headers
Gustavo Niemeyer | 6 Nov 2007 23:43
Favicon
Gravatar

Re: Database disconnection detection


Hey James!

> I've just merged support for handling database disconnections in Storm.

Thanks a lot for the feature and the nice implementation!

> There are a few caveats though:
>  1. the disconnection tests try to connect to the test database via
> TCP.  You will probably need to edit pg_hba.conf to allow this.

As we discussed, I've changed that behavior a little bit on trunk.

There are two different URIs which may be set for tests using environment
variables.  Specifically:

    STORM_POSTGRES_HOST_URI - Should be set to a URI with a hostname, such
    as "postgres://localhost/storm". This will be used for the
    disconnection test specifically.

    STORM_POSTGRES_URI - Should be set to an URI for a postgres database,
    with or without the host part.  Notice that tests run faster if this is
    set to something like "postgres:storm".

Both are optional.  The specific tests are disabled when they're not
present.  Notice that these URIs may include username/passwords as wanted.

--

-- 
Gustavo Niemeyer
http://niemeyer.net
(Continue reading)

Permalink | Reply | 0 comments |
headers
Gustavo Niemeyer | 6 Nov 2007 23:48
Favicon
Gravatar

Re: Database disconnection detection


Hey James!

> I've just merged support for handling database disconnections in Storm.

Thanks a lot for the feature and the nice implementation!

> There are a few caveats though:
>  1. the disconnection tests try to connect to the test database via
> TCP.  You will probably need to edit pg_hba.conf to allow this.

As we discussed, I've changed that behavior a little bit on trunk.

There are two different URIs which may be set for tests using environment
variables.  Specifically:

    STORM_POSTGRES_HOST_URI - Should be set to a URI with a hostname, such
    as "postgres://localhost/storm". This will be used for the
    disconnection test specifically.

    STORM_POSTGRES_URI - Should be set to an URI for a postgres database,
    with or without the host part.  Notice that tests run faster if this is
    set to something like "postgres:storm".

Both are optional.  The specific tests are disabled when they're not
present.  Notice that these URIs may include username/passwords as wanted.

--

-- 
Gustavo Niemeyer
http://niemeyer.net
(Continue reading)

Permalink | Reply | 0 comments |
headers
Brad Allen | 7 Nov 2007 04:06

Re: Sprint invitation

>Hello Stormers,
>
>In the next week we'll all be sprinting in Landscape in the
>same location where the Ubuntu Developer Summit (UDS) will
>be held (Hilton  <at>  MIT, Cambridge, MA, US).
>
>On Friday, November 2nd, some of us will have a sprint
>specifically on Storm.  We expect to get some of the on going
>work integrated and, depending on time, kick start some of the
>features we want to see supported in Storm in the near future.
>
>We invite anyone who'd like to participate to be there for the
>sprint, and you're also welcome to participate online in the
>IRC channel.

I am curious how this worked out...did anyone blog about it?

I see a lot of branches published on LaunchPad, but it's not
clear which ones are likely to make it into the trunk. What are
the planned future features?

Thanks.
Permalink | Reply | 1 comment |
headers
Andrew Holman | 7 Nov 2007 15:43

Pooling ?

I am working with CherryPy and am trying to figure out the best way  
to incorporate Storm into it. Does Storm have any connection pooling?  
If not then what is the best way to handle ensuring the connections  
are still alive with each thread?  I would assume we would have to  
write something that would do a check/auto-connect if the threads  
connection was no longer alive. How should that be handled?

- Andy
Permalink | Reply | 2 comments |
headers
Gustavo Niemeyer | 7 Nov 2007 20:56
Favicon
Gravatar

Re: Sprint invitation

Hey Brad,

> I am curious how this worked out...did anyone blog about it?

Unfortunately not.  We have had some important issues to look
over last week and that took over some of the planned time of
the sprint.

Even then, we did manage to get some things done, and we also
got some of the *unplanned* time for working on Storm.

Specifically, some exciting things were worked on:

- James Henstridge has landed the branch with support for
  auto-reconnection of stores.  We still need support for MySQL,
  but that should be easily moved on in the future.

- There were discussions on the on going work to move Launchpad
  to Storm.  Things are going pretty well in that area so far.

- There's some tentative work being done in a well known open
  source software to migrate it to Storm.  I'll keep the actual
  announcement of that with the maintainers.

> I see a lot of branches published on LaunchPad, but it's not
> clear which ones are likely to make it into the trunk. What are
> the planned future features?

