Oran Agra

from a quick look i took, we can see that in both cases, only one arena is actually being used, and the difference between `allocated` and `active` and `resident` is...

well, in theory it could be that something else was consuming RSS (not jemalloc), so it'll show in the process RSS and `used_memory_rss`, but not in allocator_resident and not in...

Thanks for taking a look. At the time I did review all the parts outside redis-cli.c and did skim through redis-cli.c to provide guidance and advice. But I felt the...

If we can pick this up and close it within a couple of weeks, I think it can fit 7.2

@zuiderkwast @jhelbaum i gave it some thought and also consulted Yossi and Itamar, i decided to take Viktor's advise and merge it without another reviewer, considering this is an isolated...

i no longer remember all the details, i suppose that considering it's a non-common edge case it doesn't worth fixing and breaking something. but on the other hand, if we...

it makes sense to me, but the fact there's an explicit test for this behavior scares me. maybe with other commands, clients expect getting an error (i.e. it's a "valid"...

the top comment needs an update, right? i.e. that's a new config that affects the generated config files, and some logging, but doesn't affect any commands at the moment, right?...

sporadic test failure: ``` *** [err]: Human nodenames are visible in log messages in tests/unit/cluster/human-announced-nodename.tcl log message of '"*Node * (nodename-1) reported node * (nodename-0) as not reachable*"' not found...

Full CI https://github.com/redis/redis/actions/runs/4176923677