Today I did a little bit digging into my RBL config and used TCPDUMP.
But after some seconds/minutes the subprocess aborted with an ACCVIO:

$ SHOW SYMBOL BG
BG == "SPAWN/NOWAIT/NOTIFY/INPUT=NL:/NOKEY/OUTPUT=BG"
$ BG TCPDUMP UDP PORT 53
....
18:20:31.362804 ns1.langstoeger.at.57175 > k.gtld-servers.net.domain: 23443 [b2&3=0x10] [1au] A? ns2.pch.net. (40)
%SYSTEM-F-ACCVIO, access violation, reason mask=00, virtual address=0000000000000870, PC=000000000009D5C0, PS=0000001B

Improperly handled condition, image exit forced.
Signal arguments: Number = 0000000000000005
Name = 000000000000000C
0000000000000000
0000000000000870
000000000009D5C0
000000000000001B

Register dump:
R0 = 00000000001A320D R1 = 0000000000000001 R2 = 0000000000016910
R3 = 00000000001A31E0 R4 = 00000000001A31F9 R5 = 0000000000000001
R6 = 0000000000000003 R7 = 0000000000000870 R8 = 0000000000000000
R9 = 0000000000000000 R10 = 0000000000000035 R11 = 000000000000DF57
R12 = 0000000000000000 R13 = 0000000000000001 R14 = 000000000000DF57
R15 = 000000007BF160F0 R16 = 000000007BF1B594 R17 = 0000000000000000
R18 = 0000000100000000 R19 = 0000000000000000 R20 = 0000000000000000
R21 = 0000000000000000 R22 = 0000000000000000 R23 = 0000000000000001
R24 = 0000000000000000 R25 = 0000000000000000 R26 = 000000000009D5C0
R27 = 000000007B646E20 R28 = 000000007BF56010 R29 = 000000007AE0D670
SP = 000000007AE0D660 PC = 000000000009D5C0 PS = 200000000000001B
18:20:31.407740 k.gtld-servers.net.domain > ns1.langstoeger.at.57175: 23443- 1/3/4 A puck.nether.net


Note the last line after the ACCVIO dump lines (must have been in the buffer)


Maybe, a subprocess is a bad idea for TCPDUMP, so I started it directly,
but alas, even on another Alpha (started a couple of minutes before) it
failed after some seconds/minutes with

....
18:17:53.378897 k.gtld-servers.net.domain > ns.langstoeger.at.8530: 20713- 1/2/2
A ns.psi.net
%SYSTEM-F-ACCVIO, access violation, reason mask=00, virtual address=000EEEB00038
0001, PC=000000000009D5C0, PS=0000001B

Improperly handled condition, image exit forced.
Signal arguments: Number = 0000000000000005
Name = 000000000000000C
0000000000000000
000EEEB000380001
000000000009D5C0
000000000000001B

Register dump:
R0 = 00000000001B4D8C R1 = 0000000000000001 R2 = 0000000000016910
R3 = 00000000001B4D60 R4 = 00000000001B4D78 R5 = 0000000000000001
R6 = 0000000000000038 R7 = 000EEEB000380001 R8 = 0000000000000000
R9 = 0000000000000000 R10 = 0000000000000035 R11 = 0000000000002152
R12 = 0000000000000000 R13 = 0000000000000001 R14 = 0000000000002152
R15 = 000000007BF160F0 R16 = 000000007BF1B594 R17 = 0000000000000000
R18 = 0000000100000000 R19 = 0000000000000000 R20 = 000000007FFD0010
R21 = 0000000000000008 R22 = FFFFFFFFFFFFFFFF R23 = 0000000000000001
R24 = 0000000000000001 R25 = 0000000000000000 R26 = 000000000009D5C0
R27 = 000000007B646E20 R28 = 0000000000000000 R29 = 000000007ADF5670
SP = 000000007ADF5660 PC = 000000000009D5C0 PS = 200000000000001B

Note: You see no timestamp of the ACCVIO itself.

I could increase the buffers of course (if this fixes it), but is this ACCVIO
not something needing to be fixed in the source?

TIA

--
Peter "EPLAN" LANGSTÖGER
Network and OpenVMS system specialist
E-mail Peter@LANGSTOeGER.at
A-1030 VIENNA AUSTRIA I'm not a pessimist, I'm a realist