RS/6000 Scalable POWERparallel Systems

This document applies to AIX Version 4.x.

1. Set the nodes to disk to unallocate all NIM resources by executing the following:

                  spbootins -r disk <Frame#> <Node#> <NumberOfNodes>
   

   An example for frame 1 with 15 nodes looks like the following:

                  spbootins -r disk 1 1 15
   

2. Now create the SPOT in debug mode by executing the following:

        FOR PSSP 2.x --> nim -Fo check -a debug=yes spot_XXXX
   

   WARNING: Make sure you are not NIM installing when using the F flag as documented in the preceding example, since this will cause the boot image to be rebuilt even if it is allocated.

   (Run lsnim -t spot to get the spot_XXXX name.)

3. When the preceding command completes, issue the following command:

        lsnim -l spot_XXXX  (PSSP-2.x)
   

   For wide/thin nodes, look for the following:

                  enter_dbg     = "rs6k 0xXXXXXXXX"
   

   For high (SMP) nodes, look for the following:

                  enter_dbg     = "rs6ksmp 0xXXXXXXXX"
   

   XXXXXXXX will be an address embedded in the text, such as 0x0013afa0. Drop the 0x and leading zeroes, and write down the remainder here : _____________. For example, the "remainder" in this example would be 13afa0.

4. Now set the nodes to install by executing the following command:

                  spbootins -r install <Frame#> <Node#> <NumberOfNodes>
   

   This is a similar command to that in step 1, but this will issue the NIM commands that will allocate the debug SPOT created in step 2.

5. Perform manual node conditioning on the nodes that you wish to debug. This will result in opening a tty to the nodes and having the > prompt displayed. Turn logging on, which should display the following:

                  st 0013afa0 2
                  g
   

   Then execute the following command:

                  st <remainder from step 3> 2
                  g
   

   It may be helpful to have the 3DigitDisplay or the Node Front Panel displayed so you can watch the LEDs change as the netboot continues. It may also be helpful to have the IBM 9076 Scalable POWERparallel Systems SP2 Diagnosis and Messages Guide (GC23-3899-03) "SP-Specific LED Values" available to determine approximately where in the boot process you are.

6. Start Logging (from aixterm).
   1. Focus the mouse on the tty window.
   2. Depress the control (Ctrl) key while pressing the left-most mouse button.
   3. Move the cursor to the "Logging" option. This will create a file called AixtermLog.<xxxxx> (where <xxxxx> is a random string). The log will be kept in the $HOME of the userid that you are logged in as. This is the file that contains the debug information.
   4. Be sure to toggle this function off when complete.

7. When the need for the debug SPOT is gone, change the SPOT back to regular state by running the following command:

        nim -Fo check spot_XXXX  (PSSP-2.x)
   

   Again, note the F flag in the preceding command syntax.

   This is a good means of obtaining useful error messages that you wouldn't otherwise see with NIM. Typical problems identified by this process are name resolution issues and network problems (routing and so on).