Difference between revisions of "DAQ test computer"

From GlueXWiki
Jump to: navigation, search
m
 
(9 intermediate revisions by the same user not shown)
Line 32: Line 32:
  
 
* and finally:
 
* and finally:
       $ mount -a
+
       $ /etc/init.d/autofs restart
  
  
Line 41: Line 41:
 
**      rsh-server
 
**      rsh-server
 
**      some-java-version  (use yum search java) (e.g. java-1.6.0-ibm-devel.i386)
 
**      some-java-version  (use yum search java) (e.g. java-1.6.0-ibm-devel.i386)
 +
**      scons.noarch
 +
  
 
Also interesting for debugging network traffic are tcpdump, wireshark, and wireshark-gnome.
 
Also interesting for debugging network traffic are tcpdump, wireshark, and wireshark-gnome.
Line 75: Line 77:
 
       192.168.1.2    pmt1     
 
       192.168.1.2    pmt1     
  
 +
* note that cfengine may disable rsh, need to talk to Sherman White to get exemption for host computer
  
* configure eth1 for static routing, IP 192.168.1.1, netMask 255.255.255.0 using gui (if remote can get gui via: $ system-config-network)
+
 
 +
* configure eth0 for DHCP on the JLab network
 +
* configure eth1 for static routing, IP 192.168.1.1, netMask 255.255.255.0 using gui (if remote can get gui via: $ system-config-network).
  
  
Line 82: Line 87:
 
*  /etc/sysconfig/network-scripts/ifcfg-eth0
 
*  /etc/sysconfig/network-scripts/ifcfg-eth0
 
       PEERDNS=yes
 
       PEERDNS=yes
 +
      ONBOOT=yes
  
 
*  /etc/sysconfig/network-scripts/ifcfg-eth1
 
*  /etc/sysconfig/network-scripts/ifcfg-eth1
 
       PEERDNS=no
 
       PEERDNS=no
 +
      ONBOOT=YES
  
 
*  /etc/sysconfig/network
 
*  /etc/sysconfig/network
Line 91: Line 98:
  
  
* activate eth1 (can also use gui):
+
* activate eth0 and eth1 (can also use gui):
 +
      /sbin/ifup eth0
 
       /sbin/ifup eth1
 
       /sbin/ifup eth1
  
Line 134: Line 142:
  
 
* reboot if something is not working properly.  Contact computer center if reboot doesn't help.
 
* reboot if something is not working properly.  Contact computer center if reboot doesn't help.
 +
 +
 +
* setting vxWorks boot parameters:
 +
 +
[VxWorks Boot]: p
 +
<br>
 +
boot device          : scsi=2,0
 +
processor number    : 0
 +
host name            : mizar
 +
file name            : /sd0/boot/vx2700_sd0
 +
inet on ethernet (e) : 129.57.14.20:fffffc00
 +
host inet (h)        : 129.57.14.33
 +
gateway inet (g)    : 129.57.12.1
 +
user (u)            : abbottd
 +
flags (f)            : 0x20
 +
target name (tn)    : davw10
 +
startup script (s)  : /sd0/drew/pem/pemv7.boot
 +
other (o)            : dc
 +
<br>
 +
  host name  -  Name of the Unix host
 +
  e  -  IP address of the board:subnet mask
 +
            (should be in /etc/hosts on unix as well)
 +
  h  -  Unix host IP address
 +
  g  -  IP gateway (this is optional - it can be cleared)
 +
  u  -  username on unix host
 +
  tn  -  name of VME board (should be in /etc/hosts on unix)
 +
  o  -  device name for the network interface (dc)
 +
 +
 +
* nfs mounts on vxWorks
 +
**nfsMount("host","file_system")
 +
 +
 +
* minicom settings
 +
** /dev/ttyS0
 +
** 9600 8N1
  
  
 
* Future possibilities
 
* Future possibilities
 
** snapshot using rsync
 
** snapshot using rsync

Latest revision as of 09:01, 26 September 2011

Instructions for configuring a Hall D DAQ computer for stand-alone operation, after installing RHEL5 from JLab CD using standard halld passwords. Instructions below assume host name moe.jlab.org and VxWorks processor pmt1.jlab.org.

Instructions also assume that the main (JLab) network connection is on eth0, and the daq system is on a private network eth1.


