Not necessarily so. I just started writing a little int keyed Object
"map" backed by an object array last week.
The key difference is that if it is okay for the map to define the key,
then it's no problem at all. For each new object, just append it to the
end of the array. So the only holes tha
And using a JSP instead of a servlet would cause strange exceptions when
checking if the form is multi-part? I don't understand why. I thought JSPs
and servlets were pretty much the same thing in the end.
-Original Message-
From: Martin Cooper [mailto:[EMAIL PROTECTED]
Sent: Tuesday, Ma
Matthew
For testing, I would suggest you tried modifying the servlet I sent you and
try to see if you get the same exception.
Kind Regards
Schalk Neethling
Web Developer.Designer.Programmer.President
Volume4.Development.Multimedia.Branding
emotionalize.conceptualize.visualize.realize
Tel: +271254
I try to use the MultipartStream class in the FileUpdate to parse the
HTTP request in Multipart.
I think I found a bug if the Content length in the part is 0, it will
mess up the rest of the data and header.
Here is an example:
"--=ABC 123456789=\r\n" +
"Content-Type: tex
Title: Message
Hi,
In our project, we
have tomcat running on 2 servers which run in a load-balanced
mode.
We need to do some
code releases on these servers, after which the servers need to be
restarted.
We can not restart
the servers straight away as no pages will be served during that ti
Hello!
I'm using an Oracle database and I have to pass a CLOB value to a stored
procedure. How can I do that if the connection DBCP returns to me is not an
oracle connection?
Thanks,
JP
hi
this type of question is best addressed at the tomcat user mailing list.
FWIW you might consider configuring, testing and updating two new
tomcat instances. once they are working correctly, you can use the load
balancer to switch from the old to the new.
- robert
On 31 Mar 2004, at 16:44,
If numActive=8 and maxActive=8 then the connection pool has handed out
all the connection it has created and the pool isn't allowed to create
more. The default pool action in this situation is:
DEFAULT_WHEN_EXHAUSTED_ACTION = WHEN_EXHAUSTED_BLOCK
So clearly your app has abandoned (lost) some of
It sounds like your server is timing out and killing the session, but I'm not sure why
it would be different on the Solaris system.
I did some experiments with various large files. I first set my session timeout to 600
minutes. I then tested to see how long it took to upload the files from vario
What is the best way to change the DEFAULT_WHEN_EXHAUSTED_ACTION? (of
course this will only prolong the problem until it fills up the queue)
I don't think my code is causing the connections to be lost, which means
the root of the problem is out of my control, or is somewhere else (The
connections
Using setClob? ;-)
http://java.sun.com/j2se/1.4.2/docs/api/java/sql/PreparedStatement.html#setClob(int,%20java.sql.Clob)
If you need to access the underlying connection then set
accessToUnderlyingConnectionAllowed to true as described here:
http://jakarta.apache.org/commons/dbcp/configuration.html
You can change the "WhenExhaustedAction" using this setter:
((GenericObjectPool) connectionPool).setWhenExhaustedAction(...)
- WHEN_EXHAUSTED_BLOCK
- WHEN_EXHAUSTED_FAIL
- WHEN_EXHAUSTED_GROW
An AbandonedObjectPool has a AbandonedConfig configuration object.
See http://jakarta.apache.org/commons
I don't know if it is timeouts or not. The stream ended unexpectedly
exception leads to believe that it isn't. In addition, literally adding one
more byte to the file size causes this to fail. I've set the timeout time
to be ridiculously high and it doesn't make a difference. The addition of
on
As you have found, [el] was donated to commons, but has not been added to as
it meets its goals. You may find [jexl] of interest, as that has seen more
activity. Alternatively, you could try asking for an example on tomcat-user
mailing list as that is where the [el] authors should reside.
Stephen
> -Original Message-
> From: Tatu Vanhanen [mailto:[EMAIL PROTECTED]
> Sent: Wednesday, March 31, 2004 2:18 AM
> To: Jakarta Commons Users List
> Subject: RE: [primitives] Looking for a primitive hashtable
>
> I'm not the original sender, but interesting issue though.
>
> > From: Rodney W
At 11:16 AM 3/31/04 -0800, Bill Simpson wrote:
>In addition, literally adding one
>more byte to the file size causes this to fail. I've set the timeout time
>to be ridiculously high and it doesn't make a difference. The addition of
>one byte changes the amount of time from 2 minutes to the length
Check out setSizeThreshold and setRepositoryPath in DiskFileUpload. Maybe this is
where your problem lies.
At 11:44 AM 3/31/04 -0800, Cindy Ballreich wrote:
>At 11:16 AM 3/31/04 -0800, Bill Simpson wrote:
>>In addition, literally adding one
>>more byte to the file size causes this to fail. I've
(just adding to stephen's comments)
commons-el is an implementation of the JSP 2.0 expression language.
examples of JSP 2.0 EL should work with commons-el.
(just to reiterate stephen's recommendations) JEXL is an extension of
EL which is in more active development so you might like to take a
l
I have (finally) introduced explicit registration of namespace prefixes in
the expression context. Sorry it took so long.
You can now register prefix/namespace pairs with JXPathContext using the
registerNamespace() and/or setNamespaceContextPointer() methods.
Please let me know if my fixes indee
Joel,
> -Original Message-
> From: Joel Shellman [mailto:[EMAIL PROTECTED]
> Subject: Re: [primitives] Looking for a primitive hashtable
>
> Not necessarily so. I just started writing a little int keyed Object
> "map" backed by an object array last week.
>
> The key difference is that if
> -Original Message-
> From: Alex Karasulu [mailto:[EMAIL PROTECTED]
...
> Now the arrays would look like so after these operations are performed:
>
> KeysIndices Object Values
> --- -
> ||| | | +-|-> "Cat"
> --- -
21 matches
Mail list logo
