Help with Solaris 8 Panic

[JTB]

I keep getting this panic when running my VI (Labview 7.1.1) in Solaris 8.  This is a repeatable crash.  Any ideas?  Are there specific patches that I should be implementing?  Thanks. 


Mar 20 09:11:10 utaho3 unix: [ID 836849 kern.notice]  
Mar 20 09:11:10 utaho3 ^Mpanic[cpu0]/thread=2a10007dd20:  
Mar 20 09:11:10 utaho3 unix: [ID 340138 kern.notice] BAD TRAP: type=31 rp=2a10007d8b0 addr=68 mmu_fsr=0 occurred in module "unix" due to a NULL poin
ter dereference
Mar 20 09:11:10 utaho3 unix: [ID 100000 kern.notice]  
Mar 20 09:11:10 utaho3 unix: [ID 839527 kern.notice] sched:  
Mar 20 09:11:10 utaho3 unix: [ID 520581 kern.notice] trap type = 0x31
Mar 20 09:11:10 utaho3 unix: [ID 381800 kern.notice] addr=0x68
Mar 20 09:11:10 utaho3 unix: [ID 101969 kern.notice] pid=0, pc=0x10033b40, sp=0x2a10007d151, tstate=0x9900001605, context=0x0
Mar 20 09:11:10 utaho3 unix: [ID 743441 kern.notice] g1-g7: 300023dba08, 0, 8, 300023dba08, 300023dba00, 0, 2a10007dd20
Mar 20 09:11:10 utaho3 unix: [ID 100000 kern.notice]  
M ar 20 09:11:10 utaho3 genunix: [ID 723222 kern.notice] 000002a10007d4c0 unix:die+80 (31, 68, 104150c8, 0, 2a10007d8b0, c25a2068)
Mar 20 09:11:10 utaho3 genunix: [ID 179002 kern.notice]   %l0-3: 0000000000000008 0000030001930590 000003000189ca08 0000000000009df1
Mar 20 09:11:10 utaho3   %l4-7: 000003000250b980 00000000c0a8e201 0000000000000000 000000000000ffff
Mar 20 09:11:10 utaho3 genunix: [ID 723222 kern.notice] 000002a10007d5a0 unix:trap+8b8 (0, 1, 5, 0, 2a10007d8b0, 0)
Mar 20 09:11:11 utaho3 genunix: [ID 179002 kern.notice]   %l0-3: 0000000000000001 0000000000000000 00000000104241a0 0000000000000000
Mar 20 09:11:11 utaho3   %l4-7: 0000000000000031 0000000000000000 0000000000010000 0000000000000000
Mar 20 09:11:11 utaho3 genunix: [ID 723222 kern.notice] 000002a10007d6e0 unix:sfmmu_tsb_miss+66c (10428fa0, 0, 3000004df88, 0, 3000004df88, 19)
Mar 20 09:11:11 utaho3 genunix: [ID 179002 kern.notice]   %l0-3: 0000000000000000 0000000000000004 000004000e781a18 00000310000bf100
Mar 20 09:11:11 utaho3   %l4-7: 0000000000000000 0000000000000000 0000000000000000 0000000000000003
Mar 20 09:11:11 utaho3 genunix: [ID 723222 kern.notice] 000002a10007d800 unix:prom_rtt+0 (0, 0, 20, 86c00000, 10, 300023dba00)
Mar 20 09:11:11 utaho3 genunix: [ID 179002 kern.notice]   %l0-3: 0000000000000006 0000000000001400 0000009900001605 000000001001a7d0
Mar 20 09:11:11 utaho3   %l4-7: 0000000000000046 0000030002176000 0000000000000006 000002a10007d8b0
Mar 20 09:11:11 utaho3 genunix: [ID 723222 kern.notice] 000002a10007d950 ceisol:cei_qintr+f8 (300023dba00, 30001d082b8, 1, 10000, 30001d082b8, 10230
)
Mar 20 09:11:11 utaho3 genunix: [ID 179002 kern.notice]   %l0-3: 000000000001c000 0000030001d1e800 00000000000003ff 0000030001cf8000
Mar 20 09:11:11 utaho3   %l4-7: 0000030001d21f00 000000000000001f 0000000080000000 0000000000010220
Mar 20 09:11:11 utaho3 genunix: [ID 723222 kern.notice] 000002a10007da60 pcisch:pci_intr_wrapper+80 (1048dd70, 1048dda8, 300001908d0, 30000193c38, 3
00001b4de8, 1)
Mar 20 09:11:12 utaho3 genunix: [ID 179002 kern.notice]   %l0-3: 0000000010335fe4 0000000000000001 00000300053800c8 0000030002d7d440
Mar 20 09:11:12 utaho3   %l4-7: 0000000000000001 0000000000020000 0000000000000640 0000000000000000
Mar 20 09:11:12 utaho3 unix: [ID 100000 kern.notice]  
Mar 20 09:11:12 utaho3 genunix: [ID 672855 kern.notice] syncing file systems...
Mar 20 09:11:12 utaho3 genunix: [ID 904073 kern.notice]  done

