------------------------------------------- # .gitignore for SQL Change Automation # ----------------------------------------------------------------- bin/ obj/ /*.dbmdl /*.ifm /*.sqlproj.user / comments
Thanks, @DanC.The actual event is now quickly becoming a blurry memory. I can't recall if the dev's VSC tab showed other files as coming through via pull.We had not mad any mods to sqlproj.user at...
Thanks, @Alex B - I don't mean to rush you and certainly appreciate your help. I'm afraid I don't have detailed info on the type of "merge" conflict that arose. i just have a vague recollection of a jumble of broken, pulled migrations when we tried to "apply to database" along with some perhaps unexpected "generate migrations". We wound up rebuilding his local dev DB from source (with the pulled code from dev2), then cherry-picking items from his old (renamed) local dev DB using SQL Compare. Finally, generating migrations for his changes. On a related topic, I'm attempting to nudge the team towards an actual branching strategy, esp after reading @Kendra_Little 's awesome posts on the subject: https://www.red-gate.com/blog/the-managers-guide-to-git-training-for-database-administrators?_ga=2.61671084.1604002557.1594680128-1689274135.1575913803&_gac=1.53677786.1591031064.CjwKCAjwztL2BRATEiwAvnALcinme8ckxPkFvrk69gYOI3ZGUujOzsawLl796KZjdWO82PZk8Ah8xBoCjY0QAvD_BwE Our team is not git-strong at this point. I would also say that the SSMS SCA plugin's git integration is not optimal. [image] But, that is a topic for a different thread. / comments
Thanks, @Alex B - I don't mean to rush you and certainly appreciate your help.I'm afraid I don't have detailed info on the type of "merge" conflict that arose. i just have a vague recollection of ...
Thanks, @Alex B. I think I'm looking for clarity in the language about changes in the dev sandbox DB vs actual migrations created and saved in eth working directory vs adding them to git staging vs committing to git. In our situation, there were changes in the dev's local sandbox DB, but no migrations created for them (and therfore no "uncommitted changes" from a src perepctive. When we pulled the other dev's migrations from remote repo and applied to DB, things broke. [image] We ended up renaming his local dev DB, rebuilding from src, and then using SQL Compare to bring his changes back into the new local DB. / comments
Thanks, @Alex B. I think I'm looking for clarity in the language about changes in the dev sandbox DB vs actual migrations created and saved in eth working directory vs adding them to git staging v...
@Alex B, Wondering if you might have some follow-up advice here. I'm still a little confused about what to do when dev 1 has some changes in their local dev DB (not yet as migrations or updated programmable objects), and we have dev 2 that has committed and pushed. Should dev 1:
pull and attempt to apply dev 2's changes?
Or - generate migrations/updated programmable objects and commit them before trying to pull and apply dev 2's changes?
/ comments
@Alex B, Wondering if you might have some follow-up advice here. I'm still a little confused about what to do when dev 1 has some changes in their local dev DB (not yet as migrations or updated pr...