Operating System - OpenVMS
1748148 Members
3952 Online
108758 Solutions
New Discussion юеВ

Re: Different VMS V8.3 I64 crashes

 
SOLVED
Go to solution
Dario Karlen
Frequent Advisor

Different VMS V8.3 I64 crashes

Hi
We had some crashes on both of our cluster nodes in the last few weeks.
I found in the clue$history following information:
Date Version System/CPU Node Bugcheck Process
PC Module Offset
----------------- -------- ------------------- ------ ------------ --------
------- -------- ----------------------- --------
11-MAR-2007 03:09 V8.3 HP rx2620 (1.60GHz ALESA1 PROCGONE SYSINIT
80B38EC0 IMAGE_MANAGEMENT 000495C0
11-MAR-2007 05:02 V8.3 HP rx2620 (1.60GHz ALESA1 PROCGONE SYSINIT
80B38EC0 IMAGE_MANAGEMENT 000495C0
11-MAR-2007 04:26 V8.3 HP rx2620 (1.60GHz ALESA2 PROCGONE SYSINIT
80B38EC0 IMAGE_MANAGEMENT 000495C0
21-SEP-2007 02:43 V8.3 HP rx2620 (1.60GHz ALESA1 CLUEXIT NULL
80D24290 SYS$CLUSTER 0003EF90
6-MAR-2008 04:55 V8.3 HP rx2620 (1.60GHz ALESA1 INSF_NONPAGE NULL
811E2470 SYS$SHDRIVER 00173670
9-APR-2008 17:27 V8.3 HP rx2620 (1.60GHz ALESA2 INSF_NONPAGE TCPIP$FT
PC046C3 811FA370 SYS$SHDRIVER 00173670

I wanted to increase the NPAGEDYN, but I didn't get it. I put the lines
"ADD_NPAGEDYN=800000"
"ADD_NPAGEVIR=800000" in the modparams.dat, executed autogen and did a restart. But when I check it in SYSGEN it shows me the same values as before. How can I increse the NPAGEDYN and NPAGEDIR???
Thank you in advance
Dario
21 REPLIES 21
labadie_1
Honored Contributor
Solution

Re: Different VMS V8.3 I64 crashes

Do not use the ADD_ in modparams.dat, better use
MIN_NPAGEDYN
and let autogen calculate NPAGEVIR.

After the autogen, you can check the values that will be used at the next boot, doing
$ mc sysgen
use current
sh npagedyn
sh npagevir

And you reboot only if the values suit you.

Vladimir Fabecic
Honored Contributor

Re: Different VMS V8.3 I64 crashes

You can do it, of course:
$ mc sysgen
SYSGEN > USE CURRECT
SYSGEN > SET NPAGEDYN new_value
SYSGEN > SET NPAGEVIR new_value
SYSGEN > WRITE CURRENT
And reebot after that.
Better way is to specify values in MODPARAMS.DAT to be:
NPAGEDYN=new_value
NPAGEVYR=new_value
and then:
$ @sys$update:autogen getdata reboot
In vino veritas, in VMS cluster
Dario Karlen
Frequent Advisor

Re: Different VMS V8.3 I64 crashes

We have an old VMS 7.3 Alpha, on which MAX NPAGEDYN and MAX NPAGEVIR is set to -1.
How can we do the same on our I64 that there is no limit for the NPAGE parameters?
labadie_1
Honored Contributor

Re: Different VMS V8.3 I64 crashes

By the way, PROCGONE means something has gone wrong very early in the boot process.

R0 was containing which value ?

labadie_1
Honored Contributor

Re: Different VMS V8.3 I64 crashes

be careful, as Npagedyn is related to your physical memory
Can you post

$ sh mem/phy


If you want, you can fix a high value for npagedyn/Npagevir, and eventually fail to boot your system, with a POOLEXPF or similar message.

Extract from the doc, at
http://h71000.www7.hp.com/doc/82FINAL/aa-pv5nj-tk/aa-pv5nj-tk.PDF


For the benefit of OpenVMS VAX systems with limited physical memory, AUTOGEN logs a warning message
in its report if NPAGEDYN exceeds 10 percent of physical memory or if NPAGEVIR exceeds 33 percent of
physical memory.
AUTOGEN also limits its own calculated value for NPAGEDYN to 20 percent of physical memory, and limits
NPAGEVIR to 50 percent of physical memory. These calculated values are adequate for most workstations
and systems with 16 or fewer megabytes of physical memory. If your system requires a larger value, you can
override the AUTOGEN calculated values by setting higher values in MODPARAMS.DAT
Dario Karlen
Frequent Advisor

Re: Different VMS V8.3 I64 crashes

ina > sh mem/phy
System Memory Resources on 10-APR-2008 18:38:32.33

Physical Memory Usage (pages): Total Free In Use Modified
Main Memory (1.99GB) 262096 174683 86990 423

