Maintenance on multiple sites – Mar 12

[Mert Cevik]

Dear Experimenters,

We will perform a maintenance on the sites below on Mar 12 to complete the flashing of the FPGAs. 

– STAR, TACC, MICH, DALL, SALT, UCSD, LOSA, KANS, RUTG, SRI

This will require cold-rebooting of the FastNet Servers that have the FPGA boards on them. VMs on these workers will be restarted after the rebooting, and all PCI devices will be re-attached to them. However, IP configurations on the dataplane interfaces will have to be done from inside the VMs (or via relevant fablib_api calls) by the experimenters. 

We set a pre-maintenance status on the following workers that will be visible on the portal.

– star-w3
– tacc-w3
– mich-w2
– dall-w2
– salt-w2
– ucsd-w3
– losa-w2
– kans-w2
– rutg-w3
– sri-w2

 

Active slices that have VM slivers on the affected servers are as follows:

– iPerf3 UDP slice(4677f196-d923-42d4-85e0-284af52f0977)
– poseidon-experiment(08083875-6f51-45e3-b652-642d252060b6)
– TACC_slice(bb1b6b43-87f9-4a11-9315-0204c90dcc5f)
-TACC_slice2(2e643024-3983-49e7-9003-7540e0379363)
– Pili_slice(51139a33-be23-4d73-aa40-6b7832c31281)
– TACC_slice(bb1b6b43-87f9-4a11-9315-0204c90dcc5f)
– Proposal1(efd59716-05bd-43c2-83ff-d9d1b1c49241)
– dkumar16_assignment4(b1d76776-4c94-4a53-a80b-a493b619fef2)
– asathwara_assignment4(397dff3a-2160-43e3-828f-31ed187b111d)
– poseidon-experiment(08083875-6f51-45e3-b652-642d252060b6)
– ServiceX_NDN_Proxy_tcp_4nodes(61c1d373-b3b6-472f-a34a-3dd6dac8c809)
– hydra_update_2(8106d9b1-0d08-43bc-90f7-10487a5f902b)
– p4sec_fabric(91d93314-0e1e-45d1-a545-d9ecd5ab4de7)
– vamsi_new_slice(ab2d5e5a-6511-46e7-9931-b7c747e3ebaa)
– PoSeiDon-TCP-Experiment-ConnectX6(a1d0db27-aef6-445e-90fc-75794b33fdd7)
– Server22(df4728b6-e830-408d-b426-aff3184c8f08)
– latency_monitoring_slice(bf124347-9e91-4ac2-9bcd-a3623b0a3d5e)
– Server22(df4728b6-e830-408d-b426-aff3184c8f08)
– backplane(b6fdd9d7-d978-45d0-85d4-55f4492db7d4)
– four_UCSD(d0ef035e-5443-4c2e-8f9f-ae59e7f5c0fb)
– hmcbz1(eb533b6b-b23b-4b12-aaa9-44efeab0f642)
– ServiceX_NDN_Proxy_tcp_4nodes(61c1d373-b3b6-472f-a34a-3dd6dac8c809)
– federated-learning-experiment(a06b00ff-8f13-42cd-af46-6d6a1dfa44da)
– MySlice(8d3d7cef-9911-440b-af9d-f963f96c3510)
– exerc_1(1f519809-edf5-4d53-91ab-1afaa110f492)
– Fabric_Demo(5031cc9a-3985-47fe-8470-12dd7b76ad7d)
– mininet@1709250873(b0ea1bd2-e62c-4fa8-8a92-091c6814f2c8)
– asathwara_assignment4(397dff3a-2160-43e3-828f-31ed187b111d)
– PoSeiDon-TCP-Experiment-ConnectX6(a1d0db27-aef6-445e-90fc-75794b33fdd7)
– test-slice-240306a(372d892b-1794-4452-a108-251cdcb657b2)
– MySlice(82d33b63-e870-4c64-a8b2-1c8c07ddedd3)
– owl-demo1-test-slice-240301a(ba521b12-57fa-4b00-8eba-3f7ed398b7ff)
– dkumar16_assignment4(b1d76776-4c94-4a53-a80b-a493b619fef2)
– MySlice11(5cbf7492-ee0f-47ab-807b-c910994c1503) 

 

We apologize for the inconvenience for your experiments before KNIT8, however we have to complete this work as part of the preparations. 

[Mert Cevik]

Dear Experimenters,

This maintenance is still in progress.  We have to add MASS to the list of the sites for mass-w2.

Following slices have slivers that will be affected as described on the previous post:

– MASS-slice1(52bda536-6362-46a3-85e4-cc6fdcefb1d6)
– eibp_tes333t(6e5a7637-5bad-425f-986b-56da40b12f96)
– MySlice(719a5bda-e3f5-477b-9122-7c10383ff1df)
– ra-upf(d5e2a22d-7465-49a6-a861-ab6df923bc7e)
– smannuru_assignment4(eb23ac03-5a75-4057-8ffc-bec1755aeed0)
– OCTFAB(59be1a6b-dcd7-4d4f-9501-1f6cd72875b2)
– Slice1(12a3bc93-60bf-416b-b341-a8b611f0e7ee)

We will send updates when the maintenance is completed.

[Mert Cevik]

Dear Experimenters,

This maintenance is completed, all sites are available for experiments. VM slivers of the current slices have their PCI devices re-attached. Missing IP addresses on the dataplane interfaces should be restored from inside the VMs or available fable_api calls.

 

[Author]

Posts