0

enter image description here

There is a multi-ring topology,combined with 8 Hubs and 2 switches,i want to ask two questions

1. If H3 is damage,but the fiber is still ok,will H2 damage too in the future? i have a thinking to explain the reason,but i don't know it is right or wrong.

Reason:because H3 is damaged,so H2 will keep sending message,m1, to H3 ,but H2 needs to send lots of other messages to H3 in the future,so this situation will cause too many messages need to be sent by H2,and it will let H2 be crashed,is this thinking right?

2. The hubs are JetNet 4006f ,and 4006f provide two redundancy technologies,rapid dual homing and rapid super ring,and my friend tells me that i have to disable the rapid dual homing ,just like the picture below,my friend and i don't know why,does anyone know the reason?i have a thinking too,but i don't it is right or wrong.

Reason:rapid dual homing and rapid super ring are both redundancy technology,so i have to choose one of them,that is,if i choose the rapid super ring as the version,i can't enable the rapid dual homing.

enter image description here

0

Your "hubs" JetNet 4006f are actually switches as well. "Hub" most commonly refers to repeater hubs only.

If H3 is damage,but the fiber is still ok,will H2 damage too in the future?

No. It is next to impossible for any switch to damage a neighbor switch, especially over fiber.

because H3 is damaged,so H2 will keep sending message,m1, to H3

No. If H3 fails, the H2-H3 link is down and no more message are passed.

For your question about proprietary rapid dual homing vs rapid super ring: I can't find decent documentation at Korenix. There are standards/drafts Ethernet Ring Protection Switching (ERPS) by ITU-T and Ethernet Automatic Protection Switching (EAPS) by IETF/ExtremeNet but it's unclear whether Korenix relates to those.

Generally, you should consider using a tree instead of (a) ring(s). With dual core switches and dual uplinks for the access layer (which I guess the rings are for), you'd have redundant uplinks, a redundant core, and absolutely no interference between access switches. Also, you could use standard RSTP/MSTP for redundancy or possibly MC-LAG (proprietary, but core-only).

4
  • If there are 30 switches,and 5 of them are damaged in the short time,what do you think the reason of this situation? Sep 29 at 11:55
  • You're having an actual issue with serial failure? I'd venture to point at the power but it could also be a production issue (one more reason to stay vendor-independent).
    – Zac67
    Sep 29 at 12:01
  • yes,i am having an actual issue with serial failure in the short time,we have checked the power,and the power is ok,and i also think the reason is the production issue,but i am not sure about that,and i hope my boss can accept this production issue Sep 29 at 12:22
  • I'd generally rule out a domino effect and seriously talk to the vendor.
    – Zac67
    Sep 29 at 12:30

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.