Roy Dahan
Roy Dahan
IIUC Both 2024.1 (@dimakr please verify) and 2024.2 (at least the one I saw) still suffers from this issue. In that case, I don't understand why don't we handle this...
Revising this should be failry easy: 1. Clone a reproducer in 2024.2 / 2024.1 release - run it with 2024.1.0 - check if it fails the same. 2. Clone a...
Sounds good, small adjustment of 10% is much better.
So, If I understand it correctly, when you tested it manually, you actually reduced the job to 1 command and this one also by 10%.
ok, nothing bad happened. I wouldn't even invest in reverting, just try to reduce more than 10% and retest it.
Is this something we still planning?
This is already in a PR stage so I prefer not throwing it on someone else. If there is further work, we can move that further work to someone else.
In case it's a responsibility that need to be transitioned, @juliayakovlev should work with @temichus who now own MVs and help with the transition. Here it's a PR, so worth...
@temichus the coverage of "Synchronous MV" isn't complete. This PR from @juliayakovlev is an addition she was working on, when you have time to allocate and prioritze this task, please...
> Tier 1 tests are switched to using racks and it showcased a fundamental issue: with 3 racks we need to have 6 nodes total ( 2 in each rack)...