You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@forrest.apache.org by CFAS Webmaster <we...@cfas.org> on 2006/01/01 00:21:41 UTC
Odd error...
All,
Happy New Year!!
I'm getting this message from forrest (svn HEAD) and I don't
understand what it's really telling me... I've removed, touched, edited
and almost anything else I could think of to get this thing to compile...
...
^ Calendars/2006Mar.html
^
skin/images/rc-t-r-5-1header-2searchbox-3searchbox.png
^ Committees/programs.html
X [0] Committees/membership.html
BROKEN: URI not found: File
/home/kronenpj/src/svn/cfas/build/site/Committees/membership.html is
already being written by another thread
java.util.ConcurrentModificationException: File
/home/kronenpj/src/svn/cfas/build/site/Committees/membership.html is
already being written by another thread
at
org.apache.excalibur.source.impl.FileSource.getOutputStream(FileSource.java:326)
at
org.apache.cocoon.bean.CocoonBean.processTarget(CocoonBean.java:545)
at org.apache.cocoon.bean.CocoonBean.process(CocoonBean.java:357)
at org.apache.cocoon.Main.main(Main.java:311)
^ Committees/MembershipApplication2.html
...
I've just updated and re-built forrest. This site is still using the
pelt skin and the forrest.properties file has this in it:
project.name=CFAS
project.skin=pelt
project.required.plugins=org.apache.forrest.plugin.input.projectInfo,org.apache.forrest.plugin.output.pdf,org.apache.forrest.plugin.input.OpenOffice.org,org.apache.forrest.plugin.input.PhotoGallery
project.configfile=${project.home}/src/documentation/conf/cli.xconf
I'm using checksums to reduce the amount of processing cocoon is doing,
but I've removed the entry for membership.html to be sure it's compiled...
It looks to be similar to Ross's discovery:
http://marc.theaimsgroup.com/?l=forrest-dev&m=112971207119409&w=2
But there was no confirmation or further discussion...
Anyone have an idea on this? Should I open an issue?
Thanks!
-Paul
Re: Odd error...
Posted by CFAS Webmaster <we...@cfas.org>.
I think I found it...
For some reason there was a .tmp file left in build/site/... Removing
it allowed the page to be built.
-Paul
CFAS Webmaster wrote:
> All,
>
> Happy New Year!!
>
> I'm getting this message from forrest (svn HEAD) and I don't
> understand what it's really telling me... I've removed, touched,
> edited and almost anything else I could think of to get this thing to
> compile...
> ...
> ^ Calendars/2006Mar.html
> ^
> skin/images/rc-t-r-5-1header-2searchbox-3searchbox.png
> ^ Committees/programs.html
> X [0]
> Committees/membership.html BROKEN: URI not found: File
> /home/kronenpj/src/svn/cfas/build/site/Committees/membership.html is
> already being written by another thread
> java.util.ConcurrentModificationException: File
> /home/kronenpj/src/svn/cfas/build/site/Committees/membership.html is
> already being written by another thread
> at
> org.apache.excalibur.source.impl.FileSource.getOutputStream(FileSource.java:326)
>
> at
> org.apache.cocoon.bean.CocoonBean.processTarget(CocoonBean.java:545)
> at org.apache.cocoon.bean.CocoonBean.process(CocoonBean.java:357)
> at org.apache.cocoon.Main.main(Main.java:311)
> ^
> Committees/MembershipApplication2.html
> ...
>
> I've just updated and re-built forrest. This site is still using the
> pelt skin and the forrest.properties file has this in it:
> project.name=CFAS
> project.skin=pelt
> project.required.plugins=org.apache.forrest.plugin.input.projectInfo,org.apache.forrest.plugin.output.pdf,org.apache.forrest.plugin.input.OpenOffice.org,org.apache.forrest.plugin.input.PhotoGallery
>
> project.configfile=${project.home}/src/documentation/conf/cli.xconf
>
> I'm using checksums to reduce the amount of processing cocoon is
> doing, but I've removed the entry for membership.html to be sure it's
> compiled...
>
> It looks to be similar to Ross's discovery:
> http://marc.theaimsgroup.com/?l=forrest-dev&m=112971207119409&w=2
> But there was no confirmation or further discussion...
>
> Anyone have an idea on this? Should I open an issue?
>
> Thanks!
> -Paul
>
Re: Checksums to improve processing time [Re: Odd error...]
