A set of test scripts for verifying Erasure Coding vs Replication.
- kill
- killAll
- status
- capture
- startInt
- killInt
- put
- get
- config
- configAll
- list
- create
- createAll
- deleteAll
- listAliases
- createAliases
To install run the following commands:
$ git clone git@github.com:rayje/ECvR.git
$ cd ECvR
$ npm install
The previous commands will clone the repository and install the node dependencies for the scripts.
The scripts contained in this repo require a config file named config.json that exists in root directory of the ECvR project.
The config file is expected to contain the following format:
{
"futuregrid": {
"host": "host at futuregrid.org",
"privateKey": "/path/to/private/key/file",
"username": "username on future grid",
"passphrase": "passphrase for keyfile"
},
"stor": {
"privateKey": "location of stor private key",
"username": "ubuntu"
},
"ec2": {
"accessKeyId": "AWS Access Key Id",
"secretAccessKey": "AWS Secret Key",
"region": "us-west-2"
},
"isEC2": true
}
The following is a description of the scripts contained in this repo.
Gets information of Future Grid or EC2 instances associated with the Stor project. This
script will write the result in JSON format to a file named instances.json.
$ ./getInstances --help
Usage: getInstances [options]
Options:
-h, --help output usage information
-V, --version output the version number
--imageId <imageId> The image id used to filter the search. (FutureGrid)
--keyName <keyName> The keyName used to filter the search. (FutureGrid)
NOTE: When run on EC2, this scripts takes no options.
To run the script on Future Grid:
$ ./getInstances --imageId <imageId> --keyName <keyName>
Note: imageId is required for Future Grid
To run the script on EC2
$ ./getInstances
Reads and displays the content found in the file instances.json in tabular format.
$ ./readInstances
instanceId | imageId | keyName | instanceType | ipAddress | privateIpAddress |
----------- ------------- ------- ------------ ---------------- ----------------
i-000xxxxx | ami-000xxxxx | my-key | m1.small | xxx.xxx.xxx.xxx | xxx.xxx.xxx.xxx |
i-000xxxxx | ami-000xxxxx | my-key | m1.small | xxx.xxx.xxx.xxx | xxx.xxx.xxx.xxx |
i-000xxxxx | ami-000xxxxx | my-key | m1.small | xxx.xxx.xxx.xxx | xxx.xxx.xxx.xxx |
Sets up tahoe on all nodes. This script will run all the required commands to
get tahoe setup on all nodes. This scripts expects the instances.json
file to have been create with the getInstances script.
$ ./tahoe-setup <introducer-ip>
Tears down all tahoe nodes. This command will remove the setup created from the
tahoe-setup script. This scripts expects the instances.json
file to have been create with the getInstances script.
$ ./tahoe-teardown <introducer-ip>
A script to run stor commands on a remote server.
$ ./stor --help
Usage: stor [options]
Options:
-h, --help output usage information
-V, --version output the version number
--start <instanceId> Start the stor server on a specific instance
--startAll Start the stor server on all instances
-p, --ringServer <ipAddress> The ip address of the Pastry ring server
-m, --capacity <capacity> The storage capacity for the Stor server [Default: 10]
-d, --storage <directory> The storage directory for the Stor server [Default: $HOME/Stor_Age]
-r, --replication <rep> The replication factor setting [Default: 5]
--stop <instanceId> Stop the stor server in a specific instance
--stopAll Stop the stor server on all instances
-e, --excludes <instanceIds> A comma delimted list of excluded instances
--stopN Stop N instances, reads from instances-ips.json
-n, --numInstances <numInstances> The number of instances to stop
-i, --startIndex <index> The index to start the stop within the stop file (instances-ips.json)
--status Displays the status of Stor on all instances
--debug Run in debug mode
--put <instanceId> Execute the stor client PUT command on the specified instanceId
--filePath <filePath> File path for the file to save in Stor
--get <instanceId> Execute the stor client GET command on the specified instanceId
--fileKey <fileKey> File key as returned by the PUT command
--getUsage Get file system space usage for all instances
--createStopFile Creates a file that can be used with the stop command (instances-ips.json)
Start the Stor server on a single instance.
