[Samba] Samba v3.0.23a BROKE my network

[Chris Hall]

On Wed, 24 Jan 2007 you wrote
>Not that I don't understand your frustration. I do.

Well, yes I'm venting a little.

But more importantly, as a software developer, I think there are repeat 
lessons here on how to write good software.  And my objective is/was to 
offer good user perspective feed back.

> I guess it boils down to whether I would prefer the developers 
>concentrate on fixing bugs and improving functionality or for them to 
>expend energy on helping prevent me from doing something bad that goes 
>against best practices. I overwhelmingly prefer the former.

I wonder how many other people were tripped up by this.

It may well be that mapping certain (or even all) groups was best 
practice.  But prior to v3.0.23 things seemed to work OK without.

The software was changed so that it no longer worked unless certain 
groups are mapped.  A small piece of configuration changed from being 
optional to being essential.

Volker Lendecke said:

   We have it made very explicit that there are big changes
   coming with this release.

There was a long piece in the release notes around the issue.  (Which 
didn't directly say "you need these groups to be mapped", but that's 
another story.)

The change in the importance of group mapping is so significant that a 
whole chapter in the HOWTO is given over to repeating the that piece of 
release note.

The HOWTO does say that it is (now) essential to map these groups.

So, this is not a small change.

So, principles of defensive programming call for the software to detect 
the (now) incomplete configuration and (at the very least) produce an 
error message.

[testparm doesn't spot the configuration problem, either.  It does 
whinge about 'printer admin' being deprecated, which seems a relatively 
minor issue !  I wonder if the required group mappings should be part of 
smb.conf...]

>If someone out there has the wherewithal to write some kind of upgrade 
>checker that has the functionality you desire, I hope they come forward 
>and do so. Otherwise, I'm content for the developers to keep doing what 
>they are doing and focus on getting Samba 4 ready for prime time. But 
>that's just me and I can't speak for anyone else.

OK.  If Samba 4 does what v3.0.23 just did -- fail to complain about 
missing bits of essential configuration, and go on to fail to work 
without giving much of a clue as to why -- then it will never be ready 
for prime time.

IMO this is a repeat of the lesson that a few lines of code are worth 
hundreds of lines of documentation.  And the corollary: if you need to 
write lots of lines of documentation to warn or to explain, then maybe 
you should go back and fix the software instead.

I'll go away, now.

Chris

>Chris Hall wrote:
>> On Wed, 24 Jan 2007 you wrote
>>> No offense,
>>
>> None taken.
>>
>>> but making any software update or change to a production system 
>>>without first testing it in a test environment is an administrative 
>>>issue, not a software issue.
>>
>> That could have reduced my users' blood pressure...
>>
>> ...but doesn't change my opinion that software should be written to 
>>avoid obscure failure caused by obvious misconfiguration -- 
>>particularly in the case of an upgrade which turns a previously 
>>working configuration into a broken one !
>>
>>> It isn't terribly difficult to configure a test environment that 
>>>would allow you to see if everything works as expected.
>>
>> Sure.  But I'm not trying to run a nuclear power station here.
>>
>> Chris
....snip....
-- 
Chris Hall   @ Home                                         +44 (0)7970 277 383