Roy Dahan
Roy Dahan
@rzetelskik IIUC and it also came up in my discussion with @fruch after reviewing the test, after you drain/delete a pod, the new pod that is added instead of it,...
I think that the wait that was added, is what let gossip to notice the changes (ring_delay?). @asias can shed more light here. Regarding the decommission that shouldn't succeed, I...
@KnifeyMoloko, (cc @soyacz who wrote the feature & @k0machi who opened https://github.com/scylladb/scylla-cluster-tests/issues/5299). The second [comment](https://github.com/scylladb/scylladb/issues/10772#issuecomment-1165569796) on the issue doesn't look related to the original issue (unless I'm missing something). In...
@aleksbykov can you please check why the resharding caused enospc here? Also, please compare it to previous runs. The coredump during enospc is a known issue.
@aleksbykov please add the information on how much data the test writes and the list of nemesis we run. AFAIK, the list of nemesis and gemini_seed should be fixed across...
Ok, so if the test schema can explain how we reached 2.9 TB with a raw dataset of 1TB (e.g. STCS, MV, SI), I think we can close the issue...
I suggested to remove the high. I don't know how often it happens and if it's just a temporary state of high load.
@asias can you please take a look?
Let’s run a test without this s-b command and see if it works or not. On Sun, Sep 18, 2022 at 09:53 Israel Fruchter ***@***.***> wrote: > @asias , @fgelcer...