Some major functionalities that may be expected for the near
future of Storm:
(Continue reading)

Permalink | Reply | 0 comments |
headers
Gustavo Niemeyer | 8 Nov 2007 05:02
Favicon
Gravatar

Storm on Maemo platform

Hello Stormers!

A friend just wrote me to let me know of some very cool news. Storm
will be part of the Nokia-pushed Maemo platform:

http://importluck.blogspot.com/2007/11/modules-modules.html

--

-- 
Gustavo Niemeyer
http://niemeyer.net
Permalink | Reply | 0 comments |
headers
Gustavo Niemeyer | 13 Nov 2007 19:08
Favicon
Gravatar

Re: Pooling ?

Hey Andy,

> I am working with CherryPy and am trying to figure out the best way

That's very cool!  I'm looking forward to see it being seamlessly
integrated in CherryPy.

> to incorporate Storm into it. Does Storm have any connection pooling?  
> If not then what is the best way to handle ensuring the connections  
> are still alive with each thread?  I would assume we would have to  
> write something that would do a check/auto-connect if the threads  
> connection was no longer alive. How should that be handled?

We actually implement a slightly different pooling mechanism, which
was recently improved by the landing of a patch from James Henstridge.

I'll explain how we do it in Zope3, and how the "pooling" system works,
so that you can imagine how to integrate it in CherryPy.

Here is how it works, from a high-level perspective: we have a ZStorm
utility which provides per-thread Store instances based on a name=>uri
mapping (code asks for zstorm.get("main-store"), for instance).  This
utility is actually very simple, except for the logic to hook it in
Zope's transaction mechanism, which you don't need.

So how would pooling work in that environment?  Well, we actually
don't need a "traditional" pooling system, because each thread has its
own Store, which is kept alive across transactions and requests.

The last step missing in that environment is how to handle disconnections,
(Continue reading)

Permalink | Reply | 0 comments |
headers
Vsevolod Balashov | 14 Nov 2007 00:18

Re: Pooling ?

Hi!

> I am working with CherryPy and am trying to figure out the best way  
> to incorporate Storm into it.

Just try middlestorm http://pypi.python.org/pypi/middlestorm/0.4
Permalink | Reply | 0 comments |
headers
Brad Allen | 15 Nov 2007 04:19

Storm talk or tutorial at PyCon?

Are any Stormers planning to give a PyCon talk? The proposal deadline 
is midnight Fri.

It would be great to have Gustavo or one of the other contributors 
give a tutorial, or at least a talk.

>Date: Fri, 09 Nov 2007 10:52:38 -0600
>From: Greg Lindstrom <gslindstrom@...>
>Subject: PyCon 2008 - Call for Tutorials
>Sender: python-announce-list-bounces@...
>To: python-list@..., pycon-organizers@...,
>  python-announce-list@...
>Reply-to: python-list@...
>X-BeenThere: python-announce-list@...
>Delivered-to: python-announce-list@...
>X-Original-To: python-announce-list@...
>List-Post: <mailto:python-announce-list@...>
>List-Subscribe: 
><http://mail.python.org/mailman/listinfo/python-announce-list>,
>	<mailto:python-announce-list-request@...?subject=subscribe>
>List-Unsubscribe:
>  <http://mail.python.org/mailman/listinfo/python-announce-list>,
>	<mailto:python-announce-list-request@...?subject=unsubscribe>
>List-Archive: <http://mail.python.org/pipermail/python-announce-list>
>List-Help: <mailto:python-announce-list-request@...?subject=help>
>List-Id: Announcement-only list for the Python programming language
>	<python-announce-list.python.org>
>Original-recipient: rfc822;bradallen@...
>
>Still thinking of presenting a tutorial at PyCon 2008 in Chicago?
(Continue reading)

Permalink | Reply | 501 comments |
headers
Gustavo Niemeyer | 17 Nov 2007 14:01
Favicon
Gravatar

Re: Storm talk or tutorial at PyCon?

Hey Brad,

> Are any Stormers planning to give a PyCon talk? The proposal deadline 
> is midnight Fri.
> 
> It would be great to have Gustavo or one of the other contributors 
> give a tutorial, or at least a talk.

I'd really being in PyCon again, but this time I will have to skip
it unfortunately. :-(

Thanks for reminding us about it,

--

-- 
Gustavo Niemeyer
http://niemeyer.net
Permalink | Reply | 501 comments |

Gmane