summaryrefslogtreecommitdiff
path: root/utils/create-cluster/README
diff options
context:
space:
mode:
authorantirez <antirez@gmail.com>2015-01-28 23:26:42 +0100
committerantirez <antirez@gmail.com>2015-01-28 23:26:46 +0100
commit8899f91a7f29500e0005fa8a19211503267d884f (patch)
treea61c1eb4533432a6a306cc1b4a95ccbfecbd7f23 /utils/create-cluster/README
parentd3ef6c94bfa2d063f0fdbf128fba20119f5c1986 (diff)
downloadredis-8899f91a7f29500e0005fa8a19211503267d884f.tar.gz
create-cluster script added.
Simple shell script to create / destroy Redis clusters for manual testing.
Diffstat (limited to 'utils/create-cluster/README')
-rw-r--r--utils/create-cluster/README27
1 files changed, 27 insertions, 0 deletions
diff --git a/utils/create-cluster/README b/utils/create-cluster/README
new file mode 100644
index 000000000..f3a3f0883
--- /dev/null
+++ b/utils/create-cluster/README
@@ -0,0 +1,27 @@
+Create-custer is a small script used to easily start a big number of Redis
+instances configured to run in cluster mode. Its main goal is to allow manual
+testing in a condition which is not easy to replicate with the Redis cluster
+unit tests, for example when a lot of instances are needed in order to trigger
+a give bug.
+
+The tool can also be used just to easily create a number of instances in a
+Redis Cluster in order to experiment a bit with the system.
+
+USAGE
+---
+
+To create a cluster, follow this steps:
+
+1. Edit create-cluster and change the start / end port, depending on the
+number of instances you want to create.
+2. Use "./create-cluster start" in order to run the instances.
+3. Use "./create-cluster create" in order to execute redis-trib create, so that
+an actual Redis cluster will be created.
+4. Now you are ready to play with the cluster. AOF files and logs for each instances are created in the current directory.
+
+In order to stop a cluster:
+
+1. Use "./craete-cluster stop" to stop all the instances. After you stopped the instances you can use "./create-cluster start" to restart them if you change ideas.
+2. Use "./create-cluster clean" to remove all the AOF / log files to restat with a clean environment.
+
+It is currently hardcoded that you start a cluster where each master has one slave, since the script is pretty basic.