[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Phpgroupware-developers] Proposed ldap patch...
From: |
Lars Kneschke(priv.) |
Subject: |
Re: [Phpgroupware-developers] Proposed ldap patch... |
Date: |
Thu, 27 Mar 2003 06:05:00 +0000 |
Izzy Blacklock <address@hidden> schrieb:
>Well I just did some digging and it looks like uidnumber and gidnumber
>are
>used in a number of places. Here's the results from the phpgw root
>dir:
>
>grep -R gidnumber * | wc -l
> 32
>grep -R uidnumber * | wc -l
> 59
>
>and from the phpgwapi dir
>
>grep -R gidnumber * | wc -l
> 23
>grep -R uidnumber * | wc -l
> 45
>
>Obviousely there would be a lot of code to change, some of which
>probably
>couldn't be changed to correct this issue the way I've proposed. This
>leaves
>two options: eliminate the option not to manage homedir and shell
>entries, or
>provide a sane default if the user chooses not to provide one.
>
>I can think of legitimate reasons why someone wouldn't want phpgw
>managing
>system accounts, so eliminating the option probably isn't a good idea.
> In
>fact, it may be worth changing the wording to read:
>
> If using LDAP, do you want to manage system accounts?
>
>Saying no should then remove the ability to add, remove, or otherwise
>modify
>system account elements. They should be limited to adding, removing,
>or
>otherwise modifying phpgw elements from existing accounts.
>
>I realize this is a lot more work then a simple patch, but I think it
>make the
>most sense. What does everyone else think?
>
>I'll document the behavior as it is now. As long as people don't try
>adding
>the demo accounts and I'd guess add users with phpgw when they've said
>no to
>managing the homedir and shell attributes, then there shouldn't be any
>
>problems.
As we have learned the last days, it is bad to have a option not to set
homedir and shell. This should be rewriten, so that this fields become
always set.
Making ldap working only with existing accounts is also a very nice idea.
I'll try to implement this.
Cu