[ad_1]
We’ve been diagnosing a problem with our redis cache after updating WordPress from 5.0.3 to 5.9.2 and Redis from 1.4.1 to 2.0.23 and utilize Elasticache in AWS.
Since the update, the cache will slowly fill to 100% DB usage, then it begins evicting and reclaiming items until the item count in the cache is at 1 item while the DB usage is still at 100%.
The remaining key is a sorted set, wp:redis-cache:metrics, and gets to 10GBs in size, taking up the entire cache and rendering it useless until it is manually flushed.
I wasn’t able to find any related issues just yet and was curious if this has been seen before. Let me know if you need any additional details!
Thanks!