1. Mert Cevik

Mert Cevik

Forum Replies Created

Viewing 15 posts - 106 through 120 (of 127 total)
  • Author
    Posts
  • in reply to: lost management network connection #4189
    Mert Cevik
    Moderator

      It seems that the root cause of the connectivity issue is known then, probably you need to review your policy based routing. Do you still need anything from us on this?

      in reply to: lost management network connection #4187
      Mert Cevik
      Moderator

        Thank you for the output. I will look at this carefully later, but with a quick view, it looks like the default IPv6 route is switched to the dataplane (can be FABv6) network (see the output from ip -6 route). Management (ssh) access is routed from the ens3 interface (IPv6 2001:400:a100:3090:f816:3eff:fe5c:dd28/64 ) and default IPv6 gateway for this subnet is 2001:400:a100:3090::1. Looks like a configuration change occurred.

        in reply to: lost management network connection #4185
        Mert Cevik
        Moderator

          Hello Fengping,

          From one VM we need to see the output of  “ip a”, “ip route”, “ip -6 route”. It can be also good to see the status of the network service (network, NetworkManager whichever is active).

           

          in reply to: Node in my slice is running but cannot SSH #4157
          Mert Cevik
          Moderator

            Hello Jeet,

            Your VM encountered the same problem as you pointed out, VM was forcefully stopped. I started the VM, checked some indicators on the hypervisor. I expect that your VM will be running fine. We have some possible fixes, however we are not able to deploy at the moment.

            Also, on the FABRIC-UCSD, all VMs (yours as well as other users’) are running on ucsd-w3 and ucsd-w4. You can consider using the other ones (ucsd-w1, ucsd-w2, ucsd-w5) for VMs that require large memory.

            Please let us know if you encounter difficulties.

            Best regards,

            Mert

            in reply to: Node active but not able to ssh #4041
            Mert Cevik
            Moderator

              This problem was caused by a transient problem on the hypervisor, and VM was forcefully stopped. VM is restarted and accessible. We will work on the root cause of the problem internally, but you should be able to continue your work on the VM.

              in reply to: Loss of SSH connectivity after Debian upgrade #4028
              Mert Cevik
              Moderator

                You can use Debian 11 Bullseye image on FABRIC slices with the name “default_debian_11”.

                in reply to: Loss of SSH connectivity after Debian upgrade #4023
                Mert Cevik
                Moderator

                  We are in the process of adding Debian 11 (Bullseye) image to all FABRIC sites. We will notify when this image can be used in your notebooks.

                  Mert Cevik
                  Moderator

                    Following subnets are used for control interfaces of the VMs on all sites.

                    [1] 10.20.4.0/23
                    [2] 10.30.6.0/23

                    Update:

                    For the EDC site (special RACK for Measurement VMs) following subnets are used:

                    [1] 10.40.4.0/23
                    [2] 10.40.6.0/23

                    • This reply was modified 1 year, 2 months ago by Mert Cevik.
                    1 user thanked author for this post.
                    Mert Cevik
                    Moderator

                      Maintenance on FABRIC-MICH is completed.

                      Mert Cevik
                      Moderator

                        Maintenance on FABRIC-MASS is completed.

                        Mert Cevik
                        Moderator

                          Maintenance on FABRIC-MASS is completed.

                          in reply to: Network outage for FABRIC Central Services [RESOLVED] #3028
                          Mert Cevik
                          Moderator

                            This outage is resolved. All FABRIC services are operational and new requests can be submitted.

                            in reply to: Network outage for FABRIC-NCSA [RESOLVED] #2726
                            Mert Cevik
                            Moderator

                              This outage is resolved. VMs on FABRIC-NCSA rack are reachable from public internet. Please let us know if you encounter any problems.

                              in reply to: Network outage for FABRIC-NCSA [RESOLVED] #2652
                              Mert Cevik
                              Moderator

                                We are working with the hosting campus for a known network issue that may require part replacement. No ETA is available, updates will be posted when the problem is resolved.

                                in reply to: Create Nodes with IPv6 address #1524
                                Mert Cevik
                                Moderator

                                  If you create an example slice to reveal the problem, we can work on it to understand better, suggest solutions, consider changes on public ACLs. Please let us know with the name of the slice and FABRIC site, if this sounds good for you.

                                Viewing 15 posts - 106 through 120 (of 127 total)