Initial daq cluster setup checklist » History » Version 5

Version 4 (Pengfei Ding, 01/13/2020 10:27 AM) → Version 5/40 (Pengfei Ding, 01/13/2020 10:28 AM)

h1. Initial DAQ cluster setup checklist.

_Objective: To reduce the number service desk tickets during the initial setup of DAQ development / production clusters._

h2. Networking

# define subnets for IPMI, fnal/public and data interfaces
# define host names for all network interfaces and make them consistent
** mydaq-br01, mydaq-eb01, mydaq-ipmi-br01, mydaq-data-br01
** the list of host names should be complete as if all hardware is available
** put all host names into /etc/hosts and distribute it across all servers
# make a consistent IP address assignment across all subnets
** use address blocks for the same server roles
** make the last octet of an IP address being the same across all NICs of the same host
# configure authentication
** Kerberos for the public interface
** publickey for the data interface
# create instructions for rebooting servers using IPMI
# enable the 9000 MTU frames on all interfaces and networking equipment by default
# configure and verify that multicasting is enabled and working all networking equipment

h2. Users

# define a shared user for
** managing UPS products
** running daq, dcs, databases
# add all people from the RSI group to the /root/.k5login
# add all known daq users to the daq and dcs shared accounts
# shared user profiles are not expected to have any customizations

h2. Storage areas

# setup a reliable NFS server for /home, /daq/products, /daq/database, /daq/log, /daq/database, /daq/tmp,.... /data /scratch, /daq/backup
# reserve adequate disk space for each area
# create a designated scratch area for doing builds on a local NVMe derive, preferably on the fastest server
** a faster NVMe drive such as Samsung 970 Pro or faster is preferred
# setup a nightly backup for /home and a weekly backup for /daq/backup areas
# the performance of the NFS should be monitored

h2. Software

# any base software such as the OS and productivity RPMs should be identical on all servers
# a default list of installed software packages should not be impeding the development/ testing work, e.g. emacs, vim, mc, tmux, perf, iperf, strace, dstat,..... VNC/MATE should be installed by default
# implement system monitoring using ganglia
or similar software


System Services

# DNS, Kerberos, NIS?
# Supervisord?
# Ganglia, graphite, influxdb, prometheus, mongoDB?