Annette Bieniusa

Something odd for the Test step in CI for Erlang 24. ``` ===> Release successfully assembled: _build/default/rel/antidote [37](https://github.com/AntidoteDB/antidote/runs/6831027292?check_suite_focus=true#step:8:38) test/release_test.sh [38](https://github.com/AntidoteDB/antidote/runs/6831027292?check_suite_focus=true#step:8:39) Starting Test transaction [39](https://github.com/AntidoteDB/antidote/runs/6831027292?check_suite_focus=true#step:8:40) Reading counter [40](https://github.com/AntidoteDB/antidote/runs/6831027292?check_suite_focus=true#step:8:41) Committing transaction [41](https://github.com/AntidoteDB/antidote/runs/6831027292?check_suite_focus=true#step:8:42)...

I find it a bit strange that only Erlang version 24 is affected.

From the documentation: > The call to start_transaction may fail if Antidote is unable to acquire the lock. What message is then sent to the client?

Does it only happen with a large number of data?

Tyler, you are doing indeed an awesome and difficult job here - thank you very much! > Differently when updates arrive that are propagated from other DCs, no snapshot is...

Can you please add some tests to make sure that the new variants are covered by the test suite?

Ping! What is the status of this PR?

@balegas ?

Thanks for spotting the issue! It seems that the function is not used, so this wasn't creating any problem. The actually used `insert_bigger` employs `vectorclock:le\2` so this should be correct....

I think the documentation is wrong here. If commitTime of T_1 < snapshotTime of T_2, the read in T_2 should return the value written in T_1, i.e. the incremented one....