The way that I used to run bruco didn't work for me today.  With help from Elenna and Camilla I was able to run it, so here is what I did:

ssh into ssh.ligo.org, select 4 for LHO cluster and c for ldas.pcdev1

made a fresh clone of https://git.ligo.org/gabriele-vajente/bruco in a directory called new_bruco

cd /home/sheila.dwyer/new_bruco/bruco

python -m bruco --ifo=H1 --channel=GDS-CALIB_STRAIN_CLEAN --gpsb=1417500178 --length=400 --outfs=4096 --fres=0.1 --dir=/home/sheila.dwyer/public_html/brucos/GDS_1417500178 --top=100 --webtop=20 --plot=html --nproc=20 --xlim=7:2000 --excluded=share/lho_excluded_channels_O4.txt

The bruco now appears at: https://ldas-jobs.ligo-wa.caltech.edu/~sheila.dwyer/brucos/GDS_1417500178/

This is for a quiet time overnight last night.

Input jitter does have a contribution, as Robert suspected based on looking at the DARM spectrum.  Jenne looked at cleaning, and plans to try out a new cleaning during Monday's commissoning window.