The test of Nik's case was not stable enough together with COMMIT AND CHAIN. So I've separated these cases into different permutations. Looking through CI logs it seems variation in sleeps and actual timeouts easily reach 30+ms. I'm not entirely sure we can reach 100% stable tests without too big timeouts. Best regards, Andrey Borodin. |
v21-0001-Introduce-transaction_timeout.patch
Description: Binary data
v21-0002-Add-better-tests-for-transaction_timeout.patch
Description: Binary data
v21-0003-Try-to-enable-transaction_timeout-before-next-co.patch
Description: Binary data
v21-0004-fix-reschedule-timeout-for-each-commmand.patch
Description: Binary data