Binbin
> I’m open to the idea of introducing this configuration, provided it defaults to off. i am glad that i got you, the default of this configuration is the same...
we can probably rename it to something like cluster-ignore-disk-write-error, just like `replica-ignore-disk-write-errors` and alian with it ``` # Replica ignore disk write errors controls the behavior of a replica when...
> Let's say we can tolerate write failures for say 5 minutes (or configurable). If the last successful write was older than that, we can crash. Then we can guarantee...
i would like to start adding some latency around it, see #1534
There is indeed does carry some risks as viktor mentioned. And in fact, it is always possible to happen, since a primary can suddenly crash, and during the failover, there...
We have a similar one, but it only works on FAIL. Perhaps we can extend this condition to work on more than just FAIL. ``` void clusterUpdateState(void) { int j,...
@zuiderkwast this is a small cleanup, do you want to take a look?
or maybe in CLUSTER REPLICATE, we should call a clusterDelNodeSlots(myself) if we find out myself is a replica. i will try to add a test to cover it later
> A replica in cluster should not possess any slot. I think we should forbid ADDSLOT and ADDSLOTRANGE to execute on a replica, and check whether "myself" has any slot...
> and check whether "myself" has any slot in CLUSTER REPLICATE we do this when master: ``` /* If the instance is currently a master, it should have no assigned...