Redis slows things?

If your site slows down with Redis caching enabled, it’s usually your Redis Server’s fault, unless your site is quite unhealthy and does 10,000 cache looks per page load.

Can you post your Redis metrics under load?

A little background on the server. The Redis server is on the same server as the Apache server. It is a dedicated server on a 24 core, Intel(R) Xeon(R) CPU X5670 @ 2.93GHz with 48GB of Ram.

It’s running ISPconfig and I have a dozen or so low traffic sites and my 1 main mid traffic site. The server typically runs at a very low load.

I am testing this on the staging site on the same server so the only load I can do is just go through the back end several times and show you metrics.

Here is a 5-minute video of me doing just that while including the server stats via the TOP command: https://nimb.ws/XBDtQH

I go to the Redis metrics a couple of times in the video but here is a quick screenshot of those: https://nimb.ws/7QrCWF

Afer I let it sit and run through it again here are the metrics: https://nimb.ws/5uAoia

Much slower it seems. These numbers are more indicative of what I noticed previously.

I’d suggest setting up dedicated Redis monitoring to see what’s causing it.

How do I do that?

Your server admin or Google will know. There are many tools/approaches.

This is a bare metal server with no major support so I am on my own for the most part here.

I found some helpful pages searching google… here is a test run with some commands running on the server: https://nimb.ws/H0v3ye

Here is the output of the info command when this is all done:

# Server
redis_version:3.2.12
redis_git_sha1:00000000
redis_git_dirty:0
redis_build_id:7897e7d0e13773f
redis_mode:standalone
os:Linux 3.10.0-1127.13.1.el7.x86_64 x86_64
arch_bits:64
multiplexing_api:epoll
gcc_version:4.8.5
process_id:1373
run_id:2f258f78b586e4d97d42d399704dd01308b8df8a
tcp_port:6379
uptime_in_seconds:2576833
uptime_in_days:29
hz:10
lru_clock:8805063
executable:/usr/bin/redis-server
config_file:/etc/redis.conf

# Clients
connected_clients:1
client_longest_output_list:0
client_biggest_input_buf:0
blocked_clients:0

# Memory
used_memory:4730608
used_memory_human:4.51M
used_memory_rss:10207232
used_memory_rss_human:9.73M
used_memory_peak:223054352
used_memory_peak_human:212.72M
total_system_memory:50465189888
total_system_memory_human:47.00G
used_memory_lua:37888
used_memory_lua_human:37.00K
maxmemory:268435456
maxmemory_human:256.00M
maxmemory_policy:allkeys-lru
mem_fragmentation_ratio:2.16
mem_allocator:jemalloc-3.6.0

# Persistence
loading:0
rdb_changes_since_last_save:141486
rdb_bgsave_in_progress:0
rdb_last_save_time:1650395398
rdb_last_bgsave_status:ok
rdb_last_bgsave_time_sec:-1
rdb_current_bgsave_time_sec:-1
aof_enabled:0
aof_rewrite_in_progress:0
aof_rewrite_scheduled:0
aof_last_rewrite_time_sec:-1
aof_current_rewrite_time_sec:-1
aof_last_bgrewrite_status:ok
aof_last_write_status:ok

# Stats
total_connections_received:6704
total_commands_processed:715918
instantaneous_ops_per_sec:0
total_net_input_bytes:30046973656
total_net_output_bytes:34998383404
instantaneous_input_kbps:0.00
instantaneous_output_kbps:0.00
rejected_connections:0
sync_full:0
sync_partial_ok:0
sync_partial_err:0
expired_keys:443
evicted_keys:0
keyspace_hits:472173
keyspace_misses:74804
pubsub_channels:0
pubsub_patterns:0
latest_fork_usec:0
migrate_cached_sockets:0

# Replication
role:master
connected_slaves:0
master_repl_offset:0
repl_backlog_active:0
repl_backlog_size:1048576
repl_backlog_first_byte_offset:0
repl_backlog_histlen:0

# CPU
used_cpu_sys:1114.31
used_cpu_user:446.39
used_cpu_sys_children:0.00
used_cpu_user_children:0.00

# Cluster
cluster_enabled:0

# Keyspace
db0:keys=3200,expires=199,avg_ttl=346706741

You’ll want a real monitoring tool that keeps track of your connections, ops/s and similar over time, since it’s not temporary issue.

This is a staging server. There isn’t really any traffic or anything to monitor at all over time.

It only seems to make sense to monitor it while I’m actually doing things in the backend or running a test load?

Is there nothing simple to check in the configs or what I provided that would clue us in to an issue?

No, there isn’t something simple to check. Your server is responding fast, then extremely slowly. Something is causing that and it’s not this plugin.

I can confirm that this plugin slows WordPress substantially. With it enabled performance is worse!

@bganush: No it doesn’t. Your Redis Server responding too slowly may slow things down. Or your codebase being a crime against humanity may. All plugin does it give WordPress an API to use, it doesn’t actually do anything. If things are slower use an APM tool to find out why.