https://github.com/r4fek/scylla-cli

Python script for managing and repairing Scylla Cluster.
https://github.com/r4fek/scylla-cli

cassandra repair scylla scylla-cluster scylladb

Last synced: 2 months ago
JSON representation

Python script for managing and repairing Scylla Cluster.

• Host: GitHub
• URL: https://github.com/r4fek/scylla-cli
• Owner: r4fek
• License: mit
• Created: 2019-03-18T13:21:20.000Z (almost 6 years ago)
• Default Branch: master
• Last Pushed: 2024-05-03T19:46:19.000Z (8 months ago)
• Last Synced: 2024-10-13T19:06:42.011Z (3 months ago)
• Topics: cassandra, repair, scylla, scylla-cluster, scylladb
• Language: Python
• Homepage:
• Size: 147 KB
• Stars: 7
• Watchers: 4
• Forks: 1
• Open Issues: 1
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# scylla-cli

Python script for managing and repairing (in a smart way) Scylla Cluster.

It repairs the primary range of a Scylla node in discrete steps using 

[advanced repair techniques](https://www.datastax.com/dev/blog/advanced-repair-techniques).

## Prerequisites

* Python >=3.5

* SSH connection or direct access to every Scylla host in the cluster should be possible

* Scylla REST API listens on every Scylla host on `:10000`

Following env variables could be set to make your life easier:

```

export SCYLLA_HOST='10.210.92.46' # initial endpoint for obtaining cluster topology

export SCYLLA_USERNAME='root' # ssh user on Scylla host

export SCYLLA_PKEY='/home/USER/.ssh/id_rsa' # Defaults to ~/.ssh/id_rsa

export SCYLLA_CONNECTION_METHOD='ssh' # or 'direct'

```

## Installation

```

pip3 install scylla-cli

```

## Developing

```

virtualenv -p python3 scylla-cli

pip3 install --editable .

```

## Checking cluster status

```

scli status

```

![](docs/status_demo.gif)

## Repairing Scylla Cluster

```

# connect to the cluster via 10.210.92.46 with root credentials and repair

# sync keyspace on every endpoint in local DC only

$ scli -u root -p repair sync --local

# connect to the cluster via 10.210.92.46 with root credentials and repair

# sync keyspace on 10.210.92.46 only

$ scli -u root -p repair sync --local --hosts 10.210.92.46

# connect to the cluster via 10.210.92.46 with root credentials and repair

# sync keyspace on every host except 10.210.92.46

$ scli -u root -p repair sync --local --exclude 10.210.92.46

```