Method or standard for selecting port numbers?

[Mark McDaid]

I was wondering if anyone would be willing to share their organization’s method for selecting port numbers for new interfaces.  Do you use specific port ranges for specific systems (lab, radiology, reg, etc.)?  Do you have a standardized methodology or best practice?  I would appreciate any insight any of the veterans could provide on the subject.

Thanks,

Topic

[Tom Rioux]

This is the way that we do it here:

20xxx Server Ports on Test

21xxx Server Ports on Prod

30xxx Intersite or interprocess connections on Test

31xxx Intersite or interprocess connections on Prod

So, if we have an interface that is using port 20123 in test, that same interface will use 21123 when we move it to prod.

There are exceptions to every rule and we deal with them when and if they arise.

Tom

[Nate Kruse]

We have no method.  It is disorganized.  It’s great.  🙄

[Robert Milfajt]

We use the same method as Thomas, except we use the third digit to differentiate interface type.

For example, 200xx are production ADT interfaces, while 213xx are results interfaces in test, etc.

Bob

Robert Milfajt
Northwestern Medicine
Chicago, IL

[Mark McDaid]

Well, we use the same method as Nate, which is why I was asking.  😳  I appreciate everyone’s responses.  Now I’ve got some ideas to think about for future implementations.

[Max Drown (Infor)]

I use a script that lists all the ports used by Cloverleaf. I often will pipe the output to grep, sort, etc. depending on what I need. Then I select the next number available for my new tcp/ip thread.

Code:

#!/qdx/qdx5.7/integrator/bin/tcl

set hciRoot /qdx/qdx5.7/integrator

# List of sites to exclude from the search
set siteExclusions {}

# Get site names from the server.ini
set sites [exec grep environs $hciRoot/server/server.ini]
regsub {environs=} $sites {} sites
set sites [lsort [split $sites ;]]

foreach site $sites {
   set site [file tail $site]
   if {[lsearch -exact $siteExclusions $site] == -1} {
       eval [exec $hciRoot/sbin/hcisetenv -root tcl $hciRoot $site]

       netcfgLoad $hciRoot/$site/NetConfig
       set connList [lsort [netcfgGetConnList]]

       foreach conn $connList {
           set protocol “”
           set host “”
           set port “”
           set pdlType “”

           set data [netcfgGetConnData $conn]

           keylget data PROTOCOL.TYPE protocol
           keylget data PROTOCOL.HOST host
           keylget data PROTOCOL.PORT port
           keylget data PROTOCOL.PDLTYPE pdlType

           # Get the name of the process
           set klst [netcfgGetConnData $conn]
           set process [keylget klst PROCESSNAME]

           # Get the “Summary:” line from the notes and output the results
           set note “”
           set notesFile “$hciRoot/$site/notes/NetConfig/$process/thread_$conn.notes”
           if {[file exists $notesFile]} {
               catch {exec /bin/grep Summary: $notesFile} note
               # Ignore this warning from grep
               if {$note == “child process exited abnormally”} {set note “”}
               # Remove the “Summary:” text
               regsub — {^Summary: ?} $note {} note
           }

           # Slightly different output for tcp/ip clients and servers
           if {$protocol == “pdl-tcpip”} {
               set protocol $pdlType
           }

           # Don’t output notes if there is no Summary: line in the notes
           if {$note == “”} {
               puts $site:$conn:$protocol:$host:$port
           } else {
               puts $site:$conn:$protocol:$host:$port:$note
           }
       }
   }
}

# End of Script

-- Max Drown (Infor)

[Tim Faul]

Nate Kruse wrote:

We have no method.

[Robert Kersemakers]

As I don’t have direct access to /etc/services (bloody network dictators…) I made a copy of /etc/services to keep track of the port numbers in use.

Zuyderland Medisch Centrum; Heerlen/Sittard; The Netherlands

[Author]

Replies