Conference csc32::consolemanager

    Alain,
      We have fixed several "memory leaks" since the ECO1 kit. These are
    available in the FT version of next ECO kit. It is more than likely
    that your problem is being caused by one of these leaks. You can get
    the latest build from
    
    CSC32::DUMPS:[BUTTERWORTH.PCM]
    
    Just do DIR and grab the appropriate saveset for your environment. As
    satted in other notes theses are not kits but just savesets with new
    images. There is alos a text file inthe same dir with the necessary
    restore commands.
    
    REgs,
      Dan

Hi Dan,

Is there a reason for which the images, coming from the csc_126_axp set, aren't link with notraceback.

Any idea when the next ECO kit will be release ?

Best regards and thanks for your support.

Alain



   
 

     Dan,
            I to am seeing these messages on v1.6-125 Vax ovms. Are you
    saying that the 'patch' kit ..126 should resolve this?   And if so can I
    install it at customer site yet?
                                                    
        Regards Alan Brodie @ eso
    

    Alan and Alain,
       First of all the AXP kit has been relinked with /NOTRACE. Sorry if
    that inconvenienced you in any way!!
    
    Alan,
      The ENS memory leaks were all resolved quite a while back but I'd
    like you to put on 126 just the same.  Bear in mind that it's possbile
    for a multi-line window to pop-up during shutdown of the software if
    you've got a firly fast system. What happpens is the shutdown event
    causes all exisiting eventlists to exit if you have that button
    selected in the eventlist options menu. This causes ENS to clean up 
    all the info about running actions. The shutdown event can also cause
    a match against a filter as it's an event and guess what, if a filter
    match occurs and it's supposed to start an eventlist then ENS does
    that. The process is created but when it tries to connect back in, ENS
    daemon has terminated and thus the new eventlist process has no one
    to talk to so it outputs the "failed to connect" message.
    
    Now it's possbile that the ENS daemon is dying for some reason which
    will also cause the message. If it happens again after you have applied
    126, please post the CONSOLE$TMP:CONSOLE$ENS_DAEMON.LOG. Be sure and
    grab it before you restart PCM otherwise the startup file will nuke it.
    
    Regs,
      Dan

Dan,

Do you also notice the following ? :

Console c3 gives me the following output (on AXP and VAX) :

POLYCENTER Console Manager
C3 Version V1.6-126
Copyright (c) 1995 Digital Equipment Corporation. All Rights Reserved

POLYCENTER Console Manager
Eventlist Version V1.6-125  <==
Copyright (c) 1995 Digital Equipment Corporation. All Rights Reserved
  
Also, the editor shows a version 125.
I checked all exe (analyze/image), they are all indicate version 126.
Is this simply text ?

Secondly, an archive problem (no response from controller...)that was solved with NSI_117_AXP reappeared with this
set.
There is an IPMT case (BRO100664) I opened some times ago, please have a look at it (you can reproduce it at will).


Regards,

Alain.

>Dan,

>Do you also notice the following ? :

>Console c3 gives me the following output (on AXP and VAX) :

>POLYCENTER Console Manager
>C3 Version V1.6-126
>Copyright (c) 1995 Digital Equipment Corporation. All Rights Reserved

>POLYCENTER Console Manager
>Eventlist Version V1.6-125  <==
>Copyright (c) 1995 Digital Equipment Corporation. All Rights Reserved
  
>Also, the editor shows a version 125.
>I checked all exe (analyze/image), they are all indicate version 126.
>Is this simply text ?
    
    Yes. This is from the Eventlist application that the C3 starts
    up automaitically. This happens because of the way builds are
    done. If the module that uses this text doesn't change between
    builds it is not recompiled thus it still has the old text in
    the object module. When the final ECO2 build is done, a full
    recompilation will be done.

>Secondly, an archive problem (no response from controller...)that was solved with NSI_117_AXP reappeared with this
>set.
>There is an IPMT case (BRO100664) I opened some times ago, please have a look at it (you can reproduce it at will).
    
    Interesting. We have never fixed this problem in the VAX stream
    (the source modules are the same) and I have had my AXP customers
    have the same problem all along. I didn't know it was ever fixed.
    Something is either skewed between the build systems or maybe
    it really was never fixed. 
    
    BTW, since you can reproduce it I would like to ask a favor. 
    Could you try doing an ARCHIVE when there are *no* CONMON
    sessions as well as when theare are several and tell me it
    makes a difference. I have a theory as to why this is 
    happening and the results of the above would be interesting.
    Oh, and also make sure there are no C3's running. The point is
    I want to see  what happens when there are no user interfaces
    running. 
    
    Regards,
       Dan

Regs,Dan

Hi Dan,

- No C3 running. 

***First test*** : there is no sessions connected to the console or monitoring it.  
		   Archive runs without problem.

***Second test*** : Connect the console.
                    archive the system, it aborts normally...
                    unlock the console (console dialog)
                    archive the system => no response from controller for xx seconds  Waiting ...
                                       => access violation (see outputs here below)

- I didn't notice any difference when c3 is running or not.
- Same behaviour for VAX and AXP. 

Best regards,

Alain.                    

------------------------------------------------------------------------------------------------------------------
***First test***

$ console arch brsphw

POLYCENTER Console Manager
Archive facility Version V1.6-126
Copyright (c) 1995 Digital Equipment Corporation. All Rights Reserved


Starting archive procedure for system brsphw
Timefile  Pass 1: Working ... done.
Timefile  Pass 2: Working ... done.
Logfile   Pass 1: Working ... done.
Logfile   Pass 2: Working ... done.
Eventfile Pass 1: Working ... done.
Updating  Pass 1: Working ... done.
Archive procedure for system brsphw completed successfully


------------------------------------------------------------------------------------------------------------------
***Second test*** 

$ console arch brsphw

POLYCENTER Console Manager
Archive facility Version V1.6-126
Copyright (c) 1995 Digital Equipment Corporation. All Rights Reserved


Starting archive procedure for system brsphw
Archive aborted - user SYSTEM is connected

$ console dialog/file=t.t brsphw
POLYCENTER Console Manager
Dialog facility version V1.6-126
Copyright (c) 1995 Digital Equipment Corporation. All Rights Reserved

$ console arch brsphw

POLYCENTER Console Manager
Archive facility Version V1.6-126
Copyright (c) 1995 Digital Equipment Corporation. All Rights Reserved


Starting archive procedure for system brsphw
 Archive: No response from Controller for 10 seconds  Waiting...
 Archive: No response from Controller for 20 seconds  Waiting...

------------------------------------------------------------------------------------------------------------------
***Second test*** ( console output) 

Connected to remote console (brsphw)
Type CTRL/E to exit or CTRL/G to escape to monitor
Console has been unlocked by another user
Type CTRL/E to exit or CTRL/G to escape to monitor

%SYSTEM-F-ACCVIO, access violation, reason mask=00, virtual address=00000019, PB

  Improperly handled condition, image exit forced.
    Signal arguments:   Number = 00000005
                        Name   = 0000000C
                                 00000000
                                 00000019
                                 00043244
                                 0000001B

    Re -1.
    
    *Thank you very much*. That gives me something to go on! Excellent
    work!
    
    REgards,
       Dan