Shay Rojansky
Shay Rojansky
@Brar my fear has less to do with `dotnet build` performance (where you have process startup, compiler bootstrapping and all the rest), and more with the inner loop when working...
Closing as per https://github.com/npgsql/npgsql/issues/3890#issuecomment-1242988454.
> I ask because every ~300 UTs, we see _dbContext.Database.MigrateAsync(cancellationToken) is getting a timeout after 30 or even 60 seconds. This sounds odd - I'd recommend investigating this a bit...
> It works like this: every test generates a unique guid as the database name and runs the EF migrations on it. Yep, so you're creating a database per test;...
Yep, TRUNCATE is a good way to manage things, and I agree with @Brar's suggestion of adding RESTART IDENTITY in case the actual generated identity values are somehow asserted on...
Decision: don't cache the PG14 async notification, and use a timeout just like pre-PG14. The alternative here would be to constantly read from the socket, in order to receive all...
Closing as no further question was posted, and this seems like a user error.
@ElmirSecerbegovic sorry for not answering earlier, this issue slipped under my radar. This looks like it belongs in https://github.com/npgsql/efcore.pg, but in any case, the issue is lacking information. Can you...
As per @vonzshik, putting on the backlog until there's support for this on the PostgreSQL side.
Don't remember any more what (if anything) we support in 5.0 for unpooled connections in TransactionScope, so it's for checking that.