Of the physical pages in use, 34834 pages are permanently allocated to OpenVMS.
ina >

We want the full memory available for the system....
Dario Karlen
Frequent Advisor

Re: Different VMS V8.3 I64 crashes

the system whe have trouble with insufficient nonpaged pool is with one of our customer. I tried to set the NPAGEDYN and NPAGEDIR to -1 in our I64 in the office, no I cannot start the machine, showing following errors:
EFI Boot Manager ver 1.10 [14.62] Firmware ver 4.13 [4634]

Please select a boot option

OpenVMS on DKA0: PKA0.0

EFI Shell [Built-in]

DVD rom device

Boot Option Maintenance Menu

System Configuration



Use ^ and v to change option(s). Use Enter to select an option

Loading.: OpenVMS on DKA0: PKA0.0

Starting: OpenVMS on DKA0: PKA0.0


**************************************************************

* Exception taken before exception handler has been loaded! *

* Unable to take crashdump. *

**************************************************************


* Exception Frame Display: *


Exception taken at IP 00000000.00056230, slot 00 from Kernel mode

Exception Frame at 00000000.2020BB80

Trap Type 00000008 (Access control violation fault)

IVT Offset 00001000 (Alternate Data TLB Fault)


* = Value read directly from the register rather than the frame


Control Registers:

CR0 Default Control Register (DCR) 00000000.00007F00

CR1 Interval Timer Match Register (ITM) * 00000023.D60C3CC9

CR2 Interruption Vector Address (IVA) * 00000000.000C0000

CR8 Page Table Address (PTA) * 00000000.0000003C

CR16 Processor Status Register (IPSR) 00001010.09026010

CR17 Interrupt Status Register (ISR) 00000802.00000000

CR19 Instruction Pointer (IIP) 00000000.00056230

CR20 Faulting Address (IFA) 00000801.FF0003C0

CR21 TLB Insertion Register (ITIR) 00000000.00000034

CR22 Instruction Previous Address (IIPA) 00000000.00056220

CR23 Function State (IFS) 80000000.0000030A

CR24 Instruction immediate (IIM) 00000000.00000000

CR25 VHPT Hash Address (IHA) 00000000.00808EA0

CR64 Local Interrupt ID (LID) * 00000000.00000000

CR66 Task Priority Register (TPR) * 00000000.00010000

CR68 External Interrupt Req Reg 0 (IRR0) * 00000000.00000000

CR69 External Interrupt Req Reg 1 (IRR1) * 00000000.00000000

CR70 External Interrupt Req Reg 2 (IRR2) * 00000000.00000000

CR71 External Interrupt Req Reg 3 (IRR3) * 00000000.00000000

CR72 Interval Time Vector (ITV) * 00000000.00010000

CR73 Performance Monitoring Vector (PMV) * 00000000.00010000

CR74 Corrected Machinecheck Vector (CMCV) * 00000000.00010000

CR80 Local Redirection Register 0 (LRR0) * 00000000.00010000

CR81 Local Redirection Register 1 (LRR1) * 00000000.00010000


Application Registers:

AR0 Kernel Register (KR0) * 00000000.20620000

AR1 Kernel Register (KR1) * 00000000.60000000

AR2 Kernel Register (KR2) * 00000000.00000000

AR3 Kernel Register (KR3) * 00000024.C5A84E1C

AR4 Kernel Register (KR4) * 00000000.00000000

AR5 Kernel Register (KR5) * 00000000.00000000

AR6 Kernel Register (KR6) * 00000000.00000000

AR7 Kernel Register (KR7) * 00000000.10001000

AR16 Register Stack Config Reg (RSC) 00000000.00000000

AR17 Backing Store Pointer (BSP) 00000000.201FC578

AR18 Backing Store for Mem Store (BSPSTORE) 00000000.201FC328

AR19 RSE NaT Collection Register (RNAT) 00000000.00000000

AR32 Compare/Exchange Comp Value Reg (CCV) 00000000.201701D0

AR36 User NaT Collection Register (UNAT) 00000000.00000000

AR40 Floating-point Status Reg (FPSR) 0009804C.8A70433F

AR44 Interval Time Counter (ITC) * 00000025.C1B93AF3

AR64 Previous Function State (PFS) 00000000.00001733

AR65 Loop Count Register (LC) 00000000.00000000

AR66 Epilog Count Register (EC) 00000000.00000000


Processor Status Register (IPSR):

AC = 0 MFL= 1 MFH= 0 IC = 1 I = 1 DT = 1

DFL= 0 DFH= 0 RT = 1 CPL= 0 IT = 1 MC = 0 RI = 0

Interrupt Status Register (ISR):

