Document revision date: 30 March 2001
[Compaq] [Go to the documentation home page] [How to order documentation] [Help on this site] [How to contact us]
[OpenVMS documentation]

OpenVMS VAX System Dump Analyzer Utility Manual


Previous Contents Index

SET RMS

Changes the options shown by the SHOW PROCESS/RMS command.

Format

SET RMS =(option[,...])


Parameter

option

Data structure or other information to be displayed by the SHOW PROCESS/RMS command. Table SDA-14 lists those keywords that you can use as options.

Table SDA-14 SET RMS Command Keywords for Displaying Process RMS Information
Keyword Meaning
[NO]ALL[: ifi] 1 All control blocks (default)
[NO]ASB Asynchronous context block
[NO]BDB Buffer descriptor block
[NO]BDBSUM BDB summary page
[NO]BLB Buffer lock block
[NO]BLBSUM Buffer lock summary page
[NO]CCB Channel control block
[NO]DRC Directory cache
[NO]FAB File access block
[NO]FCB File control block
[NO]FWA File work area
[NO]GBD Global buffer descriptor
[NO]GBDSUM GBD summary page
[NO]GBH Global buffer header
[NO]GBSB Global buffer synchronization block
[NO]IDX Index descriptor
[NO]IFAB[: ifi] 1 Internal FAB
[NO]IFB[: ifi] 1 Internal FAB
[NO]IRAB Internal RAB
[NO]IRB Internal RAB
[NO]JFB Journaling file block
[NO]NAM Name block
[NO]NWA Network work area
[NO]PIO Image I/O (NOPIO), the default, or process I/O (PIO) 2
[NO]RAB Record access block
[NO]RLB Record lock block
[NO]RU Recovery unit structures, including the recovery unit block (RUB), recovery unit stream block (RUSB), and recovery unit file block (RUFB)
[NO]SFSB Shared file synchronization block
[NO]WCB Window control block
[NO]XAB Extended attribute block
[NO]* Current list of options displayed by the SHOW RMS command


1The optional parameter ifi is an internal file identification. The default ifi (ALL) is all the files the current process has opened.
2Specifying the PIO option causes the SHOW PROCESS/RMS command to display the indicated structures for process-permanent file I/O.

The default option is ALL:ALL,NOPIO, designating for display by the SHOW PROCESS/RMS command all structures for all files related to the image I/O of the process.

To list more than one option, enclose the list in parentheses and separate options by commas. You can add a given data structure to those displayed by ensuring that the list of keywords begins with the * (asterisk) symbol. You can delete a given data structure from the current display by preceding its keyword with NO.


Qualifiers

None.

Description

The SET RMS command determines the data structures to be displayed by the SHOW PROCESS/RMS command. (See the examples included in the discussion of the SHOW PROCESS command for an indication of the information provided by various displays.) You can examine the options that are currently selected by issuing a SHOW RMS command.

Examples

#1

SDA>  SHOW RMS
RMS Display Options:  IFB,IRB,IDX,BDB,BDBSUM,ASB,CCB,WCB,FCB,FAB,RAB,NAM,
XAB,RLB,BLB,BLBSUM,GBD,GBH,FWA,GBDSUM,JFB,NWA,RU,DRC,SFSB,GBSB
 
Display RMS structures for all IFI values.
   
SDA>  SET RMS=IFB
SDA>  SHOW RMS
   
RMS Display Options:  IFB
 
Display RMS structures for all IFI values.
 
      

The first SHOW RMS command shows the default selection of data structures that are displayed in response to a SHOW PROCESS/RMS command. The SET RMS command selects only the IFB to be displayed by subsequent SET/PROCESS commands.

#2

SDA>  SET RMS=(*,BLB,BLBSUM,RLB)
SDA>  SHOW RMS
   
RMS Display Options:  IFB,RLB,BLB,BLBSUM
 
Display RMS structures for all IFI values.
      

The SET RMS command adds BLB, BLBSUM, and RLB to the list of data structures that the SHOW PROCESS/RMS command currently displays.

#3

SDA>  SET RMS=(*,NORLB,IFB:05)
SDA>  SHOW RMS
   
RMS Display Options:  IFB,BLB,BLBSUM
Display RMS structures only for IFI=5.
      

