HP3000-L Archives

June 1996, Week 4

HP3000-L@RAVEN.UTC.EDU

Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
Reply To:
Date:
Fri, 28 Jun 1996 09:10:00 +0930
Content-Type:
text/plain
Parts/Attachments:
text/plain (157 lines)
where is TCP_MAX_CONNS configured?
----------
From: HP3000-L; JEFF
To: HP3000-L
Subject: Re: PB wtith Reflection and tcpip 3.11
Date: Friday, 28 June 1996 6:38AM
 
<<File Attachment: REPBWTIT.TXT>>
On Thu, 27 Jun 1996 14:56:29 GMT Eero Laurila said:
>Steve Dirickson b894 WestWin ([log in to unmask]) wrote:
>: <<I use Reflection 5.0 and Tcpip 3.11b from Microsoft to coonect to my
>: HP968. Sometimes, I have this message at my console "NS3000 Netipc error
>: error 42" and the vt sessions never close. After 2 or 3 weeks, I'm
>: unable to connect to the HP968 trough the lan because I ghave over 100
>: sessions unclosed and I need to make ctrl a shutdown and a start
>: norecovery to be able to reconnect. HP Supports is unable to find a
>: solution>>
>
>  Ok, the VT error 42...
>  in ASCAT.NET.SYS you'll find the error as:
>  42 REMOTE NOT RESPONDING, CONNECTION CLOSED (VT WARN 042)
>  ...defined in vtserver source as follows:
>  rc_ipc_close =  042;        { Partner closed socket.  }
>  i.e. the connection went down due to some reason other than the VTSERVER
>  willing to close the socket.  Looking at the source shows the actual NetIPC
>  errors that are folded to to the same error, for example:
>   36  8774.000   2       if ( ipc_r_code.int = socket_closed  ) OR
>   37  8775.000   2          ( ipc_r_code.int = socket_aborted ) OR
>   37  8776.000   2          ( ipc_r_code.int = socket_graceful_release )
>   37  8777.000   2       then return_code := rc_ipc_close { Partner stopped }
 
<*Ding!*> this finally rang a bell.  Maybe the wrong bell, but maybe not.
 
Are you using BSD sockets much?  Check your buffers:
   :nettool.net.sys "resource;display;quit"
 
Check the number of TCP connections in use versus your max defined:
 
:print tcpuse
 symopen symnsxpt.xpt0407.telesup
 var tcp_kso_ptr:lptr=[c0000000+#333*8].[c0000000+#333*8+4]
 var max_cons = &
     symval(tcp_kso_ptr 'tcp_kso_block_type.TG_MAX_CONNECTIONS_ALLOWED')
 var cur_cons = &
  symval(tcp_kso_ptr 'tcp_kso_block_type.TG_NUMBER_OF_OPEN_CONNECTIONS')
 symclose symnsxpt
 wl cisetvar ("tcp_cur_conns", !cur_cons)
 wl cisetvar ("tcp_max_conns", !max_cons)
 c
:debug
DEBUG/iX B.79.06
 
HPDEBUG Intrinsic at: a.009dc390 hxdebug+$e4
$1 ($a0) nmdebug > use tcpuse
TRUE
TRUE
:showvar tcp@
TCP_CUR_CONNS = 93
TCP_MAX_CONNS = 4096
 
There was a sneaky problem in 5.0 with "persistant" BSD sockets staying in
use although they would only show up in debug.  When this happens, it
appears as though network services are down (but they aren't).  I have seen
cases of running out of connections and/or out of buffers causing those
unusual VT errors.  When transport gets hung up, the TCP keepalives soon
disappear (or get rejected going out) and the VT sessions drop as if the
session was aborted on the client.
 
That's my guess for what it's worth :-)
 
You should also try the NSCONTROL KILLSESS= to get rid of the pervasive
sessions.
 
Jeff Kell <[log in to unmask]>
 
.......................................................................
 
Item Subject: WINMAIL.DAT
Could not convert Microsoft Mail Message Data item to text.
Will attempt to 'shar' item as file '0080fej' at end of msg.
 
 
# This is a shell archive.  Remove anything before this line,
# then unpack it by saving it in a file and typing "sh file".
#
# Wrapped by OpenMail System <openmail@southcor> on Fri Jun 28 09:10:38 1996
#
# This archive contains:
#       0080fej
#
# Error checking via wc(1) will be performed.
 
LANG=""; export LANG
PATH=/bin:/usr/bin:$PATH; export PATH
 
 
rm -f /tmp/uud$$
(echo "begin 666 /tmp/uud$$\n#;VL*n#6%@x\n \nend" | uudecode) >/dev/null 2>&1
if [ X"`cat /tmp/uud$$ 2>&1`" = Xok ]
then
        unpacker=uudecode
else
        echo Compiling unpacker for non-ascii files
        pwd=`pwd`; cd /tmp
        cat >unpack$$.c <<'EOF'
#include <stdio.h>
#define C (*p++ - ' ' & 077)
main()
{
        int n;
        char buf[128], *p, a,b;
 
        scanf("begin %o ", &n);
        gets(buf);
 
        if (freopen(buf, "w", stdout) == NULL) {
                perror(buf);
                exit(1);
        }
 
        while (gets(p=buf) && (n=C)) {
                while (n>0) {
                        a = C;
                        if (n-- > 0) putchar(a << 2 | (b=C) >> 4);
                        if (n-- > 0) putchar(b << 4 | (a=C) >> 2);
                        if (n-- > 0) putchar(a << 6 | C);
                }
        }
        exit(0);
}
EOF
        cc -o unpack$$ unpack$$.c
        rm unpack$$.c
        cd $pwd
        unpacker=/tmp/unpack$$
fi
rm -f /tmp/uud$$
 
echo x - 0080fej '[non-ascii]'
$unpacker <<'@eof'
begin 660 0080fej
M>)\^(@   0N  0 G    .38P-C(W+C$U,S@T-BY%1%0N2D5&1BAA*5540U9-X
++E540RY%1%4 8@DG                                            X
                                                             X
end
@eof
set `wc -lwc <0080fej`
if test $1$2$3 != 0156
then
        echo ERROR: wc results of 0080fej are $* should be 0 1 56
fi
 
chmod 660 0080fej
 
rm -f /tmp/unpack$$
exit 0

ATOM RSS1 RSS2