Hello Mark,
Unfortunately, one of my tests was made using the simplest scenario
possible, but everything worked as expected and I wasn't able to reproduce
it this way.
Best regards,
Diego
On Mon, Feb 27, 2017 at 1:59 PM, Mark Thomas <ma...@apache.org> wrote:
> On 27/02/17 10:02, Diego Urenia wrote:
> > *UPDATE: *Making further tests while I was waiting for some answers I
> found
> > out that if I explode the war by myself (unziping the war file into the
> > webapps folder) I still experience the problem. So I think we can rule
> out
> > the assumption that the problem happens while the application is still
> > exploding the war file and that the file size affects it.
> >
> > Another info that might be useful:
> >
> > - I am creating this folder using java.io.File#mkdir();
> > - I am trying to check if this folder exist using
> > Class#getResources("/folder");
>
> You need to create the simplest possible test case that demonstrates the
> problem (e.g. a single servlet) and then show us the code.
>
> Mark
>
>
> >
> > Best regards,
> > Diego Urneia
> >
> > On Mon, Feb 27, 2017 at 10:09 AM, Diego Urenia <dru...@gmail.com> wrote:
> >
> >> Hello all,
> >>
> >> My application creates a folder to store the config files at the first
> >> startup and immediately after that the application tries to access this
> >> folder in order to save some information.
> >>
> >> Using Tomcat 6 and 7 everything works fine.
> >> Using Tomcat 8, 8.5, and 9 the application experiences a weird behavior
> >> and here are some tests I made which made me think it was a war
> explosion
> >> related problem:
> >>
> >> - The problem doesn't happen when I am running the application using
> >> Intellij IDEA, because I use an exploded artifact;
> >>
> >> - I've created a dummy application to simulate only the creation and
> first
> >> access to this config folder (the same way I am doing in the real
> >> application) at startup and everything worked fine, which led me to
> believe
> >> the size of my application can be what revealed the problem (bigger .war
> >> file takes longer to be exploded);
> >>
> >> - Another test I made in my application was adding a Thread.sleep(5000);
> >> after creating the folder, which then makes the access to the folder
> >> possible (maybe because Tomcat had enough time to explode the war file);
> >>
> >> It was my belief before having this problem that Tomcat exploded the war
> >> before it even starts the application, but now I have my doubts.
> >>
> >> Three questions:
> >>
> >> - Does Tomcat really explode the war before starting the application?
> >>
> >> - If it doesn't, how does it work? (pointing the right piece of
> >> documentation, which I couldn't find is as good as an explanation)
> >>
> >> - Can anyone tell me what has changed in this regard from version 6 and
> >> 7 and the newer ones that led me to this problem?
> >>
> >> Best regards,
> >> Diego Urenia
> >>
> >>
> >>
> >
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
> For additional commands, e-mail: users-h...@tomcat.apache.org
>
>