The SET RMS command removes the RLB from those data structures displayed by the SHOW PROCESS/RMS command and causes only information about the file with the ifi of 5 to be displayed.

#4

SDA> SET RMS=(*,PIO)
      

The SET RMS command indicates that the data structures designated for display by SHOW PROCESS/RMS be associated with process-permanent I/O instead of image I/O.

SHOW CALL_FRAME

Displays the locations and contents of the longwords representing a procedure call frame.

Format

SHOW CALL_FRAME [starting-address|/NEXT_FP]


Parameter

starting-address

Expression representing the starting address of the procedure call frame to be displayed. The default starting-address is the longword contained in the FP register of the SDA current process.

Qualifier

/NEXT_FP

Displays the procedure call frame starting at the address stored in the FP longword of the last call frame displayed by this command. You must have issued a SHOW CALL_FRAME command previously in the current SDA session to use the /NEXT_FP qualifier to the command.

Description

Whenever a procedure is called using CALLG or CALLS instructions, information is stored on the stack of the calling routine in the form of a procedure call frame. Figure SDA-5 illustrates the format of a call frame.6

The SHOW CALL_FRAME command interprets the contents of the designated call frame and displays whether the call frame was generated by a CALLG or CALLS instruction. If it locates nonzero bits in the portion of the second longword that represents the upper byte of the processor status word (PSW), it presents a message that indicates the fault or trap in effect. For example:


Nonzero PSW Bits (15:8) => Reserved Operand Fault on RET 

SHOW_CALL_FRAME then produces four columns of information:

Figure SDA-5 Call Frame


The SHOW CALL_FRAME command follows this listing with an indication of how many bytes were used to align the call frame to a longword boundary.

For call frames generated by a CALLS instruction, the SHOW CALL_FRAME instruction displays the argument list to the call frame in three columns containing the virtual address of each item, its contents, and its symbolic representation.

All valid procedure call frames begin on a longword boundary. If the specified address expression does not begin on a longword boundary, the call frame is invalid and SDA displays the following message:


Invalid Call Frame:   Start Address Not On Longword Boundary 

If you attempt to format an address that is not a call frame or is an invalid call frame (that is, bit 28 of the second longword is not 0), SDA displays the following message:


Invalid Call Frame:   Bit 28 is Set in "Mask-PSW" Longword 

When using the SHOW CALL_FRAME/NEXT_FP command to follow a chain of call frames, SDA signals the end of the chain by this message:


%SDA-E-NOTINPHYS, 00000000 : not in physical memory 
This message indicates that the saved FP in the previous call frame has a zero value.


Example


SDA> SHOW CALL_FRAME
Call Frame Information
----------------------
        Call Frame Generated by CALLG Instruction
 
Condition Handler       7FFE7D78  00000000
SP Align Bits = 00      7FFE7D7C  00000000
   Saved  AP            7FFE7D80  7FFE7DC0      CTL$GL_KSTKBAS+005C0
   Saved  FP            7FFE7D84  7FFE7D94      CTL$GL_KSTKBAS+00594
   Return PC            7FFE7D88  8015303F      EXCEPTION+0043F
Align Stack by 0 Bytes =>
SDA> SHOW CALL_FRAME/NEXT_FP
Call Frame Information
----------------------
        Call Frame Generated by CALLS Instruction
   
Condition Handler       7FFE7D94  00000000
SP Align Bits = 00      7FFE7D98  20FC0000
   Saved  AP            7FFE7D9C  7FFED024
   Saved  FP            7FFE7DA0  7FFE7DE4      CTL$GL_KSTKBAS+005E4
   Return PC            7FFE7DA4  801D58AA      MMG$IMGRESET+00066
        R2              7FFE7DA8  7FFE7DD0      CTL$GL_KSTKBAS+005D0
        R3              7FFE7DAC  7FFDB9F8
        R4              7FFE7DB0  8026C720
        R5              7FFE7DB4  7FFDBA00
        R6              7FFE7DB8  7FFE6300      CTL$A_DISPVEC+00500
        R7              7FFE7DBC  00000003
Align Stack by 0 Bytes =>
Argument List           7FFE7DC0  00000003
                        7FFE7DC4  7FFE7DD0      CTL$GL_KSTKBAS+005D0
                        7FFE7DC8  00000000
                        7FFE7DCC  00000000
SDA> SHOW CALL_FRAME/NEXT_FP
Call Frame Information
----------------------
        Call Frame Generated by CALLG Instruction
   
