You are viewing a plain text version of this content. The canonical link for it is here.
Posted to general@gump.apache.org by Leo Simons <ma...@leosimons.com> on 2010/09/20 11:58:46 UTC
Re: vmgump2
Hey guys,
On 8/9/10 5:17 AM, Stefan Bodewig wrote:
> On 2010-08-08, Stefan Bodewig wrote:
>> vmgump2 sends nag emails to me right now, but if the current run works
>> like I expect it to, I'll enable nagging there and turn off nagging on
>> vmgump.
>
> Done now. Starting midnight GMT vmgump2 will take over as the
> "official" machine.
Just checking in after ignoring my e-mail for some 2 months (*). Sorry I
missed all the fun! Awesome work, though!
Are you guys all up and running again now or is there anything more that
needs attention? For example did you figure out the wonders of mysql to
the degree you wanted to? Let me know.
cheers,
Leo
(*) I can totally recommend it ;)
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@gump.apache.org
For additional commands, e-mail: general-help@gump.apache.org
Re: vmgump2
Posted by Stefan Bodewig <bo...@apache.org>.
On 2010-09-23, Leo Simons wrote:
> I had some fun trying to dig through the codebase to try and remember
> what's what...hoo boy, this certainly isn't how I'd write python code
> today :)
The Gump code base is a collection of "first Python project of a Java
programmer" pieces from three or four people.
> Nevertheless I think it may be safe to disable
> # Add Historical Database storer -- ??? no such thing...
> if self.run.getOptions().isOfficial() and \
> self.run.getWorkspace().hasDatabaseInformation() and \
> self.run.getOptions().isHistorical():
> try:
> import gump.actor.history.historical
> self.run.registerActor(
> gump.actor.history.historical.Historical(self.run))
> except Exception, details:
> self.log.warning(
> 'Unable to register Historical Database Actor : %s '
> % details,
> exc_info=1)
> # Add Dynagump database populator
> if self.run.getWorkspace().hasDatabaseInformation():
> # create the database helper
> dbInfo = self.run.getWorkspace().getDatabaseInformation()
> from gump.util.mysql import Database
> database = Database(dbInfo)
> # now create the Dynagumper using that database
> import gump.actor.mysql.dynagumper
> self.run.registerActor(
> gump.actor.mysql.dynagumper.Dynagumper(self.run,database))
> in
> https://svn.apache.org/repos/asf/gump/trunk/python/gump/core/runner/runner.py
Agreed.
> that may help get rid of some confusing warnings?
We used to get funny error messages when isHistorical returned true on
the FreeBSD machine since there have been two isHistorical methods in
the same class and it depended on the platform which of the two methods
was called[1]. This is why I like compilers to catch my errors and have
made a habit out of running pylint on python code before I check it in.
Dynagumper is a an actor that doesn't do anything so it doesn't hurt too
much either.
Stefan
[1] http://svn.apache.org/viewvc/gump/trunk/python/gump/core/run/options.py?r1=988801&r2=988804
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@gump.apache.org
For additional commands, e-mail: general-help@gump.apache.org
Re: vmgump2
Posted by Leo Simons <ma...@leosimons.com>.
On 9/20/10 12:13 PM, Stefan Bodewig wrote:
> On 2010-09-20, Leo Simons wrote:
>> For example did you figure out the wonders of mysql to the degree you
>> wanted to?
>
> More than I wanted to, I'd say.
:)
> If you happen to recall the specifics
> around the "historical databaser"
IIRC the code to make use of the historical database was to go into
dynagump but was never actually written, so I don't see any reason to
run that actor.
> or the dynagump actors
IIRC we disabled dynagump (the java webapp part of it) the previous time
we rebuilt vmgump from scratch -- we weren't actually using it for
anything and no-one was developing on it, and so turning it off made
things (a) just slightly easier, (b) slightly easier to not mess up
security-wise.
> it would be nice if you could referesh
> my memory - but both work as well as they did
> before (i.e. not at all).
Yeah so I think that's "correct" behavior!
I had some fun trying to dig through the codebase to try and remember
what's what...hoo boy, this certainly isn't how I'd write python code
today :)
Nevertheless I think it may be safe to disable
# Add Historical Database storer -- ??? no such thing...
if self.run.getOptions().isOfficial() and \
self.run.getWorkspace().hasDatabaseInformation() and \
self.run.getOptions().isHistorical():
try:
import gump.actor.history.historical
self.run.registerActor(
gump.actor.history.historical.Historical(self.run))
except Exception, details:
self.log.warning(
'Unable to register Historical Database Actor : %s '
% details,
exc_info=1)
# Add Dynagump database populator
if self.run.getWorkspace().hasDatabaseInformation():
# create the database helper
dbInfo = self.run.getWorkspace().getDatabaseInformation()
from gump.util.mysql import Database
database = Database(dbInfo)
# now create the Dynagumper using that database
import gump.actor.mysql.dynagumper
self.run.registerActor(
gump.actor.mysql.dynagumper.Dynagumper(self.run,database))
in
https://svn.apache.org/repos/asf/gump/trunk/python/gump/core/runner/runner.py
that may help get rid of some confusing warnings?
cheers,
Leo
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@gump.apache.org
For additional commands, e-mail: general-help@gump.apache.org
Re: vmgump2
Posted by Stefan Bodewig <bo...@apache.org>.
On 2010-09-20, Leo Simons wrote:
> Just checking in after ignoring my e-mail for some 2 months (*).
Sounds great.
> Are you guys all up and running again now or is there anything more
> that needs attention?
Gump is happily running on the new VM as well as on a new FreeBSD zone
by now.
> For example did you figure out the wonders of mysql to the degree you
> wanted to?
More than I wanted to, I'd say. If you happen to recall the specifics
around the "historical databaser" or the dynagump actors it would be
nice if you could referesh my memory - but both work as well as they did
before (i.e. not at all).
Stefan
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@gump.apache.org
For additional commands, e-mail: general-help@gump.apache.org