On Jan 11, 2009, at 10:56 PM, Hans Bakker wrote:
please do not lecture me? I am aware of the rules.
Sorry you interpreted it as a lecture.
you write in the previous message:
Generally we want things like scheduled jobs to be loaded once and
not every time a software update is done, meaning it should be in
seed-
initial and not in seed.
if you would have told me why at that point i would accepted that
straight away....
When I first responded I had no idea what you intended, so I wrote to
find that out... and found out that it seemed you hadn't really looked
into it, hence my frustration.
but then you write:
I see where you're going with this now, and I don't like it.
i have to admit this is a good argument, no idea what you mean.... :-)
What you said could be true a long time ago but....sorry I have ran
the
loading of that scheduled job seed data several times and it did not
create multiple jobs....probably because there is a check in the job
it
cannot run jobs multiple times....?
Did the job run into between the loadings? If not it wouldn't
duplicate the job. The problem is that each time a job is run and
rescheduled a new job record is created with a new ID, so if you re-
activate the original job through a data load it will result in a
duped job.
So there is no harm and still do not agree....
If only it were the case!
but i am fine, it is not THAT important to me.
I'm sorry to hear that.
PS. I would appreciate if you discuss things with me, and not order
and
not show your power.
Well, I suppose one good accusation (ie me accusing you of not looking
into this adequately) deserves another (ie you accusing me of using a
power position to try to influence you).
If I had tried to use a power position I wouldn't have tried to
explain myself. I sincerely hope that I never push anyone into
anything. This is a community after all and I'm no one's boss, but a
powerless moderated who feels the need to write a lot in hopes that
people will read it and it will result in better results overall. I
know it doesn't always work, but I keep trying.
-David
On Sun, 2009-01-11 at 22:28 -0800, David E Jones wrote:
I see where you're going with this now, and I don't like it.
Have you tested an update with this? Do you know what will happen?
No.
So, either trust me or look back at the comments where I introduced
this to find out. I'll tell you what happens: each time the seed data
is reloaded a new job will be created and the old job (if it has been
run at least once) will still be there, meaning that each time the
job
data is reloaded it will lead to more and more of the job running.
This was found through painful experience where a client had around
30
of certain jobs running redundantly.
This is the very sort of reason why I wrote this as Rule #2 for a
committer
(http://docs.ofbiz.org/display/OFBADMIN/OFBiz+Committers+Roles+and+Responsibilities
):
"Rule #2 for a committer is the same as for a scientist: read before
you write. When you're getting started a good time ratio for read to
write is around 20 to 1. Once you're a total OFBiz pro who knows as
much as any living person about the project, you can probably reduce
that to about 3 to 1. This relates to respecting precedent. It
doesn't
mean that existing things can't or shouldn't be changed, but it does
mean that things that already exist must be understood before they
are
changed. This also relates to recognizing that whatever you are doing
chances are there are best practices or patterns already established.
So, this means you should look for those and try to understand them
and if necessary ask about them explaining what you are trying to do
before you seek to establish your own pattern."
I'm being very forward about this because it's a pattern of behavior
that is common to all of us unless we learn better and always watch
out for it, and it is one that does uncountable harm. It is also
something that gets in the way of people working together in a
community (ie people not respecting and trying to understand what
others are doing or have done).
When initially setting up a system all seed and seed-initial data
should be loaded, as described in the various comments about them.
So,
I don't see the problem that you described.
I'm going to change this back now so no one else is harmed by it
unless they decide to change it.
-David
On Jan 11, 2009, at 10:18 PM, Hans Bakker wrote:
David,
I did this because i suspected that the data has been changed. and
anyhow when the data will be updated in the future it will simply
overwrite what is there? so what is the harm having it in seed data?
It is an essential part of the system....other wise mail will not go
out...if it is not used it will do no harm and hardly takes any
system
resources. So i think there is a good reason for doing this.
i also saw that the scheduled job for fixed assets also is in seed
data...
regards,
Hans
On Sun, 2009-01-11 at 21:18 -0800, David E Jones wrote:
Hans,
Could explain why you made this change?
Generally we want things like scheduled jobs to be loaded once and
not
every time a software update is done, meaning it should be in seed-
initial and not in seed.
For more information on the different data readers, check out the
build.xml file in the ofbiz.home directory, specifically the
comments
for the run-install-seed, run-install-extseed, and run-install-
exttest
targets. To help clarify this I have also added more comments to
the
entityengine.xml file where these are declared.
Unless there is a good reason for this change that I'm missing, and
you understand all of this, please revert it (or I can certainly do
it
too).
-David
On Jan 11, 2009, at 9:01 PM, hans...@apache.org wrote:
Author: hansbak
Date: Sun Jan 11 21:01:01 2009
New Revision: 733624
URL: http://svn.apache.org/viewvc?rev=733624&view=rev
Log:
change sheduled jobs to seed data
Modified:
ofbiz/trunk/applications/party/ofbiz-component.xml
Modified: ofbiz/trunk/applications/party/ofbiz-component.xml
URL:
http://svn.apache.org/viewvc/ofbiz/trunk/applications/party/ofbiz-component.xml?rev=733624&r1=733623&r2=733624&view=diff
=
=
=
=
=
=
=
=
=
=
=
=
==================================================================
--- ofbiz/trunk/applications/party/ofbiz-component.xml (original)
+++ ofbiz/trunk/applications/party/ofbiz-component.xml Sun Jan 11
21:01:01 2009
@@ -29,7 +29,7 @@
<entity-resource type="model" reader-name="main" loader="main"
location="entitydef/entitymodel.xml"/>
<entity-resource type="data" reader-name="seed" loader="main"
location="data/PartyTypeData.xml"/>
<entity-resource type="data" reader-name="seed" loader="main"
location="data/PartySecurityData.xml"/>
- <entity-resource type="data" reader-name="seed-initial"
loader="main" location="data/ScheduledJobs.xml"/>
+ <entity-resource type="data" reader-name="seed" loader="main"
location="data/ScheduledJobs.xml"/>
<service-resource type="model" loader="main"
location="servicedef/services.xml"/>
<service-resource type="model" loader="main"
location="servicedef/services_view.xml"/>
<service-resource type="eca" loader="main" location="servicedef/
secas.xml"/>
--
Antwebsystems.com: Quality OFBiz services for competitive prices
--
Antwebsystems.com: Quality OFBiz services for competitive prices