Condition Handler       7FFE7DE4  00000000
SP Align Bits = 00      7FFE7DE8  00000000
   Saved  AP            7FFE7DEC  7FFED024
   Saved  FP            7FFE7DF0  7FFECFF8
   Return PC            7FFE7DF4  8015303F      EXCEPTION+0043F
Align Stack by 0 Bytes =>
 
      

The SHOW CALL_FRAME commands in this SDA session follow a chain of call frames from that specified in the FP of the SDA current process.

Note

6 In Figure SDA-5, the second longword contains the stack pointer alignment (SPA) bits, which indicate the zero to three bytes needed to align the frame to a longword boundary. The S bit is set if the frame resulted from a CALLS instruction; it is clear if it resulted from a CALLG instruction.

SHOW CLUSTER

Displays connection manager and system communications services (SCS) information for all nodes in a cluster.

Format

SHOW CLUSTER {/CSID=csid|/NODE=name|/SCS}


Parameters

None.

Qualifiers

/CSID=csid

Displays VAXcluster system information for a specific VAXcluster member node. The value csid is the cluster system identification number (CSID) of the node to be displayed.7

/NODE=name

Displays VAXcluster system information for a specific VAXcluster member node. The value name is the name of the node to be displayed.

/SCS

Displays a view of the cluster as seen by SCS.

Description

By default, the SHOW CLUSTER command provides a view of the VAXcluster system from the perspective of the connection manager. When you use the /SCS qualifier, however, SHOW CLUSTER provides a view of the cluster from the perspective of the port driver or drivers.

VAXcluster as Seen by the Connection Manager

The SHOW CLUSTER command provides a series of displays.

The VAXcluster summary display supplies the following information:

The CSB list displays information about the VAXcluster system blocks (CSB) currently in operation; there is one CSB assigned to each node of the cluster. For each CSB, the CSB list displays the following information:

The cluster block display includes information recorded in the cluster block (CLUB), including a list of activated flags, a summary of quorum and vote information, and other data that applies to the cluster from the perspective of the node for which SDA is being run.

The cluster failover control block display provides detailed information concerning the cluster failover control block (CLUFCB), and the cluster quorum disk control block display provides detailed information from the cluster quorum disk control block (CLUDCB).

Subsequent displays provide information for each CSB listed previously in the CSB list display. Each display shows the state and flags of a CSB, as well as other specific node information. (See the Show Cluster utility section of the OpenVMS System Management Utilities Reference Manual for information about the flags for VAXcluster nodes.)

VAXcluster as Seen by the Port Driver

The SHOW CLUSTER/SCS command provides a series of displays.

The SCS listening process directory lists those processes that are listening for incoming SCS connect requests. For each of these processes, this display records the following information:

The SCS systems summary display provides the system block (SB) address, node name, system type, system ID, and the number of connection paths for each SCS system. An SCS system can be a VAXcluster member, HSC, UDA, or other such device.

Subsequent displays provide detailed information for each of the system blocks and the associated path blocks. The system block displays include the maximum message and datagram sizes, local hardware and software data, and SCS poller information. Path block displays include information that describes the connection, including remote functions and other path-related data.


Examples

#1

SDA> SHOW CLUSTER
VAXcluster data structures 
-------------------------- 
                   --- VAXcluster Summary --- 
 
         Quorum   Votes   Quorum Disk Votes   Status Summary 
         ------   -----   -----------------   -------------- 
            2       3            1            quorum 
 
 
                        --- CSB list --- 
 
Address   Node    CSID      Votes   State     Status 
-------   ----    ----      -----   -----     ------ 
 
 
803686F0  SOLLY   000100C8    1     open    member,qf_active 
80368550  GUS     000100C9    1     open    member,qf_active 
80367B90  DORIS   000100C5    1     open    member,qf_active 
 
