06:37:44 GMT <badboy_> crayon: > the AOF persistence logs every write operation received by the server, that will be played again at server startup, reconstructing the original dataset. Commands are logged using the same format as the Redis protocol itself, in an append-only fashion. Redis is able to rewrite the log on background when it gets too big.
07:47:47 GMT <hurdos> Hi
07:51:15 GMT <hurdos> I configured cluster from 6 nodes (3M, 3S). And when operating the cluster I see that huge load is on one of master node.
07:51:22 GMT <hurdos> Is it normal?
07:53:47 GMT <hurdos> http://imgur.com/CdbWtCI
08:27:02 GMT <badboy_> if you're unlucky, yes
11:59:43 GMT <niels_> Hi, I'm trying to set up redis with 3 physical nodes where clients are also on those same nodes. I currently have Redis Cluster running and the clients just talk to the redis instance on the local IP and get redirected to the node that holds the data. This works fine, but when we disable a physical node the failover seems to fail. From what I understood Redis Cluster does automatic failover, but this doesn't appear to happen now. Does a
11:59:46 GMT <niels_> the cause?
12:01:03 GMT <niels_> I mostly followed the steps here so replicas is 1 - and the cluster nodes output has a total of 6 instances, 3 master, 3 slaves: https://redis.io/topics/cluster-tutorial
12:08:14 GMT <niels_> I have some output as well from redis-trib checks and cluster node output when all is OK and also when we try a failover: https://pastebin.com/raw/0VmKgHED
12:23:50 GMT <badboy_> check the logs
15:16:05 GMT <sgrover> New to redis.  I have a web server with redis installed locally.  Now we want to use Redis on a remote server.  So I'm thinking just set up the remote as a slave to replicate the data, and then point my website at that server after the initial sync is completed.  Am I missing anything?
15:16:16 GMT <sgrover> Is there a "promote to master" step involved?
16:02:01 GMT <crayon> badboy_, im looking to ingest logs into our SIEM and log on commands like FLUSHDB and CONFIG the AOF log wouldnt log these from my understanding. these are only output using 'MONITOR" and that output is not written to disk -- please correct me if i am entirely wrong
19:30:40 GMT <badboy_> crayon: FLUSHDB is clearly a write command, that would end up in the AOF (until compaction). Read commands won't
19:31:12 GMT <badboy_> MONITOR is the only way to see read commands. It has an overhead though, so it might not be wise to keep it running in production
19:31:48 GMT <crayon> hrm
19:32:05 GMT <crayon> so if i were to want to create security monitoring i would need MONITOR and AOF then
19:32:08 GMT <crayon> is that correct?
19:32:28 GMT <badboy_> why both?
19:32:57 GMT <crayon> in the scenario where someone dumps the database or in the scenario where someone drops the db