headers
James Peach | 1 Jun 2005 01:05
Picon
Favicon

Re: web server in Samba4

On Sun, May 29, 2005 at 01:18:24PM +1000, Andrew Tridgell wrote:
> Support for web administration in Samba4 is starting to take shape, so
> I thought it might be worthwhile to outline some design decisions here
> in case anyone has any comments.
> 
> The main aims are:
> 
>   - create a web administration interface that is good enough that
>     everyone wants to use it. It should not just be an interface for
>     inexperienced admins, it should be something that everyone
>     (including team members) finds the best way of working with the
>     server.
> 
>   - make sure it is trivial to use, while being secure.
> 
>   - provide a much richer interface than the Samba3 SWAT tool. It
>     should be more like the typical 'NAS' interfaces that vendors tend
>     to wrap around Samba, with admin and monitoring interfaces.

What's your thinking WRT monitoring interfaces? Will there be a general
performance/state monitoring infrastructure in Samba4?

--

-- 
James Peach | jpeach <at> sgi.com | SGI Australian Software Group
I don't speak for SGI.
Permalink | Reply | 8 comments |
headers
Andrew Bartlett | 1 Jun 2005 01:22
Picon
Favicon

Re: web server in Samba4

On Wed, 2005-06-01 at 09:05 +1000, James Peach wrote:
> On Sun, May 29, 2005 at 01:18:24PM +1000, Andrew Tridgell wrote:
> > Support for web administration in Samba4 is starting to take shape, so
> > I thought it might be worthwhile to outline some design decisions here
> > in case anyone has any comments.
> > 
> > The main aims are:
> > 
> >   - create a web administration interface that is good enough that
> >     everyone wants to use it. It should not just be an interface for
> >     inexperienced admins, it should be something that everyone
> >     (including team members) finds the best way of working with the
> >     server.
> > 
> >   - make sure it is trivial to use, while being secure.
> > 
> >   - provide a much richer interface than the Samba3 SWAT tool. It
> >     should be more like the typical 'NAS' interfaces that vendors tend
> >     to wrap around Samba, with admin and monitoring interfaces.
> 
> What's your thinking WRT monitoring interfaces? Will there be a general
> performance/state monitoring infrastructure in Samba4?

This is certainly one of the things I would like to see exposed by the
web interface.  Particularly if it can be in the kind of format to feed
into mrtg.

Andrew Bartlett

--

--
(Continue reading)

Permalink | Reply | 7 comments |
headers
Alan DeKok | 1 Jun 2005 01:27

Re: web server in Samba4

Andrew Bartlett <abartlet <at> samba.org> wrote:
> > What's your thinking WRT monitoring interfaces? Will there be a general
> > performance/state monitoring infrastructure in Samba4?
> 
> This is certainly one of the things I would like to see exposed by the
> web interface.  Particularly if it can be in the kind of format to feed
> into mrtg.

  Just add an SNMP server to the core.  Then people can query it
externally via another protocol, too.  This may make integration into
existing mrtg configurations easier.

  Alan DeKok.
Permalink | Reply | 2 comments |
headers
Andrew Tridgell | 1 Jun 2005 01:57

Re: web server in Samba4

Alan,

 >   Just add an SNMP server to the core.  Then people can query it
 > externally via another protocol, too.  This may make integration into
 > existing mrtg configurations easier.

If someone wants to contribute a snmp server for smbd then we could
look at it, but its not something I would consider critical for the
first release of Samba4. 

I think that snmp is in the category of 'would be nice, but we can
function OK without it', which means I'd rather see someone add
support for linking with an existing snmp implementation rather than
writing one from scratch.

One possible approach would be to optionally link in a snmp library
and expose it to ejs. We are going to need to expose our monitoring
functions to ejs anyway for the web pages, so if snmp functionality
could also be exposed to ejs it would provide a nice environment for
extending what is made visible in snmp.

Anyway, I'm not going to work on this, so if someone wants to
volunteer to do some snmp work then please speak up!

Cheers, Tridge
Permalink | Reply | 1 comment |
headers
James Peach | 1 Jun 2005 02:06
Picon
Favicon

Re: web server in Samba4

On Wed, Jun 01, 2005 at 09:22:55AM +1000, Andrew Bartlett wrote:
> On Wed, 2005-06-01 at 09:05 +1000, James Peach wrote:
> > On Sun, May 29, 2005 at 01:18:24PM +1000, Andrew Tridgell wrote:
> > > Support for web administration in Samba4 is starting to take shape, so
> > > I thought it might be worthwhile to outline some design decisions here
> > > in case anyone has any comments.
> > > 
> > > The main aims are:
> > > 
> > >   - create a web administration interface that is good enough that
> > >     everyone wants to use it. It should not just be an interface for
> > >     inexperienced admins, it should be something that everyone
> > >     (including team members) finds the best way of working with the
> > >     server.
> > > 
> > >   - make sure it is trivial to use, while being secure.
> > > 
> > >   - provide a much richer interface than the Samba3 SWAT tool. It
> > >     should be more like the typical 'NAS' interfaces that vendors tend
> > >     to wrap around Samba, with admin and monitoring interfaces.
> > 
> > What's your thinking WRT monitoring interfaces? Will there be a general
> > performance/state monitoring infrastructure in Samba4?
> 
> This is certainly one of the things I would like to see exposed by the
> web interface.  Particularly if it can be in the kind of format to feed
> into mrtg.

