| Re: IDL and NIS - cd? [message #2317 is a reply to message #2160] |
Thu, 09 June 1994 08:25  |
Geoff.Sobering
Messages: 14
Registered: June 1994
|
Junior Member |
|
|
In article <2t4bvk$r06@winx03.informatik.uni-wuerzburg.de>, Markus von
Kienlin <kienlin@physik.uni-wuerzburg.de> wrote:
> Hi again, IDL-folks,
>
> Let me be a little bit more specific about the problem with "cd" in IDL
> that I described earlier in this group:
>
> (a) the problem is not linked to the IDL environment variables which are
> correctly defined in the script "setup.idl".
> (b) the problem occurs giving the command "cd, current=some_variable",
> but only when the current directory is temporarily mounted by the
> NIS-function "automount". Then the error-message: "% Unable to get
> working directory." comes up.
> (c) this is independent from where IDL has been started: the "cd" to the
> external directory may have been done under Unix before starting IDL, or
> any time from within IDL.
> (d) "cd, current=some_variable" works correctly if the current directory
> is permanently mounted by NFS, or for any directory on the local host.
Is the 'automount'ed directory already mounted when you try the IDL 'cd'
command? If not, what happens if you cause the 'automount'er to mount the
filesystem (by accessing something in it from UNIX) and then try the 'cd'?
I ask because we have had some problems with IDL's file I/O timing out on
files accessed from an optical jukebox when there was a 10-20 second delay
while the jukebox loaded the optical platter. The first time we would try
and read a file it would fail, but after the platter was mounted (or when
using a regular file) the same 'open' command worked without fail.
Geoff Sobering
In Vivo NMR Research Center
National Institutes of Health
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
