That sounds like your Redis Server is not reachable “quite frequently”. Consider monitoring it and assigning it more resources.
I also had this same error. It appears to be a memory allocation problem for me. The Redis Cache was not big enough and plesk could not allocate the full amount I had asked it to.
Try checking how much free ram you have and restart redis cache to see if it comes back up again.
As a temp solution. Disable redis and just use whatever page caching you have in place.
@thomasjarvisdesign: How did you discover this? Would it be helpful to see the current Redis memory size in the dashboard?
@tillkruess
Sorry it has taken a while to get back to you.
This randomly happened on a site and I had to delete the file suggested in the original post to bring the site back online.
Redis is installed in a 3gb docker container on the server. When attempting to re-initialise Redis from Docker it couldnt allocate the full 3gb (error code explaning issue cannot remember exact code). The cause (unverified) is below:
Now that memory use is down again:
I think this is to do with me running Redis in Docker and Memory overcommit must be enabled. Memory Overcommit is disabled and I recently started using Optimus Cache Primer which causes the server resources to hit 100% for a hour or so every night and also spank Redis cache as it will cache all of the queries made by the cache primer.
I am going to remove Redis and Docker and then install Redis directly onto the server. Then re-configure with Overcommit enabled. I think this will be more stable. I had taken a novice approach to this and used the Bitnami version of Redis on Docker because it includes a load of common variables that would otherwise need to be added to the Redis conf file. But now I understand how it works I wont really have an issue adding/editing the conf file without using Docker.
I hope this helps. I’ll let you know if it falls over again and raise my own ticket with full details.
