Forum Replies Created
-
Posts
-
Well, one observation is that Paul said so elsewhere in the forum.
L2Bridge: These bridges are like a local network switch/bridge that connects any number of local nodes within a single site. These local bridges are directly connected to the nodes so your bandwidth will be limited by the maximum bandwidth of the NICs that you are using (i.e. ConnectX_6 NICs will provide 100Gbps). This bridge is not programable and only performs simple MAC learning. The key use of these bridges are that they can only connect to nodes within a single FABRIC site.
Another observation: suppose I create 4 VMs with a basic NIC on each, and connect each basic NIC to an L2Bridge-type network. I capture traffic on each of the NICs with tcpdump. A frame sent by host 1 with host 2’s address as the destination MAC only appears at the NIC on host 2, and not on host 3 or host 4.
Another observation: suppose I create a Linux switch connecting multiple hosts, using a L2Bridge between my Linux switch and each of the connected hosts. (as in e.g. this example.) Non-broadcast frames sent from the hosts don’t make it to the Linux switch interfaces, so the bridge does not work.
What I am trying to do: I am trying to connect multiple basic NIC interfaces with a network link, so that any frame sent by any NIC on the link appears at every other NIC on the link. Like the way an Ethernet segment behaves, or Ethernet segments connected by a hub, or Ethernet segments connected by a switch with MAC learning disabled.
Oh, I see.
Probably a good idea to mention in the upgrade announcement, if existing users are supposed to re-run the configuration notebook. The announcement just said “minimal impact to the existing notebooks” 😉
(Actually, a changelog that lists all the user-facing changes to fablib would be very helpful when there is an upgrade.)
Will the official “Configure your Environment” notebook be updated to reflect this new recommendation?
Thanks!
> If you have collaborators/students who need to use Google accounts, we need the project lead to provide a list of users/Google accounts that we should approve and we will approve them as they come in.
What is the process you would prefer for this? is there a support email we should send these emails to?
Also: I have a collaborator who just tried to join, whose institution was not on the list, so I think she did it through NCSA since she had an NCSA account from previous GENI use. Do I need to do anything special to get her approved? (Not sure where NCSA shakes out in all this.)
> One other possibility is if an institution is not listed on the CIlogon page but the institution uses Google as a email provider.
Unrelated question: if someone
1. first uses FABRIC with institution that is listed CIlogon page
2. then tries to use FABRIC with Google with institutional email
will this be one FABRIC account since it is tied to the same email, or two separate FABRIC accounts?
Thanks for the update!
It would be useful if the sphinx docs for this function (and similar ones) would briefly mention what they do “under the hood” in cases where it’s not obvious.
For example, I have some teaching experiments that remove the default route (first replacing it with a more specific routes to maintain SSH access) in order to show what happens when there is no route to a particular destination. I also have some research experiments that change the default route in order to route most Internet traffic through another host in the testbed topology. So it would be useful to know when I should expect
get_management_os_interface()to fail. (Even besides for old software versions.)FYI – the 504 Gateway Time-out error is still happening for me when I create a new topic on this forum.
Also, both today (on this post) and yesterday, when I create a new topic on this forum, it takes a long time to load and then times out with nginx error 504 Gateway Time-out. (It does still create the topic.)
I think it’s easier if all of the get_x functions behave in a similar way. What happens with add_node/get_node or add_network/get_network if you try to use the same name more than once?
Thanks. I am somehow under the impression that right now, the “sliver key” in the FABRIC portal doesn’t do anything – is that right or did I misunderstand that? So for the moment, the only key installed in a sliver is one from the Jupyter environment, and I can’t ssh/scp from my laptop.
Thanks!
Will we be able to:
* access data in persistent storage directly from the Jupyter environment, without necessarily having an active slice?
* use something like fablib download_file to transfer a file that is too big for “work” directly from a FABRIC node to the persistent storage?
* access the persistent storage space directly (without having to copy stuff to “work”) to do data analysis in the same notebook as the rest of the experiment?
