FABNetv4 Connectivity Stopped Working Between Storage VM and Sites

[Arash]

I have a Storage VM at MICH  that needs to communicate with FPGA nodes at five other sites via FABNetv4. This setup was working perfectly a couple of days ago, but now the Storage VM cannot reach any of the remote nodes (0% success rate, 100% packet loss to all sites).

I deleted and created another storage slice and noticed an error during the creation of the slice. (image attached)

Has there been a recent change to FABNetv4 cross-site routing policy, or is there a known outage affecting inter-site connectivity?  Any guidance on troubleshooting or recommended architecture would be greatly appreciated.

[Tom Lehman]

There was some maintenance last night on the underlying optical infrastructure.  I will check on status are report back here.

Tom

[Arash]

Thank you, @Tom,

I still can see that the problem exists.

 

• This reply was modified 9 months, 1 week ago by Arash.

[Tom Lehman]

The FABRIC MICH dataplane link is back in service.  FABNetv4 services are working now.

[Arash]

Thank you, Tom. I still have Tout from TACC to MICH and GPN to MICH. attached images.

These notebooks were working fine a couple of days ago.

 

 

[Arash]

I also noticed the FabnetV4 gateways are not reachable from each of these sites as well

“STAR”, “WASH”, “DALL”, “LOSA”, “SALT”

[Arash]

It is fixed now. Thank you

[Author]

Posts