Code 00000000 X = 0 W = 1 R = 0 NA = 0 SP = 0

RS = 0 IR = 0 NI = 0 SO = 0 EI = 0 ED = 1


Branch Registers: Region Registers:

B0 00000000.000D02F0 RR0 * 00000000.00000034

B1 80000000.FFD293D0 RR1 * 00000000.00000030

B2 00000000.00000000 RR2 * 00000000.00000030

B3 00000000.00000000 RR3 * 00000000.00000030

B4 00000000.00000000 RR4 * 00000000.00000030

B5 00000000.00000000 RR5 * 00000000.00000030

B6 00000000.3FF9F1F0 RR6 * 00000000.00000030

B7 00000000.3FFB2A80 RR7 * 00000000.00000334


Floating Point Registers: FPSR 0009804C.8A70433F

F6 00000000.0001003E.00000000.0000E000

F7 00000000.0001003E.00000000.00000007

F8 00000000.0001003E.00000000.00002000

F9 00000000.0001003E.00000000.00000000

F10 00000000.0001003E.00000000.00002000

F11 00000000.0001003E.00000000.00000000


Miscellaneous Registers:

Processor Identifier (CPUID 0,1) GenuineIntel

(CPUID 3) 00000000.1F020204

Interrupt Priority Level (IPL) 0000001F

Stack Align 000002D0

NaT Mask 0030

PPrev Mode 00

Previous Stack 00

Interrupt Depth 00

Preds 80000000.FF5218A3

Nats 00000000.00000000

Context 00000000.FF519AA3


General Registers:

R0 00000000.00000000 GP 00000000.00480000 R2 000007FF.FF800406

R3 000007FF.FFDF0000 R4 00000000.FFCF64C0 R5 00000000.00000004

R6 00000000.00000000 R7 80000000.FED00000 R8 00000000.00000001

R9 00000000.00000000 R10 00000000.00000000 R11 00000000.00000028

SP 00000000.2020B758 TP 00000000.0000040A R14 000000FF.FFFFFFFF

R15 00000000.000048F8 R16 00000000.00000048 R17 00000000.00000000

R18 00000001.FFFFF7C0 R19 FFFFF800.007FF800 R20 FFFFF800.00000000

R21 00000000.00000018 R22 00000000.000048F8 R23 00000000.00001401

R24 00000000.2020BE70 R25 00000000.00000003 R26 00000000.000000FC

R27 00000000.000048FC R28 00000000.00000100 R29 00000000.000000E0

R30 00000000.01000000 R31 00000000.00000004


Interrupted Frame RSE Backing Store , Size = 10 registers


00000000.201FC578: 00000801.FF0003C0 (R32)

00000000.201FC580: 00000000.000048F8 (R33)

00000000.201FC588: 00000000.00001401 (R34)

00000000.201FC590: 00000000.000D02F0 (R35)

00000000.201FC598: 00000000.00001733 (R36)

00000000.201FC5A0: 00000000.00480000 (R37)

00000000.201FC5A8: 000007FF.FF800403 (R38)

00000000.201FC5B0: 00000000.00000000 (R39)

00000000.201FC5B8: 00000000.00000028 (R40)

00000000.201FC5C0: 00000000.000048F8 (R41)


******************************************

* Using INIBRK to get help from Xdelta! *

******************************************

Brk 0 at 000D22E1

000D22E1! break.i 080003

000D22E2! nop.i 000000

does anybody have a clue? is something broken or could it caused by my manipulation?
Volker Halle
Honored Contributor

Re: Different VMS V8.3 I64 crashes

Dario,

you should not set NPAGEDYN or NPAGVIR to -1. This has most likely caused the exception early during boot.

Please consider to have AUTOGEN make the necessary changes:

$ @SYS$UPDATE:AUTOGEN SAVPARAMS GENPARAMS FEEDBACK

Then check SYS$SYSTEM:AGEN$PARAMS.REPORT to check, which changes AUTOGEN would have making. Once satisfied, replace the GENPARAMS parameter with SETPARAMS and reboot your system afterwards.

You check in the running system, whether nonpaged pool is getting expanded with:

$ ANAL/SYS
SDA> CLUE MEM/STAT

or

$ SHOW MEM/POOL/FULL

Check whether Nonpaged Dynamic Memory Current Size exceeds Initial Size. It becomes critical, when Current Size is getting near maximum size.

Volker.
Dario Karlen
Frequent Advisor

Re: Different VMS V8.3 I64 crashes

Thanks Volker for your answer. The server will be repaired in the next few days. What I do not understand: in VMS with Alphas, this limit was set to -1, why is it not the same with I64's? The server crashes running out of "nonpaged pool". So that means I have to increase this number. Do you think autogen will generate sufficient limits?

thanks a lot for your professional help! I really apreciate it.
Dario