Forum Replies Created
-
Posts
-
We need to try this on Linux I think. What distro are you using – I think we may need to spin up a VM with a similar environment.
Dear experimenters,
The update to Release 1.2 is completed. FABRIC is now open again to early users.
Dear experimenters,
The update is mostly completed at this point, however we did encounter some last minute problems we intend to resolve quickly. Please continue to use your existing slices, however we will for now keep the creation of new slices disabled to help sort out the last minute issues.
We also expect initially your projects may not always have the right permission tags to accomplish everything you need. Please report to us on the forums and we will adjust them.
Expected completion time: 2pm ET, June 7, 2022
Issue 1 above is being addressed: Maintenance on FABRIC-MAX and FABRIC-NCSA on Wed 3/16/22 – 5pm ET
Issue 2 is also being addressed transparently to users by updating IOS firmware on some of the switches.
Here is the updated topology you can expect to use. Pay attention to the key – connection types available between sites.
– 100G L1 means there is a dedicated connection between sites capable of supporting 100Gbps only carrying FABRIC traffic
– L2 means there is a shared connection between sites likely much less than 100G and subject to competing traffic from non-FABRIC sources.
– Peering locations are shown however are not available to experimenters at this time.
We suggest using virtual environments and make sure you have Python 3 (output truncated below):
$ mkvirtualenv testfabric
created virtual environment CPython3.9.2.final.0-64 in 437ms$ pip install fabrictestbed==1.0.3
Collecting fabrictestbed==1.0.3
Downloading fabrictestbed-1.0.3-py3-none-any.whl (16 kB)
Collecting fabric-orchestrator-client==1.0.3
Downloading fabric_orchestrator_client-1.0.3-py3-none-any.whl (42 kB)
|████████████████████████████████| 42 kB 1.6 MB/sHello,
The problem is you are not a member of any projects. We believe this is what is causing the notebook to fail to start up. We will work to clarify the error messages. Please have yourself added to a project and then report here if you are able/unable to login.
Could you post a fuller screenshot – you are truncating the error message. We are not seeing it for other users, so something is up with your profile.
There is nothing known in the logic of our software that requires the VM to be rebooted at 24 hours. We can test to make sure it doesn’t happen – we didn’t observe it in our recent testing related to your questions.
I don’t know that we tested reboots. I’m more surprised that NVME drive survived the reboot, than that the NIC did not. We will add ‘modify’ capability that should allow reattaching devices – this is on the roadmap.
Just to bring this up – we are going to test what happens to long-lived slices. Right now it isn’t clear if this is somehow related to your extending the slice lifetime or something happening inside the slice that cuts off access.
