Forum Replies Created
-
Posts
-
Hi Komal,
Thank you so much for your help. Currently, I am using eight tunnels, each at 20 Gbps. Since we observed some issues at that capacity, we decided to reduce them to 10 Gbps (eight tunnels at 10 Gbps). I believe the correct way to update my slice is to remove the existing tunnels and then add new ones with the lower bandwidth. I will work on making that change today, and I hope I can reserve them for a longer duration.
Thank you.
Best regards,
Fatih Berkay Sarpkaya
Hi Komal,
Thank you for the detailed explanation. Last time, I followed your suggestion and re-reserved the affected tunnels, which extended them until December 5. Today, I renewed the slice to December 16, but three tunnels showed errors and I believe they will still expire on December 5.
Given my ongoing work, I need these tunnels to remain available throughout this month. What would you recommend as the best approach to keep them active as long as possible? Also, I don’t think fablib.list_links() exposes the reservation start/end times, so I am not able to check the future reservation durations of my existing tunnels.
If there is a way for me to view the currently reserved time windows of my L2PTP links, I could try reserving them for a longer period in advance (although I’m not sure if that is appropriate for an already existing slice.)
Any guidance on how to manage these renewals effectively would be very helpful.
Thank you,
Fatih Berkay Sarpkaya
Hi Komal,
Thank you for your response. I created a new slice a few days ago and extended its expiration date to December 5 right away. The tunnels were working correctly for the last 2–3 days. However, when I tested again today, I no longer have connectivity over the tunnels with VLAN tags 100, 600, and 800.
Could you please check my slice again and confirm whether these tunnels have entered the Closed state? If so, could you help me understand the exact reason? My understanding was that once the slice is reserved until a certain date, the associated tunnels should remain reserved as well.
Additionally, could you please provide guidance on how to properly “re-declare” or “re-submit” these network services within the slice, if that is the correct approach?
slice id= 698e8e21-3926-451f-ad94-24ea5db4a822
Thank you very much for your help.
Best regards,
Fatih Berkay Sarpkaya
Dear Komal,
Thank you very much for your detailed explanation. It was very helpful for me.
Best regards,
Fatih Berkay Sarpkaya
Dear Komal,
Thank you so much for your quick reply. As I understand it, the term “host” refers to the same physical machine, and it is not distributed across multiple blades, correct?
For example, in the FABRIC resources, a single host is shown as having a total of 128 cores. Does this number represent 128 physical cores on one physical server? When I check inside a VM on that host, it reports: “AMD EPYC 7543 32-Core Processor “. This makes me think that the machine has 32 physical cores, which raises a question: is the “128 cores on the host” total coming from four separate machines (32 × 4)?
So, to confirm, you are saying that this interpretation is not correct, and that a “host” refers to one single physical machine. If so, could you please clarify that 128 cores and “AMD EPYC 7543 32-Core Processor ” difference?
Thank you very much for your clarification.
Best regards,
Fatih Berkay Sarpkaya
