Re: Web based Retrieval Mechanism???

Bizarre, I never received anything back from mailman, yet my message was
online. Have cut-and-pasted to rejoin :)
 
> Hrrm, I can see that would be quite useful for some people.  You'd need
> to be VERY careful about security though.

Yeah, that is one area of concern (along with the actual programming, am a
sys admin, not a programming geek :)
 
> This would be a very cool wheel to contribute!

Do I sense I've dug myself a hole here...  :)
 
> I'm guessing from your mention of LDAP that you have consistent userids
> and groups across all your machines, so at least for a first stab at it
> you could just assume that they are indeed consistent.  But I think you
> should leave hooks in place so that someone can put some different
> mechanism in place.

Correct, I have a LDAP backend across all of my servers and hence consistent
uid/guids.
 
> For example, in my case I do *not* have consistent userids and groups
> between the backup machine and the clients - the backup machine only
> does backups, and I don't allow users to log in.  So I have users on
> some machines who simply don't exist on the backup machine.

My backups get punched to a couple of different boxes. Server config files
and other more sensitive items are sent to a different box in a secure
location (physical as well as file-level access). Home directories, public
stuff and group areas are fed to an OpenFiler based NAS/SAN box. It has
built-in systems to enable these shares to be shared out via
NFS/SMB/WebDAV/etc, however I wanted a more simplistic interface if
possible. Have used MS' VSS on XP at at previous job and it works pretty
well. Just right-click the parent of your deleted file/folder, find the copy
you want and bring it back. Veritas now do a web based retrieval system for
Backup Exec 10d, that can grab files back that are stored on near-line
(before sending to tape) backup storage. I tested the beta and it worked
well.

> If I were
> to use this tool of yours, then I'd want to somehow shim in some code so
> that I can have it parse the .../daily.0/hostname/etc/passwd file to get
> the relevant userids and do password matching.
 
Agreed, security is the primary concern. But I guess that's the same as any
normal webapp that accesses the filesystem. As you mentioned, some kind of
hook to appropriate auth mechanisms would allow ppl to use/build whatever
auth they required. In your case some parsing code.

Have been checking to see if any generic web based file managers would be a
good place to start, but as yet they all seem to be a bit lacking in certain
areas. A couple of nice AJAX based ones are starting to pop up though
(http://ecosmear.com/relay).

Will investigate and start contributing to that wheel :)

Thanks

Dan
--

Dan Hawker
Linux System Administrator
EADS Astrium

-- 

This email is for the intended addressee only.
If you have received it in error then you must not use, retain, disseminate or 
otherwise deal with it.
Please notify the sender by return email.
The views of the author may not necessarily constitute the views of Astrium 
Limited.
Nothing in this email shall bind Astrium Limited in any contract or obligation.

Astrium Limited, Registered in England and Wales No. 2449259
Registered Office: Gunnels Wood Road, Stevenage, Hertfordshire, SG1 2AS, England

-------------------------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642