Reports until 16:13, Wednesday 18 October 2017
H1 DAQ (CDS)
david.barker@LIGO.ORG - posted 16:13, Wednesday 18 October 2017 (39078)
h1fw0 down for RAID upgrade, h1nds0 serving h1fw1 data.

WP7161. Dan and Dave:

I ran four pairs of 10Gig multi-mode fibers between the E-18 raid box 0 (fw0) and the following locations:

Q-logic switch in DAQ-1 rack (one pair)

Q-logic switch in DAQ-0 rack (one pair)

h1ldasgw0 (two pairs)

h1fw0 frame writer code was turned off.

h1ldasgw0 (solaris machine, X4270) was powered off. We removed the dual port 10GIG-E card which had been installed as part of the frame writer instability investigation. We installed a dual port fibre-channel card in its place, and Dan re-organized the FC card layout.

We connected both ports of the dual-FC card to the primary FC port on the two controllers on the E-18-0 raid box. We connected the secondary FC ports on these controller cards to one port each on the Q-logic switches, using the ports previously connected directly to h1ldasgw0.

In lieu of a management connection to the E-18-0 controller cards (scheduled for later this week), Dan connected a laptop directly to these units.

An updated DAQ as-built drawing will be available soon.

Since h1fw0 will be 'starting from scratch' writing its frame files, I have transitioned h1nds0 over to serving h1fw1 data in the interim.