Example:
$ ./stor --start 192.168.1.100 -p 192.168.1.100 -m 10 -r 5
or
$ ./stor --start i-00004545 -p 10.1.2.123
Starts all stor servers.
This command will read the addresses of the instances found in instances.json and run the start stor
command for each instance minus the ring server.
Note: It is expected that the ring server already be started before this script runs. This script WILL NOT try to start the ring server.
Example:
$ ./stor --startAll -p 10.1.2.123
Stop the Stor server on a single instance.
Example:
$ ./stor --stop 192.168.1.100
or
$ ./stor --stop i-00004545
Stops all the Stor servers found, minus those specified with the excludes flag.
Example:
$ ./stor --stopAll -e i-00001234
or
$ ./stor --stopAll -e <public address>
Stops the Stor server on N instances.
Note: This command requires that the createStopFile command was run at least once
and there exists a instance-ips.json file in the same directory as instances.json.
Arguments:
-i The index of the first instance
-n The number of instances to stop
Example:
$ ./stor --stopN -i 0 -n 10
The above command will read the instance-ips.json file and read the first 10 ip
addresses, and run the stor -k command on those instances.
Executes the PUT command on the instance of the Stor server identified.
Example:
$ ./stor --put i-00001234 --filePath /tmp/A
Executes the GET command on the instance of the Stor server identified.
Example:
$ ./stor --get i-00001234 --fileKey KEY
Display the status of the stor server on all instances
Example:
$ ./stor --status
instanceId | keyName | ipAddress | privateIpAddress | running |
------------ -------- ---------------- ---------------- -------
i-000xxxx1 | my-key | xxx.xxx.xxx.xx1 | xxx.xxx.xxx.xx1 | true |
i-000xxxx2 | my-key | xxx.xxx.xxx.xx2 | xxx.xxx.xxx.xx2 | false |
i-000xxxx3 | my-key | xxx.xxx.xxx.xx3 | xxx.xxx.xxx.xx3 | false |
Display the file system usage of for all instances
Optional parameters:
-e, --exclude xxx.xxx.xxx.xx1,xxx.xxx.xxx.xx2
-d, --storage <storage directory>
Example:
$ ./stor --getUsage
xxx.xxx.xxx.xxa - Ok - 12345
xxx.xxx.xxx.xxb - Err - Error info
Creates a file named instance-ips.json that consists of the public addresses
of all instances.
Note: This command will shuffle the addresses, so they will not appear
in the same order as they appear when the readInstances or getInstances
scripts are run.
This command also takes the -e option which is a comma delimited list of public
ip addresses to exclude from the list of addresses generated.
Example:
$ ./stor --createStopFile -e xxx.xxx.xxx.xx1,xxx.xxx.xxx.xx2
A script to run commands on a tahoe node
$ ./tahoe --help
Usage: tahoe [options]
Options:
-h, --help output usage information
-V, --version output the version number
--kill <nodeIp> Stop a storage node
--killAll Stop all storage nodes
-e, --excludes <instanceIds> A comma delimted list of excluded instances
--status Displays the status of Tahoe on all instances
--debug Run in debug mode
--capture Capture storage measurement
--startInt <nodeIp> Start an introducer node on the specified instance
--killInt <nodeIp> Kill an introducer running on the specified instance
--startAll <introducerNodeIp> Start tahoe storage nodes on all but the introducer node
--put <nodeId> Put a file on tahoe.
--filePath <filePath> The path to the file used for the put command.
--get <nodeId> Get a file from tahoe
--filename <filename> The name of the remote file
--config <nodeId> Set a tahoe configuration
--configAll Set a tahoe configuration on all nodes
-s, --storage <size> reserved_space
-n, --needed <needed> shares.needed
-t, --total <total> shares.total
-p, --happy <happy> shares.happy
-x, --port <happy> tub.port
-l, --listConfig Display the current config value
--list <nodeIp> List the files stored on the tahoe node
--create <node> Creates a storage node, if not already created
--createAll Create a storage node on all instances
--deleteAll Delete all storage nodes from all instances
--listAliases List aliases on all nodes
--createAliases Creates the tahoe alias on all nodes, if it does not exist
Stop the tahoe storage node on a single instance.