I suggest that this information be exposed in a more general way so that
it is useable by more consumers than just the web interface.
(Continue reading)

Permalink | Reply | 3 comments |
headers
Alan DeKok | 1 Jun 2005 02:08

Re: web server in Samba4

Andrew Tridgell <tridge <at> osdl.org> wrote:
> One possible approach would be to optionally link in a snmp library
> and expose it to ejs. We are going to need to expose our monitoring
> functions to ejs anyway for the web pages, so if snmp functionality
> could also be exposed to ejs it would provide a nice environment for
> extending what is made visible in snmp.

  I would recommend SMUX over an integrated SNMP server. :)

  I've done some work on it myself in another project, but I'm not
familiar enough with Samba yet to see how I'd add it in here.

  Alan DeKok.
Permalink | Reply | 0 comments |
headers
Jeremy Allison | 1 Jun 2005 02:18
Picon
Favicon

Re: Timeouts in HEAD ldap code.

On Sat, May 28, 2005 at 10:04:22PM +0200, Volker Lendecke wrote:
> 
> Are you referring to lib/smb_ldap.c? The LDAP parsing/unparsing stuff? As far
> as I know this is not yet really used anywhere. I've programmed
> idmap_smbldap.c that uses these functions, but that's about it I think. In
> Samba 3 this has not gone anywhere yet, maybe it will sometime, but not yet.
> 
> This LDAP lib does not use alarm, and I'd be happy to see read_data_until and
> write_data_until corrected.
> 
> I'm much more concerned with smbldap.c, the stuff that uses the alarm()
> function. The whole reason for the homegrown ldap libraries was the inability
> to get the OpenLDAP libs to *reliably* timeout after a certain amount of
> time. I don't think they are ready to run over async sockets, or am I wrong
> here? In my test of quite a while ago, the OpenLDAP libs did not check for
> EINTR and happily returned if the read/write calls were interrupted. This
> might change however.

The problem is the changes you made to HEAD to cause read_data() to
depend on read_data_until() broke the callers of read_data() until
I made read_data_until() use the signal-safe sys_select_eintr() and sys_read()
calls.

Now read_data_until() isn't quite right as it calls sys_read() which will
block indefinately as it ignores EINTR.

We need to separate out the read paths for the smb calls, which need
to be signal safe (using the EINTR ignoring wrapper functions) and
to create parallel read paths using read_data_until() which allow
signal interrupts to return EINTR. Then from functions that depend
(Continue reading)

Permalink | Reply | 8 comments |
headers
Ingo Steuwer | 1 Jun 2005 08:27
Picon
Favicon

Re: net rpc vampire performance

Hi

we've looked at this and decided to implement and contribute a "general 
export", preferred is xml. We don't want to conflict with the announced 
enhancements for the ldif-export and would use a similar command line ("net 
rpc vampire xml").

So the questions are:

- is the command line ok? (Don already asked it, but I can't find an answer)
- can I find a patch for the ldif-export somewhere or is it in SVN?

Greetings
Ingo Steuwer

--

-- 
Ingo Steuwer       steuwer <at> univention.de         fon: +49 421 22 232- 0
Entwicklung        Linux for Your Business
Univention GmbH    http://www.univention.de/     fax: +49 421 22 232-99
Permalink | Reply | 1 comment |
headers
Volker Lendecke | 1 Jun 2005 09:38
Picon
Favicon

Re: svn commit: samba r7155 - in trunk/source/nsswitch: .

On Tue, May 31, 2005 at 10:02:46PM +0000, jerry <at> samba.org wrote:
> a little bit of paranoia until we gain more confidence in the fd_events list

Thanks. I already had problems with that already.

Volker
Permalink | Reply | 1 comment |
headers
Volker Lendecke | 1 Jun 2005 11:07
Picon
Favicon

Re: Timeouts in HEAD ldap code.

On Tue, May 31, 2005 at 05:18:59PM -0700, Jeremy Allison wrote:
> Now read_data_until() isn't quite right as it calls sys_read() which will
> block indefinately as it ignores EINTR.

You're sure? sys_read is only called if select indicated that something is
readable. Reading http://www.opengroup.org/onlinepubs/007908799/xsh/read.html
I get the impression that EINTR is only returned when there's no data
available. In all other cases it should return the amount of data actually
read:

[EINTR]
    The read operation was terminated due to the receipt of a signal, and no
    data was transferred.

Is read(2) still an interruptible system call when select has indicated
readability?

Volker
Permalink | Reply | 2 comments |

Gmane