I am building an indexing application that scans a file system using VFS2
and indexes with lucene.
After running for a long while, it eats up all the memory and often hits
OOME -- when I look at the heap dump, 75% of all memory usage is from
SoftRefFilesCache. I plan to switch to LRUFilesCache, b
> > > Java version: 1.7.0_11, vendor: Oracle Corporation
> > > Java home: C:\Program Files\Java\jdk1.7.0_11\jre
> > > Default locale: en_US, platform encoding: Cp1252
> > > OS name: "windows 7", version: "6.1", arch: "amd64", family: &qu
Also failing here:
https://builds.apache.org/job/commons-vfs-trunk/
but not for the same reason (looks more like jenkins config/resources issue)
On Tue, Jan 29, 2013 at 4:49 PM, Ryan McKinley wrote:
> are trunk tests failing for anyone else?
>
> I get:
> junit.framework.TestSu
are trunk tests failing for anyone else?
I get:
junit.framework.TestSuite@3cf7d6ff(org.apache.commons.vfs2.test.ProviderTestSuite)
Time elapsed: 8426 sec <<< ERROR!
org.apache.commons.vfs2.FileSystemException: Unknown message with code
"org.apache.commons.vfs2.FileNotFoundException Could not rea
Sorry for the double post.
Since my account sent a bunch of spam, I was kicked off the list and my
first message failed. I think signed up again and now both messages went
through.
thanks
ryan
On Tue, Jan 29, 2013 at 2:56 PM, Ryan McKinley wrote:
> Hello-
>
> I am using trun
Hello-
I am using trunk VFS and hitting an error when trying to resolve files.
This simple example demonstrates the problem:
FileObject root = VFS.getManager().resolveFile( "
ftp://atlas.ca.gov/pub/casil/"; );
FileObject file = root.resolveFile("
ftp://atlas.ca.gov/pub/casil/transportation/READM
everybody else, I'm busy with a family, a job, and hobbies...
>
> All that being said, a house is built one brick at a time, so this could be
> started with one provier...
>
> Volunteers welcome!
>
> Gary
>
>
> On Mon, Jan 14, 2013 at 3:00 PM, Ryan McKinley wro
Hello-
I just switched from my own exec wrappers to commons exec, and it is
much better -- thank you thank you!
However, one thing I notice (especially while profiling) is that every
task and watchdog gets a new Thread. This is ok -- but gets kinda
overwhelming with lots of short lived processe
