Changes between Version 2 and Version 3 of Internal/Repair


Ignore:
Timestamp:
May 30, 2007, 4:17:05 PM (17 years ago)
Author:
Joseph F. Miklojcik III
Comment:

changes resulting from 5/29 meeting and broader understanding of calbiration

Legend:

Unmodified
Added
Removed
Modified
  • Internal/Repair

    v2 v3  
    1818particular round of repairs.
    1919
    20 4) Comment out lines for these nodes from dhcp1:/etc/dhcp3/dhcpd.conf.
    21 Restart dhcpd on dhcp1.
    22 
    23 5) For each node to be repaired, remove each node from its mounting in
     204) For each node to be repaired, remove each node from its mounting in
    2421the grid, leaving the node id box attached.  As you remove nodes, take
    2522them and their node id box back to the node repair area.  One or two
     
    2825hardware or incorrectly installed connections on the wiki page.
    2926
    30 6) Once in the node repair area, remove the node id box and then the
     275) Once in the node repair area, remove the node id box and then the
    3128yellow node enclosure.  Verify that the node id boxes match the list
    3229of nodes to be repaired on the wiki page, and that the 802.11 hardware
    3330vendor matches what is expected.  Note exceptions on the wiki page.
    3431
    35 7) Replace the power supply.  Take care to put old power supplies in
     326) Replace the power supply.  When attaching the new power supply to
     33the chassis, put in all four screws loosely first, then tighten once
     34all four are in place. Take care to put old power supplies in
    3635the garbage bin.  If the 802.11 hardware vendor did not match what is
    3736expected, correct the hardware.  Replace the enclosure.  Replace the
    38 node id box.
     37node id box.  Do not replace the side enclosure screw, we no longer
     38use them.
    3939
    40 8) Calibrate the node (NYI).
     407) Bring the node to the calibration shelf.  Attach the CM cable,
     41CONTROL cable, power cable, and two antennas.  Run the calibration
     42experiment.
    4143
    42 8) Replace the node in the grid.  Verify the node id box against two
    43 adjacent nodes.
     448) Replace the node in the grid.  Re-attach all peripherals, such as USB
     45devices.  Verify the node id box visually against two adjacent nodes.
    4446
    45479) Once all nodes have been repaired and replaced, verify that the
    4648nodes are not red on the orbit-lab.org/wiki/Staus page.  That is, that
    47 the CM reports back to the CMC correctly.
     49the CM reports back to the CMC correctly.  Note exceptions in the wiki
     50page.
    4851
    49 10) Turn the repaired nodes on.  Because they obtain pool addresses
    50 from dhcp, they will load an 'inventory' image (NYI).  Wait five
     5210) Comment out lines for these nodes from dhcp1:/etc/dhcp3/dhcpd.conf.
     53Restart dhcpd on dhcp1.
     54
     5511) Turn the repaired nodes on.  Because they obtain pool addresses
     56from dhcp, they will load an 'inventory' image.  Wait five
    5157minutes for the inventory image to finish loading, and for the inventory
    5258script to run.
    5359
    54 11) Run the gendhcpconf script on repository2.  Compare its output
     6012) Run the gendhcpconf script on repository2.  Compare its output
    5561with the entries you commented out in step 4.  Correct
    5662dhcp1:/etc/dhcp3/dhcpd.conf if needed.
    5763
    58 12) During the following maintenance slot, verify that you can image
    59 all nodes that have been repaired since the last maintenance slot by
    60 running the CM stress experiment (NYI).
     6413) If there is time, verify that imaging works on the repaired nodes.