Example:
$ ./tahoe --kill 192.168.1.100
or
$ ./tahoe --kill i-00004545
Stops all the storage nodes found, minus those specified with the excludes flag.
Example:
$ ./tahoe --killAll -e i-00001234
or
$ ./tahoe --killAll -e <public address>
Display the status of the tahoe storage nodes on all instances
Example:
$ ./tahoe --status
instanceId | keyName | ipAddress | privateIpAddress | running |
------------ -------- ---------------- ---------------- -------
i-000xxxx1 | my-key | xxx.xxx.xxx.xx1 | xxx.xxx.xxx.xx1 | true |
i-000xxxx2 | my-key | xxx.xxx.xxx.xx2 | xxx.xxx.xxx.xx2 | false |
i-000xxxx3 | my-key | xxx.xxx.xxx.xx3 | xxx.xxx.xxx.xx3 | false |
Display the size of the contents within the .tahoe/storage directory on all instances
Example:
$ ./tahoe --capture
instanceId | keyName | ipAddress | privateIpAddress | storage |
------------ -------- ---------------- ---------------- -------
i-000xxxx1 | my-key | xxx.xxx.xxx.xx1 | xxx.xxx.xxx.xx1 | 31960 |
i-000xxxx2 | my-key | xxx.xxx.xxx.xx2 | xxx.xxx.xxx.xx2 | 0 |
i-000xxxx3 | my-key | xxx.xxx.xxx.xx3 | xxx.xxx.xxx.xx3 | 31959 |
Starts an introducer node on the IP address of the instance
--startInt <nodeIp> Start an introducer node on the specified instance
Example:
$ ./tahoe --startInt xxx.xxx.xxx.xx1
Terminates the introducer node on the IP address of the instance, also deletes the .tahoe-int directory
--killInt <nodeIp> Kill an introducer running on the specified instance
Example:
$ ./tahoe --killInt xxx.xxx.xxx.xx1
Put a file on tahoe. This command will copy a local file to the tahoe node and run
the tahoe put command on the node.
Example:
$ ./tahoe --put i-000xxxx3 --filePath /tmp/test.txt
The put command expects the --filePath flag to be a local file path.
NOTE: The --filePath is required.
Gets the contents of a file that was previously PUT on a tahoe node.
Example:
$ ./tahoe --get i-000xxxx3 --filename test.txt
The get command expects the filename flag to be the name of a file that
was previously PUT on a tahoe node.
NOTE: The --filename is required.
Sets a value in the tahoe.cfg file on a storage node.
This command supports the following flags to set the related config values:
- -s reserved_space
- -n shares.needed
- -t shares.total
- -p shares.happy
- -x tub.port
Example:
$ ./tahoe --config i-000xxxx3 -s 1G
The above example will set the following config value:
reserved_space = 1G
The config command also supports the following helper flags:
--debug Outputs debug content
-l Displays the remote config value without setting it
The same as Config, except applies selected configuration to all nodes
Example:
$ ./tahoe --configAll -s 1G
Lists the files on a tahoe node
Example:
$ ./tahoe --list i-000xxxx3
List: xxx.xxx.xxx.xx1
---------------------
my_test.txt
list.out
Creates a storage node, if not already created. This command will run the
tahoe create-node -d .tahoe tahoe command on a single node.
Example:
$ ./tahoe --create i-000xxxx3
Create a storage node on all instances. This will run the above --create
command on all nodes.
Example:
$ ./tahoe --createAll
Delete all storage nodes from all instances. This command will remove the
.tahoe directory from all nodes.
Example:
$ ./tahoe --deleteAll
List aliases on all nodes. This command will run the tahoe list-aliases
command on all nodes.
Example:
$ ./tahoe --listAliases
Creates the tahoe alias on all nodes, if it does not exist. This command will
run the tahoe create-alias tahoe command on all nodes.
Example:
$ ./tahoe --createAliases