Rafi Shamim

As mentioned above, the SQL standard (and the Postgres docs) don't guarantee ordering across subqueries, but PG often does preserve the order in practice. It seems potentially brittle to rely...

Here is some more discussion on the matter, with better sources: https://dba.stackexchange.com/a/184167 Quote from Postgres docs: > If sorting is not chosen, the rows will be returned in an unspecified...

I don't know why the SIGSEGV is happening but I stumbled upon this: https://github.com/golang/go/issues/27505#issuecomment-713706104 Do you think we need something like that? Do we need `Setpgid=true`?

hi @danaugrs! the docs here are more up to date: https://www.cockroachlabs.com/docs/stable/build-a-go-app-with-cockroachdb-gorm.html we will also plan to update the code in this repo.

Thanks for the issue. The examples-orms docker builder image uses Java 8, so it should work with Java 8. But I think the right direction would be to update examples-orms...

Update: still working on this, but my local ruby environment had broken since the last time I used it. I should have things working locally soon.

I have the tests running locally. I used the test file below then ran with `RUBYOPT="-W0" TESTOPTS="-v" TEST_FILES=test/cases/transactions_test.rb bundle exec rake test`. I have reproduced so far: - on e723a85739...

RIP i wish i knew about the `TESTOPTS='-n=/ConcurrentTransactionTest#test_transaction_isolation__read_committed/'` trick before

~Here's a debug.zip from a repro on 247332b running with `--vmodule=replica_evaluate=2`. I'm working on the bisect as well.~ [debug.zip](https://github.com/cockroachdb/cockroach/files/7368539/debug.zip) Whoops.... I ran against the wrong commit

OK, here's the debug.zip from https://github.com/cockroachdb/cockroach/commit/247332bc191a6d7bbcb043bf45509dc9517f78cf [debug.zip](https://github.com/cockroachdb/cockroach/files/7368603/debug.zip)