Disable Persistence of Redis

[Georg Höllrigl]

Hello,

Im running redis 2.6.14 on an EC2 instance. About every two minutes it writes out the dump.rdb file which means with about 10GB memory, this blocks the whole redis process for 10 - 20 seconds which is rather annoying for the app!

The save value is already disabled in the config, and the server restarted:

redis 127.0.0.1:6379> CONFIG GET save

1) "save"

2) ""

I've also tried setting the save value rather high - like save it every 100years (about 3153600000 seconds) and some high number of changes.

Anyone can give me an idea, why redis still saves to disk?

Kind Regards,

Georg

Here is the used config (from the slave)

daemonize yes

pidfile /var/run/redis/redis-server.pid

port 6379

timeout 0

loglevel notice

logfile /var/log/redis/redis-server.log

databases 16

stop-writes-on-bgsave-error yes

rdbcompression yes

rdbchecksum yes

dbfilename dump.rdb

dir /var/lib/redis

slaveof redis.master.xxxxxxxxxx 6379

slave-serve-stale-data yes

slave-read-only yes

slave-priority 100

maxmemory-policy volatile-ttl

appendonly no

appendfsync everysec

no-appendfsync-on-rewrite no

auto-aof-rewrite-percentage 100

auto-aof-rewrite-min-size 64mb

lua-time-limit 5000

slowlog-log-slower-than 10000

slowlog-max-len 128

hash-max-ziplist-entries 64

hash-max-ziplist-value 512

list-max-ziplist-entries 512

list-max-ziplist-value 64

set-max-intset-entries 512

zset-max-ziplist-entries 128

zset-max-ziplist-value 64

activerehashing yes

client-output-buffer-limit normal 0 0 0

client-output-buffer-limit slave 256mb 64mb 60

client-output-buffer-limit pubsub 32mb 8mb 60

[Henrique Mecking]

Hi Georg,

can you paste here the output of "info Persistence"?

Att,
Henrique Mecking.

--
You received this message because you are subscribed to the Google Groups "Redis DB" group.
To unsubscribe from this group and stop receiving emails from it, send an email to redis-db+u...@googlegroups.com.
To post to this group, send email to redi...@googlegroups.com.
Visit this group at http://groups.google.com/group/redis-db.
For more options, visit https://groups.google.com/groups/opt_out.

[Yiftach Shoolman]

This could be because of replication issue. I would assume your master client buffer is overflow (please look at the master Redis log) and therefore your slave is trying to sync and fails in a loop. Each sync your master snapshot your dataset to persistent storage (I can guess EBS), and because your dataset is relatively large, fork time takes too long and blocks your Redis, see more details here:http://redis.io/topics/latency.

To solve the issue (assuming it's indeed this one), just set something like this:

config set client-output-buffer-limit "slave 536870912 536870912 60"

--

Yiftach Shoolman
+972-54-7634621

[Georg Höllrigl]

Hello,

The solution to problem was somewhat different. 

Our Devs missed a script that issued the save command from time to time.

So it was not a redis problem, but a problem in our app.

Thank you for your time and sorry to bother you.

I guess the final solution will be to disable the save command with something like:

rename-command SAVE ""
rename-command BGSAVE ""

Kind Regards,

Georg