[LaurenLL]

Hi JTB,

You mentioned that this behavior is repeatable.  What are you trying to do, and what steps must be carried out in order to recreate this?  I am looking to report this to R&D for further investigation, but it would be useful to get some specific information about what is causing the crash first.

Regards,

Lauren

[JTB]

I've tried to isolate between device I/O and simple processing but I was unsuccessful in doing so.  This panic may have something to do with the ethernet threads since I'm using GPIB-Enet.  That was my initial assumption until I caused the panic without using the GPIB-Enet traffic.  I have seen the panic with dumping the results of various solaris commands (i.e prtconf, psinfo, pkginfo, kstat, netstat ...).  The process sometimes is successful, but for the most part the panic occurs.

[LaurenLL]

Hi JTB,

Without being able to isolate the specific process that triggers this error, it is difficult to find a solution for it.  I am passing this information on to R&D, but I am not sure that there will be any results.  In addition, I have taken a look at some Solaris forums that report similar errors.  You may be able to get some more information from them.

Keep me updated if you can find anything specific on it.

Regards,

Lauren

[JTB]

I've continued to try and isolate this issue to one of three things.  The application includes hardware interfaces to 2 DIO cards and 2 Analog cards, a Condor 1553 card (QPMC 1553) , and a NI GPIB Enet interface to 9 Sorenson power supplies.  Upon eliminating the I/O to the GPIB-Enet, the application completes without failure.  As soon as the interface to the GPIB Enet is included, the application fails with the panic as mentioned above.  We are running Solaris 8.  I will attempt to capture the Ethernet traffic with the use of ethereal and post this later.  In the mean time,  are there any known Solaris patches that I should implement with the configuration I'm using?  I'm especially leary of the GPIB Enet as Solaris has spent considerable amount of effort to redesign the I/O stack in Solaris 10.

[LaurenLL]

Hi JTB,

There is not currently a patch for Solaris, but you may want to take a look at the following articles that address the use of GPIB-ENET devices with Solaris 8:

GPIB-ENET/100 Fails with Solaris 8

Unable to Use the GPIB-ENET with LabVIEW under Solaris

Again, I recommend trying to monitor the specific commands you are executing and where the error is occurring in order to isolate the issue.

Regards,

Lauren

[JTB]

I think that I may have been mis-understood.  Is there a list of NI recommended Solaris patches (to the Solaris 8 OS and developed by Sun) that are required for the National Instruments LabVIEW 7.1.1 PDS available to the public. 

BTW  Thanks for your prompt responses.  This is a bewildering problem that I would love to have a solution for. 

[chris_amadasun]

Hello JTB,
Is your Solaris running on a Sparc or Intel processor?
As this article states, LabVIEW will work on Solaris systems with the Sparc processors.

Please let us know in your reply.