The following tools are available for MAGI experiments in CREATE:
magi_bootstrap.py: Install MAGI
Used with the testbed start command,
magi_bootstrap.py installs MAGI and supporting tools on all nodes at startup.
In the topology file of the desired experiment, include the following command:
tb-set-node-startcmd $NodeName "sudo python /share/magi/current/magi_bootstrap.py"
$NodeName is the control node.
magi_graph.py: Create graphs for a MAGI experiment
magi_graph.py is a graph generator for experiments executed on CREATE using MAGI. The tool fetches the required data using MAGI’s data management layer and generates a graph in PNG format. This tool may be executed from either the Deter Ops machine or a remote computer with access to internet. The data to be plotted and other graph features are configurable.
The various command line options are as follows:
Usage: magi_graph.py [options] Plots the graph for an experiment based on parameters provided. Experiment Configuration File OR Project and Experiment Name needs to be provided to be able to connect to the experiment. Need to provide build a graph specific configuration for plotting. Options: -h, --help show this help message and exit -e EXPERIMENT, --experiment=EXPERIMENT Experiment name -p PROJECT, --project=PROJECT Project name -x EXPERIMENTCONFIG, --experimentConfig=EXPERIMENTCONFIG Experiment configuration file -c CONFIG, --config=CONFIG Graph configuration file -a AGENT, --agent=AGENT Agent name. This is used to fetch available database fields -l AAL, --aal=AAL AAL (experiment procedure) file. This is also used to fetch available database fields -o OUTPUT, --output=OUTPUT Output graph file. Default: graph.png -t, --tunnel Tell the tool to tunnel request through Deter Ops (users.deterlab.net). -u USERNAME, --username=USERNAME Username for creating tunnel. Required only if different from current shell username.
This tool expects the user to provide a configuration file. The format of the configuration file needs to be similar to the sample configuration file provided below.
graph: type: line xLabel: Time(sec) yLabel: Bytes title: Traffic plot db: agent: monitor_agent filter: host: servernode peerNode: clientnode trafficDirection: out xValue: created yValue: bytes
The configuration is divided into two parts a) Graph options and b) Database options. Graph options are used to configure the type of graph and the various labels. The database options help the tool fetch the data to be plotted.
Each record stored in the database using MAGI’s database layer has the following three fields along with any other that an agent populates.
agent: Agent Name host: Node of which the agent is hosted created: Timestamp of when the record is created
In the above mentioned example, data populated by the agent named “monitor_agent” hosted on the node named “servernode” will be fetched. The data would further be filtered on the configured values of peerNode and trafficDirection, which are agent specific fields.
Among the fetched data, values corresponding to the fields, created and bytes, will be plotted correspoding to the x and the y axis, respectively.
magi_orchestrator.py: Run MAGI
magi_orchestrator.py reads the procedure's AAL file and orchestrates an experiment based on the specified procedures. The Orchestrator does the following in this order:
- '''Joins Groups''' - The Orchestrator iterates over the list of groups and for each group sends a request to all the mapped nodes to join the group. A corresponding reply adds the replier to the group. Messages addressed to a group are sent to all the nodes that are part of the group.
- '''Loads Agents''' - The Orchestrator iterates over the list of agents and for each agent sends an agent load request to the mapped groups. An agent load message tells the Orchestrator to start an instance of the agent implementation and to put it in a listening mode, where the instance waits for further messages.
- '''Executes Event Streams''' - Next, the Orchestrator concurrently executes all the event streams listed as part of
streamstarts. The Orchestrator has a predefined event stream called
exit. The purpose of this event stream is to unload all the agents and disjoin groups. All workflows should end with executing this stream for a clean exit.
From your home directory on
users.isi.deterlab.net, run the following command:
/share/magi/current/magi_orchestrator.py --control clientnode.myExp.myProj --events procedure.aal
clientnode equals the node you want to start with
myExp is the Experiment Name
myProj is the Project Name
procedural.aal is the name of the AAL file.
The various command line options are as follows:
Usage: magi_orchestrator.py [options] Options: -h, --help show this help message and exit -c CONTROL, --control=CONTROL The control node to connect to (i.e. control.exp.proj) -f EVENTS, --events=EVENTS The events.aal file(s) to use. Can be specified multiple times for multiple AAL files -l LOGLEVEL, --loglevel=LOGLEVEL The level at which to log. Must be one of none, debug, info, warning, error, or critical. Default is info. -o LOGFILE, --logfile=LOGFILE If given, log to the file instead of the console (stdout). -e EXITONFAILURE, --exitOnFailure=EXITONFAILURE If any method call fails (returns False), then exit all streams, unload all agents, and exit the orchestrator. Default value is True -g GROUPBUILDTIMEOUT, --groupBuildTimeout=GROUPBUILDTIMEOUT When building the initial groups for agents in the given AAL, use the timeout given (in milliseconds) when waiting for group formation to complete. --nocolor If given, do not use color in output. -v, --verbose Tell orchestrator to print info about what its doing -n, --tunnel Tell orchestrator to tunnel data through Deter Ops (users.deterlab.net).
magi_status.py: Check status of a MAGI experiment, reboot MAGI daemon, download logs
check MAGI’s status on experiment nodes
reboot MAGI daemon process
* download logs from experiment nodes
This tool is run for one experiment at a time. The user needs to provide the project name and the experiment name to the tool.
This tool, by default, works for all of the nodes corresponding to the given experiment. However, it can be made to work with a restricted set of nodes, either by directly providing the set of interested nodes, or by providing an AAL (experiment procedure) file to fetch the set of desired nodes.
This tool by default only informs if the MAGI daemon process on a node is reachable or not. Specific options can be used to fetch group membership details and information about active agents.
If you want to reboot the MAGI daemon,
magi_status.py first reboots MAGI daemon processes on the experiment nodes, and then fetches their status.
If the tool is asked to download logs, it just does that, and does not fetch the status.
Usage: magi_status.py [options] Script to get the status of MAGI daemon processes on experiment nodes, to reboot them if required, and to download logs. Options: -h, --help show this help message and exit -p PROJECT, --project=PROJECT Project name -e EXPERIMENT, --experiment=EXPERIMENT Experiment name -n NODES, --nodes=NODES Comma-separated list of the nodes to reboot MAGI daemon -a AAL, --aal=AAL The yaml-based procedure file to extract the list of nodes -l, --logs Fetch logs. The -o/--logoutdir option is applicable only when fetching logs. -o LOGOUTDIR, --logoutdir=LOGOUTDIR Store logs under the given directory. Default: /tmp -g, --groupmembership Fetch group membership detail -i, --agentinfo Fetch loaded agent information -t TIMEOUT, --timeout=TIMEOUT Number of seconds to wait to receive the status reply from the nodes on the overlay -r, --reboot Reboot nodes. The following options are applicable only when rebooting. -d DISTPATH, --distpath=DISTPATH Location of the distribution -U, --noupdate Do not update the system before installing MAGI -N, --noinstall Do not install MAGI and the supporting libraries