Elliott Wolin, 20-Aug-2008


From the root account:

  • create halld group gid=267 using gui (if remote can get gui via: $ system-config-users)


  • create halld account id=500 with standard password and halld as primary group using gui


  • have computer center mount /group/halld and /group/da, or at least allow the computer to mount them. If the latter:
     $ mkdir /group
     $ chmod a+rwx /group
     $ ln -s /misc/da /group/da
     $ ln -s /misc/halld /group/halld


  • Edit /etc/auto.misc and add:
     da              -rw             jlabgrp:/vol/vol1/da
     halld           -rw             jlabgrp:/vol/vol1/halld
  • and finally:
     $ /etc/init.d/autofs restart


  • yum install:
    • minicom
    • xinetd
    • ant
    • rsh-server
    • some-java-version (use yum search java) (e.g. java-1.6.0-ibm-devel.i386)
    • scons.noarch


Also interesting for debugging network traffic are tcpdump, wireshark, and wireshark-gnome.


  • enable rsh in /etc/xinetd.d/rsh on the local private network only:
     socket_type             = stream
     wait                    = no
     user                    = root
     log_on_success          += USERID
     log_on_failure          += USERID
     server                  = /usr/sbin/in.rshd
     disable                 = no
     only_from               = 192.168.0.0/16
     bind                    = 192.168.1.1


  • configure autostart of xinetd/rsh and ncsd:
     /sbin/chkconfig xinetd on
     /sbin/chkconfig nscd   on


  • start xinetd and nscd:
     /sbin/service xinetd start
     /sbin/service nscd   start

or

     /etc/init.d/xinetd start
     /etc/init.d/nscd   start


  • edit /etc/hosts:
     127.0.0.1       moe localhost localhost.localdomain moe.jlab.org
     ::1             localhost.localdomain   localhost
     192.168.1.2     pmt1    
  • note that cfengine may disable rsh, need to talk to Sherman White to get exemption for host computer


  • configure eth0 for DHCP on the JLab network
  • configure eth1 for static routing, IP 192.168.1.1, netMask 255.255.255.0 using gui (if remote can get gui via: $ system-config-network).


Edit network files:

  • /etc/sysconfig/network-scripts/ifcfg-eth0
     PEERDNS=yes
     ONBOOT=yes
  • /etc/sysconfig/network-scripts/ifcfg-eth1
     PEERDNS=no
     ONBOOT=YES
  • /etc/sysconfig/network
    • turn IPv6 off
    • set hostname to moe


  • activate eth0 and eth1 (can also use gui):
     /sbin/ifup eth0
     /sbin/ifup eth1


  • edit /etc/sudoers and add:
     halld   ALL=(ALL)       ALL


  • fix ttyS0 permisions for minicom. May be needed after every reboot:
     chmod a+rw /dev/ttyS0
  • or from halld account:
     sudo chmod a+rw /dev/ttyS0


From the halld account:


  • copy files from /group/da/ejw/daq-config into halld home dir:
    • .complete.tcsh
    • .emacs
    • .cshrc
    • .alias
    • .login


  • create directory /home/halld/.xemacs and copy in from /group/da/ejw/daq-config:
    • init.el
    • custom.el


  • create /home/halld/.rhosts:
     192.168.1.2
     pmt1.jlab.org


  • set .rhost permissions:
     chmod u=rw .rhosts
     chmod go-rw .rhosts


  • reboot if something is not working properly. Contact computer center if reboot doesn't help.


  • setting vxWorks boot parameters:
[VxWorks Boot]: p

boot device  : scsi=2,0 processor number  : 0 host name  : mizar file name  : /sd0/boot/vx2700_sd0 inet on ethernet (e) : 129.57.14.20:fffffc00 host inet (h)  : 129.57.14.33 gateway inet (g)  : 129.57.12.1 user (u)  : abbottd flags (f)  : 0x20 target name (tn)  : davw10 startup script (s)  : /sd0/drew/pem/pemv7.boot other (o)  : dc
host name - Name of the Unix host e - IP address of the board:subnet mask (should be in /etc/hosts on unix as well) h - Unix host IP address g - IP gateway (this is optional - it can be cleared) u - username on unix host tn - name of VME board (should be in /etc/hosts on unix) o - device name for the network interface (dc)


  • nfs mounts on vxWorks
    • nfsMount("host","file_system")


  • minicom settings
    • /dev/ttyS0
    • 9600 8N1


  • Future possibilities
    • snapshot using rsync