 
                 --- Cluster Block (CLUB) 801C3F70 --- 
 
Flags: 10080001 cluster,init,quorum 
 
 
Quorum/Votes                  2/3    Last transaction code          02 
Quorum Disk Votes               1    Last trans. number           1126 
Nodes                           3    Last coordinator CSID    00000000 
Quorum Disk             $255$DUA2    Last time stamp       26-MAR-1993 
Found Node SYSID     0000000008A0                             18:52:32 
Founding Time          3-DEC-1992    Largest trans. id        00000466 
                         00:01:44    Resource Alloc. retry           0 
Index of next CSID           00D2    Figure of Merit          00000000 
Quorum Disk Cntrl Block  80334E00    Member State Seq. Num        0190 
Timer Entry Address      00000000    Foreign Cluster          00000000 
CSP Queue                   empty 
 
 
         --- Cluster Failover Control Block (CLUFCB) 801C407C --- 
 
Flags: 00000000 
 
Failover Step Index      00000028    CSB of Synchr. System    803686F0 
Failover Instance ID     00000466 
 
 
         --- Cluster Quorum Disk Control Block (CLUDCB) 80334E00 --- 
 
State: 0001 qs_not_ready 
Flags: 0000 
 
Iteration Counter               0              UCB address    00000000 
Activity Counter                0              TQE address    80419F40 
Quorum file LBN          00000000              IRP address    803665A0 
 
 
         --- SOLLY Cluster System Block (CSB) 803686F0 --- 
 
State: 01 open 
Flags: 02020302 member,cluster,qf_active,selected,status_rcvd 
 
 
Quorum/Votes       2/1    Next seq. number    0247    Send queue      00000000 
Quor. Disk Vote      1    Last seq num rcvd   0314    Resend queue    00000000 
CSID          000100C8    Last ack. seq num   0247    Block xfer Q.      empty 
Eco/Version       0/12    Unacked messages       1    CDT address     801C28F0 
Reconn. time  00000059    Ack limit              4    PDT address     801CEA20 
Ref. count           2    Incarnation  18-DEC-1993    TQE address     00000000 
Ref. time  18-DEC-1993                    08:52:20    SB address      8041B6E0 
              08:53:58    Lock mgr dir wgt       1    Current CDRP    00000000 
 
      

This example shows the screen displays for the SHOW CLUSTER command. (Displays for nodes GUS and DORIS, similar to that for node SOLLY, are also included in the SHOW CLUSTER output but have been omitted from this example.)

#2

SDA> SHOW CLUSTER /CSID=000100C8
VAXcluster data structures 
-------------------------- 
         --- SOLLY Cluster System Block (CSB) 803686F0 --- 
 
State: 01 open 
Flags: 02020302 member,cluster,qf_active,selected,status_rcvd 
 
Quorum/Votes       2/1    Next seq. number    0247    Send queue      00000000 
Quor. Disk Vote      1    Last seq num rcvd   0314    Resend queue    00000000 
CSID          000100C8    Last ack. seq num   0247    Block xfer Q.      empty 
Eco/Version       0/12    Unacked messages       1    CDT address     801C28F0 
Reconn. time  00000059    Ack limit              4    PDT address     801CEA20 
Ref. count           2    Incarnation  18-DEC-1993    TQE address     00000000 
Ref. time  18-DEC-1993                    08:52:20    SB address      8041B6E0 
              08:53:58    Lock mgr dir wgt       1    Current CDRP    00000000 
 
      

This example shows the use of the /CSID qualifier to obtain information about a specific node (in this instance, node SOLLY).

#3

SDA> SHOW CLUSTER /NODE=LEON01
VAXcluster data structures 
-------------------------- 
 
      --- LEON01 Cluster System Block (CSB) 9863BC00 --- 
 
State: 01 open 
Status 0206E1A2 member,qf_noaccess,cluster,selected,status_rcvd 
                cwps,rangelock,dyn_remaster,dts,vcc 
Cpblty 00000001 rm8sec 
 
Quorum/Votes       4/1    Next seq. number    5D8B    Send queue      987C3F80 
Quor. Disk Vote     10    Last seq num rcvd   3302    Resend queue    00000000 
CSID          00200093    Last ack. seq num   5D8A    Block xfer Q.      empty 
Eco/Version       0/24    Unacked messages       0    CDT address     9830C600 
Reconn. time  00000000    Ack limit              3    PDT address     98388590 
Ref. count           2    Incarnation  26-JAN-1993    TQE address     00000000 
Ref. time  26-JAN-1993                    15:14:37    SB address      98638140 
              15:28:43    Lock mgr dir wgt       1    Current CDRP    00000000 
 
 
      

This example shows the use of the /NODE qualifier to obtain information about a specific node (in this instance, node LEON01).

