Ran Shidlansik
Ran Shidlansik
@KarthikSubbarao Overall LGTM. I do think we need to add more tests so that future changes will not introduce degradation. For example lets include function and redis.call cases ARE getting...
> > @KarthikSubbarao Overall LGTM. I do think we need to add more tests so that future changes will not introduce degradation. For example lets include function and redis.call cases...
LGTM but would like to hear @lipzhu
@KarthikSubbarao Thank you for driving this! I am reading through the PR and will provide technical comments there (if I have any) but wanted to ask here: Why is that...
@enjoy-binbin did we have a test failure here, or should we introduce new test for this case?
On a general level I think that (given the pr ID) this PR is very risky :)
> > On a general level I think that (given the pr ID) this PR is very risky :) > > Why you think it is the risky? it is...
> It occurred to me that maybe we also want to cancel server.rdb_bgsave_scheduled? I also thought about that during the implementation. I think the main goal was to kill the...
> Looks like it could be a complement to SCHEDULE. I am ok with both. > > Another question here, `kill the current running bgsave`, does it include the disk-based...
I would also change the top comment to indicate this is not only fixing the test. the issue is probably also related to keyspace notifications and pubsub.