Mac OSX Leopard and server connections

Paul Looney support at ahsomme.com
Sat Feb 6 01:05:53 CST 2010


Dan,
We have found this problem on OS X going back to Panther, 10.3.9.
In our experience it happens when attempting to mount two volumes (or  
share points) with the same name. OS X will label the second one "The  
Same Name-1" but the "-1" is hidden.
This can happen when a volume is remounted before it is completely  
dismounted. It can happen when inserting a flash drive (or any other  
type of drive) with the same name as one already mounted. Jeanne  
DeVoto mentioned that she noticed it when reinserting ejectable  
cartridges too rapidly.
The esteemed Phil Davis has written us a program called "Connection  
Repair" which fixes this for the network-connected volume we use; it  
unmounts all volumes with that name. If there is a sufficient need  
for this we could generalize it and make it available to all Rev users.
Paul Looney

On Feb 5, 2010, at 8:57 PM, Sarah Reichelt wrote:

>> I have a rather odd problem using rev applications on Mac OSX  
>> Leopard, connecting to a server on my network. I connect to a  
>> server with an Applescript command: do theScript as Applescript. I  
>> can do this just fine. But if I unmount the server  by dragging  
>> it's icon to the trash, then remount from within Rev eventually  
>> Rev reads the name of the server differently. Let's say the name  
>> of the server is "George". I can connect to "George" several  
>> times, but eventually Rev or Leopard appends a digit to the name  
>> of the mounted volume.
>>
>> Originally, the path to file "Textfile.txt" on the server named  
>> "George" will be "/Volumes/George/Textfile.txt" However, after a  
>> few connections (don't know how many or exactly when this  
>> happens), the path to the file becomes "/Volumes/George-1/ 
>> Textfile.txt". Oddly enough, the icon on the desktop for the  
>> volume is still "George". However, any reference to the file  
>> "Textfile.txt" from within Rev now will fail, since Rev reads the  
>> connection as "/Volumes/George-1/Textfile.txt" but is still  
>> looking for "/Volumes/George/Textfile.txt"!  This gets somewhat  
>> frustrating, since I never know when it will happen, and once it  
>> does, I can no longer transfer files back and forth between the  
>> server and the computer.
>>
>> Has anyone seen this behavior? Do you have a workaround or a  
>> suggestion to get around this problem? I never had the problem  
>> working with pre-Leopard operating systems on the Mac, but it is  
>> now cropping up since all our computers in a lab are now running  
>> Leopard.
>>
>> Another issue may be that the physical server (now running OSX  
>> Snow Leopard) has several share points, each of which may be  
>> mounted as individual volumes, one of which is the aforementioned  
>> "George". Mounting and unmounting several of these share points  
>> may bring about the re-numbering issue. Rebooting the individual  
>> computer doesn't seem to help either.
>
>
> This isn't just a Snow Leopard problem but has been around for a long
> time. Here is the script I use for making network connections. The
> example connects to a Public folder on a remote machine.
> It includes a check for the shared folder not having been properly
> disconnected, so avoids the -1 problem.
>
>
> local sLastLoginAttempt
>
> -- connect to remote server
> --
> on doRemoteLogin
>     -- retrieve these 3 parameters from custom props or other data  
> storage
>     put the cRemoteIP of this stack into remoteIP
>     put the cLoginName of this stack into login
>     put base64decode(the cLoginPass of this stack) into pword
>
>     put "/Volumes/" & login & "/Public/" into tRemoteFolder
>
>     if the volumes contains login or there is a folder  
> tRemoteFolder then
>         exit doRemoteLogin  -- already logged in
>     end if
>
>     if remoteIP is empty OR remoteIP = "localhost" OR remoteIP =
> "127.0.0.1" then
>         exit doRemoteLogin  -- getting data locally
>     end if
>
>     if login is empty OR pword is empty then
>         exit doRemoteLogin  -- not set up yet
>     end if
>
>     -- don't try more than once every 10 minutes, or it can lock up  
> the program
>     if the seconds - sLastLoginAttempt < 600 then
>         exit doRemoteLogin
>     else
>         put the seconds into sLastLoginAttempt
>     end if
>
>     --  Shell commands:
>     --  mkdir /Volumes/emonitor
>     --  mount_afp afp://emonitor:jade05@192.168.0.101/emonitor / 
> Volumes/emonitor
>     -- for a background task: "command > file_path 2>&1 &" or "command
>> /dev/null 2>&1 &"
>
>     set the cursor to watch
>     if there is a folder ("/Volumes/" & login) then
>         -- was connected but didn't disconnect properly
>         put "rm -d /Volumes/" & login into tShellCmd
>         get shell(tShellCmd)
>     end if
>     put "mkdir /Volumes/" & login into tShellCmd
>     get shell(tShellCmd)
>     put "mount_afp afp://" & login & ":" & pword & "@" into tShellCmd
>     put remoteIP & "/" & login && "/Volumes/" & login after tShellCmd
>     put " > /dev/null 2>&1 &" after tShellCmd
>     get shell(tShellCmd)
>
>     -- keep checking for 10 seconds, then give up
>     repeat 10 times
>         set the cursor to busy
>         wait 1 second with messages
>         if the volumes contains login or there is a folder  
> tRemoteFolder then
>              exit doRemoteLogin
>         end if
>     end repeat
> end doRemoteLogin
>
>
> HTH,
> Sarah
> _______________________________________________
> use-revolution mailing list
> use-revolution at lists.runrev.com
> Please visit this url to subscribe, unsubscribe and manage your  
> subscription preferences:
> http://lists.runrev.com/mailman/listinfo/use-revolution




More information about the use-livecode mailing list