Recommended next steps by onsite:

They missed some settings on their end, IPv4 is up but IPv6 still down. Followup sent to email thread.

Edit addition.

Cathal and I chatted about this in IRC so I'm summarizing my points here for task visibility.

I've sent a followup to the migration thread with the vendor, cc'd in all of NetOps, to determine whats up.

Please double-check and fill out the racking details section of this racking task.

Rob Halsell,

Your case # 01207593: Remote hands for IP Transit migration from copper to fiber has been updated with the following:

"Created by: mmariscalmata Fibers have been swapped. TX/RX has been properly aligned."

Ok, confirmed the parts have arrived.

I've made some assumptions regarding hostnames and networking. Please review and also provide racking deails.

In the racking details for these 3 new hosts you listed hostnames dumps100[123], but we already have existing 'clouddumps100[12]' so shouldn't these be clouddumps100[345]?

In T438952#12357353, @wiki_willy wrote:

@RobH - can you request quote for the disk? Since we're in the middle of figuring out whether we should extend out the server life cycle for things, it might be good to just buy the disk for this one.

In T438952#12354980, @BTullis wrote:

Oh, OK. I didn't realise that the warranty had expired. We could survive without it, but a replacement would be nice.

Please review and update this rack task. This task requires update for all hostnames, racking location, and checklist entries for hostnames.

Please review and update this rack task. This task requires update for all hostnames, racking location, and checklist entries for hostnames.

Please update the site.pp file with the insetup role for your team (detailed on https://wikitech.wikimedia.org/wiki/SRE/Dc-operations) and add the new servers to preseed.yml for partition info.

Please update the site.pp file with the insetup role for your team (detailed on https://wikitech.wikimedia.org/wiki/SRE/Dc-operations) and add the new servers to preseed.yml for partition info.

I've moved the ordering tasks along as they were going to expire if not placed to order, and I've now taken some assumptions on hostnames to create the racking task. Please review the above information in the task description and correct as needed.

I've moved the ordering tasks along as they were going to expire if not placed to order, and I've now taken some assumptions on hostnames to create the racking task. Please review the above information in the task description and correct as needed.

Please double check the racking details for this, as I made assumptions based on previous gpu hosts.

Please update this task description, checklists, and title with the new hostnames, as well as the racking details section.

In T418012#12309116, @cmooney wrote:

In T418012#12309022, @RobH wrote:

• All switches except the lsw and ssw in A8 are racked, connected to SCS

Can we confirm what ports the lsw and ssw in rack A1 are connected to? Port 17 - labelled asw-a1-eqiad - does not seem to have anything connected? Plus no port on the scs is labelled ssw1-a1-eqiad, and that switch in Netbox shows nothing connected to its console port.

The remaining switches, apart from those in A8, are reachable from management and configured.

IRC Sync update:

Turns out the reply may not have included all the right parties so I did another reply today and included the additional parties on the DR side.

@ssingh: The new memory is dectected, same speed as the old just I can compare the serial output to the photos, throwing the photos here for reference in future. New memory installed, 2 spare dimms left to use, and the old memory set aside in case it wasn't the issue. You can resolve this task once you are aware, thanks!

This had remote hands update that it was done, double checking the work now.

Pushed SAS-RAID_Firmware_2WJX5_WN64_52.38.0-6897_A13_01.EXE into the R660 storage directory on cumin1003. should be able to run and see it there and on all other cumin hosts after the nightly rsync.

Dear customer,

We acknowledge receipt of your case and will proceed to complete your requested task within the timeframe you specified. We will update you upon completion or reach out if we have any questions.

Best Regards,

remote hands update:

There is a 48 hour minimum window to have remote hands work on things without an urgency fee applied. I picked Friday with a note that the work can actually take place at any time. Typically they'll read that and just get to it sometime in the next day or so.

The ram landed yesterday, filing remote hands shortly to work on this host.