[PHILIP CUMMINS]

ok i will close this now, thanks for the help

[PHILIP CUMMINS]

yes it works after making the changes you suggested

[PHILIP CUMMINS]

hi Charles, were you able to complete the grafana/node exporter example?

I can’t see how this can be an issue with the prometheus configuration

[PHILIP CUMMINS]

I forgot to extend the slice, sorry.

are you able to follow the example?
jupyter-examples-rel1.8.0/fabric_examples/complex_recipes/grafana_prometheus_servers

[PHILIP CUMMINS]

Komal Thareja wrote:

#Create Slice slice = fablib.new_slice(name=slice_name) # Network net1 = slice.add_l2network(name=network_name, subnet=IPv4Network(“192.168.1.0/24”)) # Node1 node1 = slice.add_node(name=node1_name, site=site) iface1 = node1.add_component(model=’NIC_Basic’, name=’nic1′).get_interfaces()[0] iface1.set_mode(‘auto’) iface1.set_bandwidth(10) net1.add_interface(iface1) # Node2 node2 = slice.add_node(name=node2_name, site=site) iface2 = node2.add_component(model=’NIC_Basic’, name=’nic1′).get_interfaces()[0] iface2.set_mode(‘auto’) iface2.set_bandwidth(10) net1.add_interface(iface2) #Submit Slice Request slice.submit();

even following your example and using an L2 network and creating/adding the interface, I am unable to set the bandwidth. After the slice has been created I still see the 100 bandwidth for the interface.

• This reply was modified 3 weeks, 6 days ago by PHILIP CUMMINS.
• This reply was modified 3 weeks, 6 days ago by PHILIP CUMMINS.

[PHILIP CUMMINS]

if it helps anyone from the future:

– I am removing the bootstrap nodes before starting IPFS daemon

– I am adding only the enp7s0 internal interface IP address to the announce list

It should work exactly as you expect, once i’ve progressed a little further i’ll try to remember to update this with a link to the jupyter notebook to get the code

[PHILIP CUMMINS]

MySlice ID: 8b22983e-2165-483e-a37b-7c75df1b5804

GrafanaServer ID: c2ef6dc8-c05f-4d14-be93-f86457e59016

[PHILIP CUMMINS]

i am following:
jupyter-examples-rel1.8.0/fabric_examples/complex_recipes/grafana_prometheus_servers

I have the two slices open: GrafanaServer and MySlice.

the prometheus.yml looks as expected but i don’t see the data source.

as i am in EU, i probably won’t have the SSH tunnel open when you check.

Thanks,

Phil

[PHILIP CUMMINS]

is this supported by fabnet?

when creating the slice I do:

node = slice.add_node(name=f”{swarm_node_name_prefix}-{idx+1}”, site=site, disk=disk, image=image)
node.add_fabnet()
iface = node.get_interface(network_name=f’FABNET_IPv4_{node.get_site()}’)
iface.set_mode(‘auto’)
iface.set_bandwidth(30)

after I am trying to confirm the new bandwidth:

for idx in range(swarm_node_count):
node = slice.get_node(f”{swarm_node_name_prefix}-{idx+1}”)
iface = node.get_interface(network_name=f’FABNET_IPv4_{node.get_site()}’)
iface.set_mode(‘auto’)
iface.set_bandwidth(30)
print(iface.get_bandwidth())
nodes.append(node)

but I see that the bandwidth is 100 for the interface, any suggestions? thanks!

[PHILIP CUMMINS]

and if the devices are not able to communicate with the outside world then the same probably goes for pulling images from docker hub, installing with apt…

maybe an FAQ section here could be very useful: https://learn.fabric-testbed.net/knowledge-base/network-services-in-fabric/

thanks,

[PHILIP CUMMINS]

hi Komal, thanks for the information,

I understand from reading about network services that FABNetv4 is “An IPv4 service with RFC1918 addressing which can connect any number of interfaces across any number of sites. Communication with the outside world (public Internet) is not possible.”

is this used when executing node.add_fabnet() in the Fabric hosted Jupyter notebook?

I can see that FABNetv4Ext can communicate with the outside world, is this the default when executing node.add_fabnet()?

The reason I ask, I am sorry to report is that I can see on start up IPFS opening what look like connections with the outside world. I stop the docker container without interacting with the network. I can lock down the connections using a swarm key, but this results in double encryption which may affect the testing I am doing.

Overall though I am really excited with the experience of using the fabric testbed, and just wanted to say good job!

[PHILIP CUMMINS]

deleting the slice, re-running the configuration and validation produces this node list:

ID	Name	Cores	RAM	Disk	Image	Image Type	Host	Site	Username	Management IP	State	Error	SSH Command	Public SSH Key File	Private SSH Key File
4ba39cec-e226-4c75-b905-07abfa96f2b8	Node1	2	8	10	default_rocky_8	qcow2	max-w1.fabric-testbed.net	MAX	rocky	2001:468:c00:ffc4:f816:3eff:fe6f:3234	Active		ssh -i /home/fabric/work/fabric_config/slice_key -F /home/fabric/work/fabric_config/ssh_config rocky@2001:468:c00:ffc4:f816:3eff:fe6f:3234	/home/fabric/work/fabric_config/slice_key.pub	/home/fabric/work/fabric_config/slice_key

the experiment now runs as expected. Hello, Fabric!

Thanks!

[Author]

Posts