inakisoriamrf

Hi @agosalvez you have more details at https://github.com/Snapchat/KeyDB/issues/460 If you need more info I'll be more than happy to solve any questions

Thanks @JohnSully If you need any further information I'm 100% available

Hi @JohnSully We use simple strings like `r:1567:{d2}0c5731-d778-4b2b-9a57-3daf9926fd8e` (expiring) or `e:123:912ec803b2ce49e4a541068d495ab570` (not expiring). About 5% of the keys never expire, the other 95% expire in 48 hours although most of...

Hi @JohnSully , any update on this? We may have a way to reproduce it or at least some hints on what's happening we moved from 9 servers 3 shards...

It's getting worse. Fragmentation grows crazy at some point taking the shard down Reported memory usage:  Real memory usage: 

Hello @michael-grunder we still having this problem, maybe you can guide me on where you think we have to look at and we can implement the patch. Thank you!

Hello @michael-grunder we have the following configuration ```PHP $redis = new RedisCluster( null, $this->config->getNodes(), $this->config->getTimeout(), $this->config->getReadTimeout(), $this->config->getPersistent(), $this->config->getAuth() ); $redis->setOption(RedisCluster::OPT_SLAVE_FAILOVER, $this->config->getReadsDistribution()); //With current values $redis = new RedisCluster( null, ['1.1.1.1:6379','1.1.1.2:6379',...],...

The main motivation for this change is to improve availability under failure scenarios. In our workload, if all replicas in a shard fail, it’s preferable to continue serving reads from...

@michael-grunder something I have to change here? Maybe it is a KeyDB configuration issue? I see there are a lot of tests Skpped for KeyDB

I've added the WAITs, hopefully this helps.