This is a discussion on Find the source IP address of a telnet connection. within the HP-UX Operating System forums, part of the Unix Operating Systems category; --> Can anyone help with tracking the source IP address of a telnet connection under 11.23i. We currently do this ...
| |||||||
| FAQ | Members List | Calendar | Search | Today's Posts | Mark Forums Read |
| ||||
| Can anyone help with tracking the source IP address of a telnet connection under 11.23i. We currently do this using lsof but lsof under 11.23i has a problem with identifying telnet connections. I need to be able to track it by using the connection's tty (e.g. pts/1) as a starting point. TIA |
| |||
| robwlindsay@gmail.com wrote: > Can anyone help with tracking the source IP address of a telnet > connection under 11.23i. 11.23i? Is that perhaps mashing the kernel naming - ie 11.23 with the Operating environment naming - ie 11iv2? > We currently do this using lsof but lsof under 11.23i has a problem > with identifying telnet connections. Can you be more specific? > I need to be able to track it by using the connection's tty (e.g. > pts/1) as a starting point. > TIA -- a wide gulf separates "what if" from "if only" these opinions are mine, all mine; HP might not want them anyway... feel free to post, OR email to rick.jones2 in hp.com but NOT BOTH... |
| |||
| robwlindsay@gmail.com wrote: > Can anyone help with tracking the source IP address of a telnet > connection under 11.23i. We currently do this using lsof but lsof under > 11.23i has a problem with identifying telnet connections. > > I need to be able to track it by using the connection's tty (e.g. > pts/1) as a starting point. > > TIA To be more precise, the HP-UX kernel's PSTAT implementation from which lsof obtains open file information has a problem. That problem was once fixed in HP-UX 11.11, but the fix was lost in HP-UX 11.23 for all architectures. I have been told that fix will be restored by a depot available "real soon now." I've tested a preliminary version of it and it appears to restore the delivery of telnet TCP/IP connection parameters to lsof. To track or locate the depot, refer to defect JAGag05130. Unfortunately the patch in the depot does not fix a similar problem that exists for nfsd, reported in defect JAGaf96371 in November 2005. The PSTAT kernel implementation does not deliver its UDP/IP parameters, either. While both are related to the failure of PSTAT to report stream sockets properly by following their module chains to their private TCP/IP and IDP/IP information structures, the nfsd UDP/IP solution to this defect has been declared an enhancement request by the HP NetUX Transport Lab and there is so far no commitment to fix it. Vic Abell, lsof author |
| |||
| Thanks for the info ... Any suggestions on tracking the source of a telnet connection without using lsof ... until the patch is delivered? abe@purdue.edu wrote: > robwlindsay@gmail.com wrote: > > Can anyone help with tracking the source IP address of a telnet > > connection under 11.23i. We currently do this using lsof but lsof under > > 11.23i has a problem with identifying telnet connections. > > > > I need to be able to track it by using the connection's tty (e.g. > > pts/1) as a starting point. > > > > TIA > > To be more precise, the HP-UX kernel's PSTAT implementation from which > lsof > obtains open file information has a problem. That problem was once > fixed in HP-UX > 11.11, but the fix was lost in HP-UX 11.23 for all architectures. > > I have been told that fix will be restored by a depot available "real > soon now." I've > tested a preliminary version of it and it appears to restore the > delivery of telnet TCP/IP > connection parameters to lsof. To track or locate the depot, refer to > defect > JAGag05130. > > Unfortunately the patch in the depot does not fix a similar problem > that exists for nfsd, > reported in defect JAGaf96371 in November 2005. The PSTAT kernel > implementation > does not deliver its UDP/IP parameters, either. While both are related > to the failure of > PSTAT to report stream sockets properly by following their module > chains to their > private TCP/IP and IDP/IP information structures, the nfsd UDP/IP > solution to this > defect has been declared an enhancement request by the HP NetUX > Transport Lab > and there is so far no commitment to fix it. > > Vic Abell, lsof author |
| |||
| The only suggestion I can offer is to see if your HP support people can get access to the pre-release copy of the patch against which I have tested lsof. Vic robwlindsay@gmail.com wrote: > Thanks for the info ... > > Any suggestions on tracking the source of a telnet connection without > using lsof ... until the patch is delivered? > > abe@purdue.edu wrote: > > > robwlindsay@gmail.com wrote: > > > Can anyone help with tracking the source IP address of a telnet > > > connection under 11.23i. We currently do this using lsof but lsof under > > > 11.23i has a problem with identifying telnet connections. > > > > > > I need to be able to track it by using the connection's tty (e.g. > > > pts/1) as a starting point. > > > > > > TIA > > > > To be more precise, the HP-UX kernel's PSTAT implementation from which > > lsof > > obtains open file information has a problem. That problem was once > > fixed in HP-UX > > 11.11, but the fix was lost in HP-UX 11.23 for all architectures. > > > > I have been told that fix will be restored by a depot available "real > > soon now." I've > > tested a preliminary version of it and it appears to restore the > > delivery of telnet TCP/IP > > connection parameters to lsof. To track or locate the depot, refer to > > defect > > JAGag05130. > > > > Unfortunately the patch in the depot does not fix a similar problem > > that exists for nfsd, > > reported in defect JAGaf96371 in November 2005. The PSTAT kernel > > implementation > > does not deliver its UDP/IP parameters, either. While both are related > > to the failure of > > PSTAT to report stream sockets properly by following their module > > chains to their > > private TCP/IP and IDP/IP information structures, the nfsd UDP/IP > > solution to this > > defect has been declared an enhancement request by the HP NetUX > > Transport Lab > > and there is so far no commitment to fix it. > > > > Vic Abell, lsof author |
| |||
| robwlindsay@gmail.com writes: > Thanks for the info ... > > Any suggestions on tracking the source of a telnet connection without > using lsof ... until the patch is delivered? "who -R"? Regards, Ulrich |
| ||||
| # Status de Conecciones Telnet en HP-UX por Ricardo Pelcastre # MexLord - mexlord@hotmail.com #!/opt/perl/bin/perl %dias = (Sun => "Dom", Mon => "Lun", Tue => "Mar", Wed => "Mie", Thu => "Jue", Fri => "Vie", Sat => "Sab" ); %meses = (Jan => "Ene", Feb => "Feb", Mar => "Mar", Apr => "Abr", May => "May", Jun => "Jun", Jul => "Jul", Aug => "Ago", Sep => "Sep", Oct => "Oct", Nov => "Nov", Dec => "Dic" ); $ip=$ARGV[0]; $cmd="/usr/bin/grep \"$ip\" /var/adm/syslog/syslog.log\n"; @lineas=qx($cmd); foreach (@lineas) { if (/telnet\/tcp: Connection from\s+(\S+)\s+\((\S+)\) at (\S+) (\S+)\s+(\S+) (\S+) (\S+)$/) { $salida=sprintf("[%s %2s %s %s] - %s => %s\n",$dias{$3},$5,$meses{$4},$6,$1,$2); print $salida; } } |