On 11/20/06, David R Robison <drrobison@openroadsconsulting.com> wrote:
>
> I'm sorry, I'm unsure what your asking.
Sorry for the confusion. My question is do you use rev="latest.milestone" or
rev="latest.release" in any of your ivy files. But if the question is not
clear I'm pretty sure you don't use them. So in this case the slowness may
be due to a fix side effect or something like that, but it's difficult to
know. To improve your perf you can set allownomd="false" on your resolvers
for which you always have an ivy file. You can also set
alwaysCheckExactRevision="false" on your resolvers so that Ivy does not try
to find "latest.integration" revision, but only consider it as a dynamic
version. Anyway, you seem to say that Ivy doesn't trust its cache anymore
with 1.4.1, which is strange. It shouldn't be the case unless you use
checkModified="true" and/or changing revisions. If you don't maybe it's a
bug in 1.4.1. If you post your debug trace (what you see on the console when
you run "ant -d") then maybe I'll be able to better understand what's
happening.
Xavier
Here is my ivy config file.
>
> <ivyconf>
> <properties file="${ivy.conf.dir}/ivyconf.properties" />
> <conf defaultCache="${ivy.conf.dir}/ivy-cache"
> defaultResolver="default-chain"/>
> <resolvers>
> <chain name="default-chain">
> <url name="orci">
> <artifact
> pattern="
> http://www.openroadsconsulting.com/maven/[module]/[type]s/[artifact]-[revision].[ext]
> "
> />
> <artifact
> pattern="
> http://www.openroadsconsulting.com/maven/[module]/[type]s/[artifact].[ext]
> "
> />
> </url>
> <url name="ibiblio" >
> <artifact
> pattern="
> http://mirrors.dotsrc.org/maven/[module]/[type]s/[artifact]-[revision].[ext]
> "
> />
> <artifact
> pattern="http://mirrors.dotsrc.org/maven/[module]/[type]s/[artifact].[ext]
> "
> />
> <artifact
> pattern="
> http://mirrors.dotsrc.org/maven2/org/samba/jcifs/[module]/[revision]/[artifact]-[revision].[ext]
> "
> />
> </url>
> <url name="astrogrid">
> <artifact
> pattern="
> http://www.astrogrid.org/maven/[module]/[type]s/[artifact]-[revision].[ext]
> "
> />
> </url>
> <url name="bluesunrise">
> <artifact
> pattern="
> http://www.bluesunrise.com/maven/[module]/[type]s/[artifact]-[revision].[ext]
> "
> />
> </url>
> <url name="jibx">
> <artifact
> pattern="
> http://jibx.sourceforge.net/maven/[module]/[type]s/[artifact]-[revision].[ext]
> "
> />
> </url>
> <url name="berlios">
> <artifact
> pattern="
> http://osgirepo.berlios.de/repository/[module]/[type]s/[artifact]-[revision].[ext]
> "
> />
> </url>
> <url name="geotools">
> <artifact
> pattern="
> http://maven.geotools.fr/repository/org/geotools/[module]/[revision]/[artifact]-[revision].[ext]
> "
> />
> <artifact
> pattern="
> http://maven.geotools.fr/repository/com/vividsolutions/[module]/[revision]/[artifact]-[revision].[ext]
> "
> />
> <artifact
> pattern="
> http://maven.geotools.fr/repository/org/opengis/[module]/[revision]/[artifact]-[revision].[ext]
> "
> />
> <artifact
> pattern="
> http://maven.geotools.fr/repository/javax/units/[module]/[revision]/[artifact]-[revision].[ext]
> "
> />
> </url>
> <url name="orci">
> <artifact
> pattern="
> http://www.openroadsconsulting.com/maven/[module]/[type]s/[artifact]-[revision].[ext]
> "
> />
> <artifact
> pattern="
> http://www.openroadsconsulting.com/maven/[module]/[type]s/[artifact].[ext]
> "
> />
> </url>
> </chain>
> <filesystem name="projects">
> <artifact
> pattern="${repository.dir}/[artifact]-[revision].[ext]" />
> <ivy pattern="${repository.dir}/[module]-[revision].xml"
> />
> </filesystem>
> </resolvers>
> <modules>
> <module organisation="orci" name=".*" resolver="projects" />
> </modules>
> </ivyconf>
>
> Does this help? David
>
> Xavier Hanin wrote:
> > On 11/20/06, David Robison <drrobison@openroadsconsulting.com> wrote:
> >>
> >> I just upgraded from 1.3.1 to 1.4.1 and now my resolves are taking
> >> a substantially longer time. It seems like it is requiring that
> >> each resolve has to find the file on the Internet even though it
> >> already exists in the cache. Is there some configuration I'm missing?
> >> Thanks, David
> >>
> > No there is no strange configuration. Are you using latest.milestone or
> > latest.release? If you are, prior to 1.4 the behaviour was rather
> > different,
> > and now the proper implementation causes slowness. Otherwise could you
> > provide more details on your resolve (verbose or debug log)?
> >
> > Xavier
> >
>
> --
>
> David R Robison
> Open Roads Consulting, Inc.
> 708 S. Battlefield Blvd., Chesapeake, VA 23322
> phone: (757) 546-3401
> e-mail: drrobison@openroadsconsulting.com
> web: http://openroadsconsulting.com
> blog: http://therobe.blogspot.com
> book: http://www.xulonpress.com/bookstore/titles/1597816523.htm
>
>
>
>
>
|