Posted by Ross Gardler <rg...@apache.org>.
CFAS Webmaster wrote:
> Ross,
>
> I do find them useful for one thing. Since I haven't yet been able to
> get forrestbot working, I use a different program to upload files to the
> "production" site. That utility primarily uses date/time stamps from
> files to figure out which ones to upload.
> The checksums stop this utility from uploading the whole site every
> time I update something. Unfortunately the checksums do not, as David
> and others have pointed out, save generation time as they compare the
> generated file with what is on disk.
That sounds useful for the Google Sitemap thing too. In that case we
want to list which pages have changed since the last time the site was
built.
It would also be useful in the new DevTools plugin, we could use the
date and time stamp to calculate how many pages have changed since the
last full site build - thereby giving another indication of the
likelyhood of a corrupt site.
> Does anyone think there would be a way to capture checksums of the
> *source* files so that Cocoon doesn't even look at unchanged files?
THat would be one for the Cocoon list.
> This would need to be done in such a way that if site-related data such
> as copyrights or left-side menu items changed, the whole site is
> regenerated... Probably very tricky to do correctly.
Not really, that data is in specific files, site.xml, tabs.xml and
skinconf.xml - all we need to do is detect when those files have changed.
We need this in the DevTools plugin (see thread on saving files from
Forrest Run).
Ross
Re: Checksums to improve processing time [Re: Odd error...]
Posted by CFAS Webmaster <we...@cfas.org>.
Ross,
I do find them useful for one thing. Since I haven't yet been able to
get forrestbot working, I use a different program to upload files to the
"production" site. That utility primarily uses date/time stamps from
files to figure out which ones to upload.
The checksums stop this utility from uploading the whole site every
time I update something. Unfortunately the checksums do not, as David
and others have pointed out, save generation time as they compare the
generated file with what is on disk.
Does anyone think there would be a way to capture checksums of the
*source* files so that Cocoon doesn't even look at unchanged files?
This would need to be done in such a way that if site-related data such
as copyrights or left-side menu items changed, the whole site is
regenerated... Probably very tricky to do correctly.
-Paul
Ross Gardler wrote:
> CFAS Webmaster wrote:
>
>> I'm using checksums to reduce the amount of processing cocoon is
>> doing, but I've removed the entry for membership.html to be sure it's
>> compiled...
>
>
> This has been discussed a few times an some initial findings have been
> reported. The above seems to imply that you've been using it for a
> while and find it to be useful.
>
> Is there any chance of you knocking up a Jira issue to referesh our
> memories about how to enable this (perhaps you can even attech a aptch
> ;-)
>
> If you can add any observations from your experience about
> effectiveness that would be cool too.
>
> Ross
Re: Checksums to improve processing time [Re: Odd error...]
Posted by David Crossley <cr...@apache.org>.
David Crossley wrote:
> CFAS Webmaster wrote:
> > Ok. I'll probably do a howto, but it'll be reasonably short as it's not
> > complicated.
> >
> > -Paul
> >
> > Ross Gardler wrote:
> > >CFAS Webmaster wrote:
> > >
> > >> I'm poised to "knock up" a JIRA issue on this, but I want to make
> > >>sure that David's observation isn't telling me not to...
> > >
> > >No he was pointing to the fact that the information I was asking for
> > >is already in the mail archives and should be captured in
> > >documentation. If you don't have the time to write a doc, some links
> > >to the archives in a Jira issue will be a good start.
>
> I hope that you do not waste time, Paul.
>
> Please find the mails in the archives.
> There was big discussion and analysis and
> conclusions. I don't think that it needs a howto,
> perhaps just an FAQ to say that there is no effect
> on Cocoon processing and that a little time can be
> saved on filesystem writes, which will accumulate
> to a big saving for a site with thousands of files.
Compare with other FAQs ... some are quite detailed.
If you reckon that it needs more, then sure, make a Howto
doc (perhaps with a link from a concise summary FAQ).
-David
Re: Checksums to improve processing time [Re: Odd error...]
Posted by David Crossley <cr...@apache.org>.
CFAS Webmaster wrote:
> Ok. I'll probably do a howto, but it'll be reasonably short as it's not
> complicated.
>
> -Paul
>
> Ross Gardler wrote:
>
> >CFAS Webmaster wrote:
> >
> >>All,
> >>
> >> I'm poised to "knock up" a JIRA issue on this, but I want to make
> >>sure that David's observation isn't telling me not to...
