Thread: Fate of pgsnmpd

Fate of pgsnmpd

From
"Florian G. Pflug"
Date:
Hi

Does anyone know if pgsnmpd is still actively developed?
The last version (0.1b1) is about 15 months old.

greetings, Florian Pflug


Re: Fate of pgsnmpd

From
Stefan Kaltenbrunner
Date:
Florian G. Pflug wrote:
> Hi
> 
> Does anyone know if pgsnmpd is still actively developed?
> The last version (0.1b1) is about 15 months old.

there seems to be quite a lot of work going on in the cvs tree:

http://cvs.pgfoundry.org/cgi-bin/cvsweb.cgi/pgsnmpd/pgsnmpd/

so i would guess it is still active though without an release for a while.



Stefan


Re: Fate of pgsnmpd

From
Magnus Hagander
Date:
Florian G. Pflug wrote:
> Hi
> 
> Does anyone know if pgsnmpd is still actively developed?
> The last version (0.1b1) is about 15 months old.

It is.
There is a team (Josh Tolley, me and Hiroshi Saito) working for RFC 1697
compliance. When that's done, there are some other additions in the
pipeline.

There's been a rewrite of several things since 0.1b1, and what we have
now is not ready for production use. But it will be :-)

//Magnus


Re: Fate of pgsnmpd

From
"Dave Page"
Date:

> ------- Original Message -------
> From: "Florian G. Pflug" <fgp@phlo.org>
> To: "pgsql-hackers@postgresql.org" <pgsql-hackers@postgresql.org>
> Sent: 06/04/07, 20:12:39
> Subject: [HACKERS] Fate of pgsnmpd
> 
> Hi
> 
> Does anyone know if pgsnmpd is still actively developed?
> The last version (0.1b1) is about 15 months old.

Yes, it is. There have been a number of commits recently and I believe Josh(?) is giving  a talk about it at pgCon.

Regards, Dave


Re: Fate of pgsnmpd

From
"Josh Tolley"
Date:
On 4/6/07, Dave Page <dpage@postgresql.org> wrote:
> Yes, it is. There have been a number of commits recently and I believe Josh(?) is giving  a talk about it at pgCon.
>
> Regards, Dave

Josh just found his passport, which will make giving that talk a lot
easier ;) As Magnus said, we're aiming at RFC 1697 compliance first.
Since the RFC's MIB is designed to apply to *any* database, it doesn't
cover lots of the specific statistics a pgsql person would likely want
to see, so after the RFC work is done we'll be adding a pgsql-specific
MIB..

- Josh Tolley


Re: Fate of pgsnmpd

From
Michael Fuhr
Date:
On Sat, Apr 07, 2007 at 06:01:10AM -0600, Josh Tolley wrote:
> Josh just found his passport, which will make giving that talk a lot
> easier ;) As Magnus said, we're aiming at RFC 1697 compliance first.
> Since the RFC's MIB is designed to apply to *any* database, it doesn't
> cover lots of the specific statistics a pgsql person would likely want
> to see, so after the RFC work is done we'll be adding a pgsql-specific
> MIB..

I might be interested in contributing to this effort, if not with
code then at least with discussion and testing.  I'm currently using
MRTG to execute Perl scripts that query the statistics views and
I've been thinking about rewriting those scripts to be AgentX
subagents so they'd be queryable via SNMP.

-- 
Michael Fuhr


Re: Fate of pgsnmpd

From
Magnus Hagander
Date:
Michael Fuhr wrote:
> On Sat, Apr 07, 2007 at 06:01:10AM -0600, Josh Tolley wrote:
>> Josh just found his passport, which will make giving that talk a lot
>> easier ;) As Magnus said, we're aiming at RFC 1697 compliance first.
>> Since the RFC's MIB is designed to apply to *any* database, it doesn't
>> cover lots of the specific statistics a pgsql person would likely want
>> to see, so after the RFC work is done we'll be adding a pgsql-specific
>> MIB..
> 
> I might be interested in contributing to this effort, if not with
> code then at least with discussion and testing.  I'm currently using
> MRTG to execute Perl scripts that query the statistics views and
> I've been thinking about rewriting those scripts to be AgentX
> subagents so they'd be queryable via SNMP.
> 

We're definitely interested in listening :-) Please join us over at the
pgsnmpd list.

//Magnus


Re: Fate of pgsnmpd

From
Josh Berkus
Date:
Josh,

> Josh just found his passport, which will make giving that talk a lot
> easier ;) As Magnus said, we're aiming at RFC 1697 compliance first.
> Since the RFC's MIB is designed to apply to *any* database, it doesn't
> cover lots of the specific statistics a pgsql person would likely want
> to see, so after the RFC work is done we'll be adding a pgsql-specific
> MIB..
FYI, the MySQL folks want to talk to you about maybe lobbying to change the 
RFC.  They feel that an awful lot of RFC1697 is Oracle-specific, and are 
wondering if we can do anything about it.

Anyway, if you want to ping them, Brian Aker <brian@mysql.com>.

-- 
--Josh

Josh Berkus
PostgreSQL @ Sun
San Francisco


Re: Fate of pgsnmpd

From
"Josh Tolley"
Date:
On 4/7/07, Josh Berkus <josh@agliodbs.com> wrote:
> FYI, the MySQL folks want to talk to you about maybe lobbying to change the
> RFC.  They feel that an awful lot of RFC1697 is Oracle-specific, and are
> wondering if we can do anything about it.

Indeed... I've had brief discussions with a Mark Atwood, IIRC, who's
working on the MySQL implementation. I'll drop them a line, and cc
pgsnmpd-devel. Thanks.

- Josh Tolley