#4

SDA> SHOW CLUSTER /SCS
VAXcluster data structures 
-------------------------- 
 
                 --- SCS Listening Process Directory --- 
 
Entry Address     Connection ID     Process Name           Information 
-------------     -------------     ------------           ----------- 
 
  80419D60          08EE0000        SCS$DIRECTORY 
  80419E20          08EE0001        VMS$VAXcluster 
 
 
                   --- SCS Systems Summary --- 
 
  SB Address     Node      Type       System ID         Paths 
  ----------     ----      ----       ---------         ----- 
 
  8041A120      PINTO      HSC       00000000F10E         1 
  8041AA20      DORIS      VMS       0000000008A9         1 
  8041AB40      GUS        VMS       0000000008A1         1 
  8041B6E0      SOLLY      VMS       0000000008A0         1 
  8041D420      DODGER     HSC       00000000F00F         1 
 
 
                    --- PINTO System Block (SB) 8041A120 --- 
 
System ID            00000000F10E    Local software type           HSC 
Max message size               66    Local software vers.         X301 
Max datagram size              62    Local software incarn.   8355FE00 
Local hardware type          HS50                             008DA59A 
Local hardware vers. 022702220222    SCS poller timeout           000F 
                     022202220222    SCS poller enable mask         01 
 
 
                    --- Path Block (PB) 8041C400 --- 
 
                 Status:  0000 
 
Remote sta. addr.    00000000000E    Remote port type              HSC 
Remote state         00000000000E    Number of data paths            2 
Remote hardware rev.     00000225    Cables state            A-OK B-OK 
Remote func. mask        4F710200    Local state                  OPEN 
Resetting port                 0E    Port dev. name               PAB0 
Handshake retry cnt.            1    SCS MSGBUF address       80390270 
Msg. buf. wait queue        empty    PDT address              801CEA20 
 
 
                    --- DORIS System Block (SB) 8041AA20 --- 
 
System ID            0000000008A9    Local software type           VMS 
Max message size              112    Local software vers.         V5.0 
Max datagram size             576    Local software incarn.   A9D31760 
Local hardware type          V780                             008DA59B 
Local hardware vers. 010E0138207A    SCS poller timeout           000C 
                     000030030E10    SCS poller enable mask         00 
 
 
                    --- Path Block (PB) 80437E80 --- 
 
                Status:  0000 
 
Remote sta. addr.    000000000002    Remote port type            CI780 
Remote state                 ENAB    Number of data paths            2 
Remote hardware rev.     00040003    Cables state            A-OK B-OK 
Remote func. mask        FFFFFF00    Local state                  OPEN 
Resetting port                 02    Port dev. name               PAB0 
Handshake retry cnt.            1    SCS MSGBUF address       8036F0B0 
Msg. buf. wait queue        empty    PDT address              801CEA20 
 
      

This example shows a subset of a typical output for the SHOW CLUSTER/SCS command. In this system, there are three nodes (DORIS, GUS, and SOLLY), and there are two HSCs (PINTO and DODGER). After the summary information in the first two screen displays, specific information for each system block and its associated path block is shown.

Note

7 You can find the CSID for a specific node in a cluster by examining the CSB list display of the SHOW CLUSTER command. Other SDA displays refer to a system's CSID. For instance, the SHOW LOCK command indicates where a lock is mastered or held by CSID.
8 For information about the state and status of nodes, see the description of the ADD command in the Show Cluster utility section of the OpenVMS System Management Utilities Reference Manual.


Previous Next Contents Index

  [Go to the documentation home page] [How to order documentation] [Help on this site] [How to contact us]  
  privacy and legal statement  
4556PRO_008.HTML