tomcat-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Adam Hardy <>
Subject Re: charset problems coming up during runtime
Date Wed, 05 Nov 2003 19:45:08 GMT
On 11/05/2003 07:02 PM Christoph Lechleitner wrote:
> Basically the problem manifests in producing the html output.
> i.e., the so far correct data from the DB or file are not getting encoded,
> the browser gets ? instead of &auml; (for รค).
> If this happens when filling forms with current values of some parameter,
> the error indirectly finds its way into the DB or file.
>>Which character set are you using? iso-8859-1 or iso-8859-15? You say 
>>you have your OS, your Java, appserver and database all set to use one 
>>of these, if I understand correctly. Presumably consistently the one or 
>>the other and not a mix?
> AFAIK they are the same except the EURO symbol which we don't need so far.
> Nevertheless, all components are supposed to use iso-8859-1.
>>Is this app in production? 
> Yes.
> The problem occurs in a new Application as well as in another which is
> in production for 2 years now. (So far, we tried to notice the problem
> each time it occurs and to correct eventual damages manually)
>>What sort of load is it handling? I ask to 
>>see what the feasibility of changing the appserver is - try the 90 day 
>>trial of weblogic for instance. Does that suffer the problem too?
> I don't think it's a general system problem.
> The production site is a shared host server with about 200 virtual 
> webservers, about 50 of it use PHP4 (this includes comples CMS appps), 
> 30 mysql, 20 Perl (including the Webmailer neomail); none of these
> has a problem with these characters.
> Btw., the problem also has occured on the development and on the test 
> system, both without any load.
>>What about changing to IBM Java? 
> I am actually thinking about this, perhaps I just try.
>>Or even from Linux to Windows?
> Windows?
> The server room has no Windows, it's underground ;->>>
> Seriously, Windows is not an option, we need reliable, calculable and 
> trustworthy systems.
> kr
> Christoph Lechleitner
> ---------------------------------------------------------------------
> To unsubscribe, e-mail:
> For additional commands, e-mail:

Well OK so now we know the problem:

- it lies with the character set interpretation in the HTML output from 
the database and files - which files? I'm not familiar with file reading 
  or whether you have to specify a character set.

- it occurs after random time intervals after starting tomcat. It 
doesn't ever go away again does it?

- the problem is overcome currently by restarting tomcat. This narrows 
it down to Java or Tomcat basically. So you can forget the OS or the 

- you manually encode umlauted letters into HTML code. I presume you 
have a filter with a big switch case construct that parses your output 
strings. Perhaps you should log the output of the ASCII value of the 
umlauted letters, to see what you are really getting instead of what you 
should be getting.

Where does the filter operate in your app? Obviously you have already 
fetched your output string from either JDBC or from a file at the point 
when you apply the filter. It is unlikely that both JDBC and the 
file-reading processes always go wrong simulataneously. That is the case 

struts 1.1 + tomcat 5.0.12 + java 1.4.2
Linux 2.4.20 RH9

To unsubscribe, e-mail:
For additional commands, e-mail:

View raw message