I was directed to this mailing list as a good source of knowledge about accessing files on Netapp. I hope this question is appropriate; if not, I apologize.
I am working in an environment in which a Netapp filer containing files with mixed-case names is being accessed using both CIFS and NFS. It is important that I be able to access the files using NFS in a case-insensitive manner. Unfortunately, converting all of the filenames to a consistent case is not an option. The NFS access is being done from a RedHat Linux server running Apache. I have considered writing an Apache module to parse the directory structure to determine the actual case of the filename, although this option seems too inefficient.
Any suggestions would be greatly appreciated. Thank you very much for your time,
eric seidman eseidman@cs.stanford.edu
I am working in an environment in which a Netapp filer containing files with mixed-case names is being accessed using both CIFS and NFS. It is important that I be able to access the files using NFS in a case-insensitive manner.
You can't do it. UNIX and NFS support both upper and lower case filenames.
Unfortunately, converting all of the filenames to a consistent case is not an option.
Then you are out of luck.
The NFS access is being done from a RedHat Linux server running Apache. I have considered writing an Apache module to parse the directory structure to determine the actual case of the filename, although this option seems too inefficient.
Altering Apache is the best option, but you have to decide what to do in the case of conflicts. It would seem that you already have an idea of which case of filenames is the "correct" one... in which case, you should have no trouble having the filenames saved to a consistent case.
Perhaps if I (or others) understood why you can't convert filenames to a consistent case it would be easier to resolve the problem.
Bruce
Bruce Sterling Woodcock wrote:
Perhaps if I (or others) understood why you can't convert filenames to a consistent case it would be easier to resolve the problem.
Bruce
I'm wondering why the need for case conversion at all.
It sounds like the filenames are part of the web servers document tree, and URL's are case sensitive.
Have you considered getting an http license for the NetApp, and having it source the files directly? The Apache server would handle the dynamic pages, while the NetApp handles the static pages.
-- Matthew Lee Stier * Fujitsu Network Communications Unix Systems Administrator | Two Blue Hill Plaza Ph: 914-731-2097 Fx: 914-731-2011 | Sixth Floor Matthew.Stier@fnc.fujitsu.com * Pearl River, NY 10965
+-- "Bruce Sterling Woodcock" sirbruce@ix.netcom.com once said: | > The NFS access is being done from | > a RedHat Linux server running Apache. I have considered writing | > an Apache | > module to parse the directory structure to determine the actual case of | > the filename, although this option seems too inefficient. | | Altering Apache is the best option, but you have to decide what | to do in the case of conflicts. It would seem that you already have | an idea of which case of filenames is the "correct" one... in which | case, you should have no trouble having the filenames saved to a | consistent case.
Look at mod_spelling for apache. I think it will do what you want (i.e. handle when people mis-case a URL).
Oz
actually,
it's "mod_speling"
=)
-corris
On Wed, 30 Aug 2000, Ozzie Sabina wrote:
Date: Wed, 30 Aug 2000 19:02:33 -0400 From: Ozzie Sabina ors@cimedia.com To: Bruce Sterling Woodcock sirbruce@ix.netcom.com, Eric Seidman eseidman@CS.Stanford.EDU, toasters@mathworks.com Subject: Re: accessing mixed-case files on Netapp
+-- "Bruce Sterling Woodcock" sirbruce@ix.netcom.com once said: | > The NFS access is being done from | > a RedHat Linux server running Apache. I have considered writing | > an Apache | > module to parse the directory structure to determine the actual case of | > the filename, although this option seems too inefficient. | | Altering Apache is the best option, but you have to decide what | to do in the case of conflicts. It would seem that you already have | an idea of which case of filenames is the "correct" one... in which | case, you should have no trouble having the filenames saved to a | consistent case.
Look at mod_spelling for apache. I think it will do what you want (i.e. handle when people mis-case a URL).
Oz
On Wed, 30 Aug 2000, Eric Seidman wrote:
The NFS access is being done from a RedHat Linux server running Apache. I have considered writing an Apache module to parse the directory structure to determine the actual case of the filename, although this option seems too inefficient.
Try mod_speling (note the spelling ;-)). We use that here when Micro$oft HTML publishing programs refer to StudlyCapped.HTM URL's when the filename is actually "studlycapped.htm" on disk. Personally, I think this is silly... URL's and filenames on NFS are case-sensitive. People and software should deal with that.