> >
> >No he was pointing to the fact that the information I was asking for
> >is already in the mail archives and should be captured in
> >documentation. If you don't have the time to write a doc, some links
> >to the archives in a Jira issue will be a good start.
I hope that you do not waste time, Paul.
Please find the mails in the archives.
There was big discussion and analysis and
conclusions. I don't think that it needs a howto,
perhaps just an FAQ to say that there is no effect
on Cocoon processing and that a little time can be
saved on filesystem writes, which will accumulate
to a big saving for a site with thousands of files.
-David
Re: Checksums to improve processing time [Re: Odd error...]
Posted by CFAS Webmaster <we...@cfas.org>.
Ok. I'll probably do a howto, but it'll be reasonably short as it's not
complicated.
-Paul
Ross Gardler wrote:
> CFAS Webmaster wrote:
>
>> All,
>>
>> I'm poised to "knock up" a JIRA issue on this, but I want to make
>> sure that David's observation isn't telling me not to...
>
>
> No he was pointing to the fact that the information I was asking for
> is already in the mail archives and should be captured in
> documentation. If you don't have the time to write a doc, some links
> to the archives in a Jira issue will be a good start.
>
> Ross
Re: Checksums to improve processing time [Re: Odd error...]
Posted by Ross Gardler <rg...@apache.org>.
CFAS Webmaster wrote:
> All,
>
> I'm poised to "knock up" a JIRA issue on this, but I want to make sure
> that David's observation isn't telling me not to...
No he was pointing to the fact that the information I was asking for is
already in the mail archives and should be captured in documentation. If
you don't have the time to write a doc, some links to the archives in a
Jira issue will be a good start.
Ross
Re: Checksums to improve processing time [Re: Odd error...]
Posted by CFAS Webmaster <we...@cfas.org>.
All,
I'm poised to "knock up" a JIRA issue on this, but I want to make sure
that David's observation isn't telling me not to...
-Paul
David Crossley wrote:
>Ross Gardler wrote:
>
>
>>CFAS Webmaster wrote:
>>
>>
>>>I'm using checksums to reduce the amount of processing cocoon is doing,
>>>but I've removed the entry for membership.html to be sure it's compiled...
>>>
>>>
>>This has been discussed a few times an some initial findings have been
>>reported. The above seems to imply that you've been using it for a while
>>and find it to be useful.
>>
>>Is there any chance of you knocking up a Jira issue to referesh our
>>memories about how to enable this (perhaps you can even attech a aptch ;-)
>>
>>If you can add any observations from your experience about effectiveness
>>that would be cool too.
>>
>>
>
>There is a very clear description of the effect
>in the forrest mail archives. Someone needs to
>turn that into documentation.
>
>By the way, it has no effect on Cocoon processing.
>That still happens. Some little time is saved on
>filesystem writes.
>
>-David
>
>
Re: Checksums to improve processing time [Re: Odd error...]
Posted by David Crossley <cr...@apache.org>.
Ross Gardler wrote:
> CFAS Webmaster wrote:
> >I'm using checksums to reduce the amount of processing cocoon is doing,
> >but I've removed the entry for membership.html to be sure it's compiled...
>
> This has been discussed a few times an some initial findings have been
> reported. The above seems to imply that you've been using it for a while
> and find it to be useful.
>
> Is there any chance of you knocking up a Jira issue to referesh our
> memories about how to enable this (perhaps you can even attech a aptch ;-)
>
> If you can add any observations from your experience about effectiveness
> that would be cool too.
There is a very clear description of the effect
in the forrest mail archives. Someone needs to
turn that into documentation.
By the way, it has no effect on Cocoon processing.
That still happens. Some little time is saved on
filesystem writes.
-David
Checksums to improve processing time [Re: Odd error...]
Posted by Ross Gardler <rg...@apache.org>.
CFAS Webmaster wrote:
> I'm using checksums to reduce the amount of processing cocoon is doing,
> but I've removed the entry for membership.html to be sure it's compiled...
This has been discussed a few times an some initial findings have been
reported. The above seems to imply that you've been using it for a while
and find it to be useful.
Is there any chance of you knocking up a Jira issue to referesh our
memories about how to enable this (perhaps you can even attech a aptch ;-)
If you can add any observations from your experience about effectiveness
that would be cool too.
Ross