Conference abbott::mailworks-vms

1559.0. "AXP system crash upon clients INBOX opening " by GIDDAY::LEH () Wed Mar 19 1997 21:54

An AXP system running UCX 4.1-ECO 2 with MailWorks 1.3A-3 ICF#35 crashed 
upon some activity from a dial up PPP client on Win95 running TeamLisnk 2.5 
ECO3. It was recorded:

1997031616152084 CON>S009BA408, Connect request from user a_john, node
        netrider-2.minenco.cra.com.au.
1997031616152093 CON> Node, netrider-2.minenco.cra.com.au, too long to log.
1997031616152104 CON>S009BA408, User a_john connected from node
        netrider-2.minenco.cra.com.au

and later confirmed by the user that the connection seemed to work OK normal 
with 'usual' mail tasks, except trying to open the MailWorks' Inbox when the 
client PC seemed to be hang in hourglass mode and later, system crash occurred. 

Rollback to 1.3A-3 without ICF#35 still made no difference on this 
reproducible problem; however, UCX rollback to 4.0 at some ECO seemed to have 
averted the problem for the time being.

As Netrider (network access) server service was involved in connecting this 
client to DMW server, the client mail notify address was  
	netrider-2.minenco.cra.com.au

and hence the warning/error "too long to log"

Is this address acceptable to the Connect/MUAS servers, as the system dump
showed:
        R2  = FFFFFFFF 88052268  UCX$BGDRIVER_NPRW+05C68
        R3  = FFFFFFFF 80FB04A0
        R4  = 00000000 00000020
        R5  = 00000000 00000001
        R6  = 00000000 00000000
        R7  = FFFFFFFF 880552B0  UCX$BGDRIVER_NPRW+08CB0
        PC  = FFFFFFFF 803E3FA0  UCX$BGDRIVER_NPRO+39FA0
        PS  = 20000000 00000803

Saved Scratch Registers in Mechanism Array
------------------------------------------

R0  = 00000000 00000150   R1  = 00000000 00000001   R16 = 00000000 00000000
R17 = 00000000 00000158   R18 = 00000000 00000000   R19 = 00000000 00000007
R20 = 00000000 00010000   R21 = 00000000 00000007   R22 = 00000000 00000000
R23 = 00000000 00000007   R24 = 00000000 00000000   R25 = 00000000 80F9FED8
R26 = 00000000 00000000   R27 = 00000000 00000000   R28 = FFFFFFFF 803DCA9C

CPU 00 reason for Bugcheck: INVEXCEPTN, Exception while above ASTDEL

Process currently executing on this CPU: A1MAIL$CONNECT

Current image file: $4$DKB0:[SYS0.SYSCOMMON.][SYSEXE]A1MAIL$CONNECT.EXE;2
..................

Thanks for any comment

Hong

    
    Also see note 1557 - interesting that this also reports problems after
    UCX upgrade.
    
    Cheers,
    Arfon.