DAQ Networking

See also - DAQ Doc Portal

DAQ Network Components

Portserver

A portserver is a telnet (insecure network interface) capable, serial cable connected, network interface. The advantage to using serial connections is it allows a network connection during boot and enables easy remote power cycling and NVRAM restoration to ROCs.

See here for some useful commands.

NFS Network File System

Main Article - NFS Installation Guide

To mount an NFS filesystem to a ROC (linux or VXworks) there are several steps involved (see How To Forge - NFS Guide)

• Have NFS serve the CODA 2.6.2 directory to localhost (127.0.0.1).
• Mount it, in a different location (pretending to me the VME controller):

• • example /etc/exports:

/coda/2.6.2/    localhost(rw,sync,no_root_squash,no_subtree_check)
mount it /somewhere_else

• • (as root)

mkdir /somewhere_else 
mount localhost:/coda/2.6.2 /somewhere_else

FAQ

For linux systems:

• Make sure that the ethernet ports being used are configured to DHCP over the ethernet network
  • Edit the /etc/sysconfig/network-scripts/ifcfg-eth# configuration files are correctly set to allow the desired ethernet port to DHCP on-boot (on RHEL and Centos only)
  • Acquire a static IP from the ethernet network you are using (if its a LAN you manage set these in the router or switch yourself)
  • Identify the hostnames and IPs of the local machine and remote server you want to connect to in the /etc/hosts file
  • Make sure the /etc/sysconfig/network file has the localhost's hostname set in it (can also be done with the 'host' unix command)
  • On the NFS host server ensure that the /etc/exports file correctly targets the IPs or entire network (using .0 as the last digit of the IP address) that you want to server the NFS contents to
    • After updating a network connection or target computer in /etc/exports run the "service autofs restart" or "exports -a" command as root
    • Make sure the /etc/hosts file includes the same information for all computers on the local network involved in the NFS file sharing, with 127.0.0.1 assigned to 'localhost' on all
  • Make sure that the permissions for the shared files allow for appropriate read, write, and execute privileges for all users and groups, on local and remote computers
    • To check the user and group permissions are set correctly, the naming conventions can be deceptive - use ls -n to show the number of users and groups, which is what really matters
    • Even if two separate computers have the same user and group names but their user and group numbers are different you are able to tell this and assign the rwx privileges correctly on the shared folders (using 'chown -R' and 'chmod -R 777' or whatever correct permissions you want to set)

For VXworks systems:

• The same things apply as for ethernet network mounted file systems, except that VXworks uses the ~/.rhosts file to determine which IPs to allow to mount an NFS shared folder
  • Therefore, follow the /etc/exports file convention, except with ~/.rhosts instead for VXworks ROC file sharing
  • The ROC_name and ROC_name.jlab.org addresses both need to be listed for it to work

DAQ Network Map

For History, see DAQ Network Map

As of July 27th 2018 DAQ Network Map
Location	SSH Hostname	Software ROC#	Hardware ROC# and Branch	Serial Access Method	Access Command (udp Write Failure error messages are normal)
RHRS	'happex7'	26	02 branch 4	serial port via portserver hatsv3 port 2	'telnet hatsv3 2002'
LHRS	'hallavme14'	25	05 branch 4	serial port via portserver hatsv21 port 4	'telnet hatsv21 2004'
TS	'happex4'	no ROC	no ROC	serial port via network console server acons1 port 8	'ssh adaq:ttyS8\@acons1'
Counting House	'halladaq6'	23	00 branch 1	serial port via network console server acons1 port 4	'ssh adaq:ttyS4\@acons1'
Injector	'qweak1'	24	00 branch 2	serial port via network rpi12 server 'minicom'	'ssh pi@rpi12' the 'run_minicom'

Accessible Computers

DAQ Computers

Workspace Computers
Location	SSH Hostname	Username
Hall A LHRS	halla-lpt1	ahut
Hall A RHRS	halla-lpt2	ahut
	rpi4	pi
Hall A Counting House	adaq1	apar
	adaq2	apar
	adaq3	apar
	adaq4	apar

MCC/ACE/Injector

There is also a computer on wheels in the Injector area. To log in to this computer you need to have a Machine Control Center (MCC) account, which is independent of the main JLab CUE system (but not agnostic of it). To get a MCC/ACE account follow these steps.

• Go to the main JLab CCPR system here
• Click on the link at the top for ACE CCPRs here
• Then select New User Account in the category menu and proceed with your request
• They will process your request and send you an email asking you to log in with your JLab CUE account to change your ACE password
• You need to log in to a computer that has a static DHCP managed JLab IP address and follow that link to change your password (which cannot be the same as your current CUE password)

Phone Numbers

Commonly Used Phone Numbers
Who	Why	Number
Machine Control Center (MCC)	To control the accelerator	(757) 269 7047 or 7048
Security Guards	Emergency	(757) 269 5822
	Normal	(757) 269 7169
Hall A Counting House	Back Room	(757) 269 5505
		(757) 269 5509
	Control Room	(757) 269 5501
Radcon	Radiation help	(757) 876 1743
Computer Helpdesk	Computer or network help	(757) 269 7155
Health Clinic	Health questions	(757) 269 7539

For table formatting help see MediaWiki Help:Tables