ubuntu-phone team mailing list archive

[Evan Dandrea <evan.dandrea@xxxxxxxxxxxxx>]

Good morning, everyone. The 1SS move continues. The nitty gritty details
can be found below. Larry's goal is to have everything powered on and
networked by the end of business today (US Eastern).

---------- Forwarded message ----------
From: *Larry Works*
Date: Tuesday, November 12, 2013
Subject: Lab update from 11 November
To: Evan Dandrea <evan.dandrea@xxxxxxxxxxxxx>


Evan,

The networking configuration stopped shortly after the switches were
stacked and the link in was connected. Brian didn't like the way the
racks were populated and thought all of the CDUs needed to be flipped
endwise so the input power cables were going straight up to the top of
the racks instead of being looped up from the bottom. I uninstalled and
reinstalled all of the CDUs while he and Rick started shifting some of
the servers around within the various racks. After a bite of lunch I
headed back to Lex to pick up the last of the hardware (the 4 new Intel
SDPs for the kernel team, our 2 DMZ systems and IS' gateway server for
the DMZ. When I got back from that a more extreme shifting of hardware
was ongoing. I couldn't stay to see how extensive as I had to get the
rental back to the airport. When I got back from the rental turn in most
of the re-racks had been completed. I helped finish up the last 18 to 20
systems and then proceeded to rack most of the hardware that was just
brought in from Lex.

The end result of the day was:
  - The switches are all in place and the Cisco switches are stacked.
  - The network link for our internal network has been run.
  - All of the big hardware is at 1SS.
  - All of the server and test systems at 1SS are now racked.
  - The only hardware remaining at Lex are the phones/tablets and the 4
Veritons used for bootspeed testing (those probably won't get brought
over until Wednesday).
  - Brian was going to run the network link for the openstack suite
(since he will not be onsite on Tuesday) and maybe the DMZ link (if he
gets their gateway server racked).

For Tuesday, and I made this clear, networking in the systems is the
first order of business and the primary operation for the day. I will
get all of the infrastructure re-IP'd and connected to the switches,
Rick will ensure they are on the appropriate VLANs. We will start
bringing up systems (the internal DNS/DHCP server initially) and we will
test connectivity to them through the VLAN. Once we are satisfied we can
reach all of the required network segments we will start bring up
additional servers then test clients. Once they are all on and can be
reached over the VPN link (and can communicate to each other internally)
we will start enabling jenkins instances. At that point the remote hands
will be called upon to start running basic tests and verifying
connectivity from remote locations but that probably won't be until late
afternoon/early evening our time. My goal is to have everything powered
on and on the network before I end my day tomorrow. I'm not going to
make plans for what tasks I will work on Wednesday until after I see
where we are tomorrow (well, later today really).

Now to get some sleep. 10 hours over the past three days isn't quite
cutting it; might see if I can squeeze in 5 tonight/this morning.

~w