Artsiom Mishuta

Results 103 comments of Artsiom Mishuta

> I think we should resolve the known issues first so that we can merge the test without needing to skip injections. > > I commented on each issue you...

> @temichus my problem is that these issues will not be fixed because they describe expected behavior of the system. I can close them as "not planned". Then what? so,...

(latency during grow-shrink 650gb) - test_latency_write_with_nemesis ![image](https://github.com/user-attachments/assets/99e36c23-59b9-4021-a57e-268580300575) (latency during grow-shrink 650gb) - test_latency_mixed_with_nemesis ![image](https://github.com/user-attachments/assets/6c1d71cf-0415-4b6f-8d3c-a4806975de82) (latency during grow-shrink 650gb) - test_latency_read_with_nemesis - has latency >15 ms but hints are not creating...

(latency during grow-shrink (tablets) 650gb) - test_latency_write_with_nemesis - has latency >15 ms but there are only few hints at this time ![image](https://github.com/user-attachments/assets/a8013858-a0fc-469a-853d-74cc13786286) ![image](https://github.com/user-attachments/assets/2e05cb2c-d31b-47f5-bf6f-f31a570d8cd9) (latency during grow-shrink (tablets) 650gb) - test_latency_mixed_with_nemesis...

> @temichus > > we know of an issue in decomission [scylladb/scylla-enterprise#4204](https://github.com/scylladb/scylla-enterprise/issues/4204) but in those results there is also high latency in one of the add nodes, > > can...

@fruch rerun results - no issues ![image](https://github.com/user-attachments/assets/f5af41e6-b275-4caf-b828-35a6fbdd4abe) ![image](https://github.com/user-attachments/assets/4122ab64-876d-4af3-af79-03749eff35c4) ![image](https://github.com/user-attachments/assets/f2a2c8e1-119c-42e9-a17f-78faf7309d5c)

fresh bisect runs for hinted-handoff-grow-shrink-latency-test and branch-perf-v15 branches write hinted-handoff f3725882-16ec-4e86-87e4-082aa5eaf781 ![image](https://github.com/user-attachments/assets/71fb6262-5cbc-4ba3-8a90-ec750c30f977) ![image](https://github.com/user-attachments/assets/0c72e2fb-708e-49e9-9746-a5cf55a782f9) write branch-perf-v15 3dbfeda8-3947-41f9-bda1-596c2b4e1e42 ![image](https://github.com/user-attachments/assets/20687af2-5efc-464d-80cb-66e82b847766) ![image](https://github.com/user-attachments/assets/9db729c3-f472-4a19-a7f9-35c85d54fad7) read hinted-handoff 7bab0369-2308-434b-8316-a62a0dab2c39 (Total test HDR is RED) ![image](https://github.com/user-attachments/assets/9ad260e7-a886-487d-bcbd-1345230363a3) ![image](https://github.com/user-attachments/assets/68f6c123-18b8-417d-913f-e83dee6a0c7c) read branch-perf-v15...

for latency-650gb-grow-shrink-tablets-v2 both ci runs are green hinted-handoff https://jenkins.scylladb.com/view/staging/job/scylla-staging/job/artsiom_mishuta/job/hints/job/scylla-master-perf-regression-latency-650gb-grow-shrink-tablets-v2/4/ branch-perf-v15 https://jenkins.scylladb.com/view/staging/job/scylla-staging/job/artsiom_mishuta/job/hints/job/scylla-master-perf-regression-latency-650gb-grow-shrink-tablets-v2/5/

@roydahan, @juliayakovlev according to https://github.com/scylladb/qa-tasks/issues/1705#issuecomment-2491220067. fresh bisect runs for hinted-handoff-grow-shrink-latency-test and branch-perf-v15 branches showed the same results. this PR can be merged

Test: performance_regression_test.PerformanceRegressionTest.test_latency_write_with_nemesis Test start time: 2024-07-18 15:04:15 Test id: aafbc802-1c7b-4421-8bf7-97c8453349b3 Scylla Server Version: 6.1.0~dev.20240717.21c67a5a649f with build-id 2cf8162ba323ae6faef5a1e7562200f1d25a6834 ![image](https://github.com/user-attachments/assets/c6ff444e-9fd3-48d0-a34e-5c1d169e5c99) ![image](https://github.com/user-attachments/assets/a83e7729-0e49-4b32-901a-7df618c339b3) ![image](https://github.com/user-attachments/assets/1d4ba89d-b0cf-4545-9d76-9abee0847014)