You are viewing a plain text version of this content. The canonical link for it is here.
Posted to users@ws.apache.org by John Southerland <jo...@southerland-consulting.com> on 2005/12/29 20:43:47 UTC
AIX 5.2 Java 1.4 WebServer Garbage in the stream?
Wade knocked out my ssl issue in short order, thanks again, but I still have
a connection problem on AIX using http.
I enabled debug for XmlRpc and tried once again to send a message the first
test, a simple telnet to the port and "GET / HTTP/1.0" returned nothing,
but printed on the remote in all it's accurate glory, however the actually
xmlrpc request did something more interesting. I didn't capture the raw
request but it was generated from a linux box running jvm 1.5 against the
remote AIX 5.2 server with a 1.5 jvm as well.
To make a long setup short, here is the resulting output on the server:
b9 /32
java.util.NoSuchElementException
java.util.NoSuchElementException
at java.util.StringTokenizer.nextToken(StringTokenizer.java:347)
at org.apache.xmlrpc.WebServer$Connection.run(WebServer.java)
at org.apache.xmlrpc.WebServer$Runner.run(WebServer.java)
at java.lang.Thread.run(Thread.java:788)
Anyone seen this before? Is it network byte order versus host byte order
conversion error or something?
Once again the same code runs brilliantly on Linux and Solaris, fun fun
Thanks, John
John Buren Southerland
Southerland Consulting
801.467.8090(office)
214.734.8099(cell)
john@southerland-consulting.com
_____
From: John Southerland [mailto:john@southerland-consulting.com]
Sent: Wednesday, December 28, 2005 2:17 PM
To: xmlrpc-user@ws.apache.org
Subject: AIX 5.2 Java 1.3.1 and 1.4 SecureWebServer IOException
All,
I have recently tried to move my xmlrpc server app to a new os, AIX 5.2
running Java 1.3.1 and tried 1.4, every time I try to startup a
SecureWebServer, that works fine in Solaris and Linux, I get an IOException
from the constructor with the text: "com.sun.net.ssl.internal.ssl.Provider"
I realize this is most likely an IBM JRE issue, but the security properties
look fine, and I am definitely not overriding the system wide configuration.
This is actually a simple configuration, and since it works well on other
operating systems, I was wondering if anyone else has encountered errors
like this, or has any ideas on how I should continue. I'm now officially
stumped.
Also,
To add some comedy to this https problem, my non secure "backup" non ssl
webserver (also the builtin xmlrpc server) starts fine on the server but the
client gets an IOException on connect with the text "Remote host closed
connection during handshake", not much of a backup connectivity method is
it?
So please help me if you can, am I crazy, or is IBM to blame?
Thanks in advance, John
John Buren Southerland
Southerland Consulting
801.467.8090(office)
214.734.8099(cell)
john@southerland-consulting.com
RE: AIX 5.2 Java 1.4 WebServer Garbage in the stream?
Posted by John Southerland <jo...@southerland-consulting.com>.
PS: for posterity that string " b9 /32" is the text equivalent of an ssl
handshake.
Enjoy, John
John Buren Southerland
Southerland Consulting
801.467.8090(office)
214.734.8099(cell)
john@southerland-consulting.com
-----Original Message-----
From: John Southerland [mailto:john@southerland-consulting.com]
Sent: Thursday, December 29, 2005 4:11 PM
To: hwadechandler-apache@yahoo.com
Cc: xmlrpc-user@ws.apache.org
Subject: RE: AIX 5.2 Java 1.4 WebServer Garbage in the stream?
Woops, it turns out httpd is not working on linux either, I did not fully
implement it because I didn't need it until now, and it got left behind in
the code march.
Sorry for the interrupt, RTFM
Thanks again, John
John Buren Southerland
Southerland Consulting
801.467.8090(office)
214.734.8099(cell)
john@southerland-consulting.com
-----Original Message-----
From: Wade Chandler [mailto:hwadechandler-apache@yahoo.com]
Sent: Thursday, December 29, 2005 4:01 PM
To: xmlrpc-user@ws.apache.org
Subject: Re: AIX 5.2 Java 1.4 WebServer Garbage in the stream?
--- John Southerland <jo...@southerland-consulting.com>
wrote:
> Wade knocked out my ssl issue in short order, thanks
> again, but I still have
> a connection problem on AIX using http.
>
> I enabled debug for XmlRpc and tried once again to
> send a message the first
> test, a simple telnet to the port and "GET /
> HTTP/1.0" returned nothing,
> but printed on the remote in all it's accurate
> glory, however the actually
> xmlrpc request did something more interesting. I
> didn't capture the raw
> request but it was generated from a linux box
> running jvm 1.5 against the
> remote AIX 5.2 server with a 1.5 jvm as well.
>
> To make a long setup short, here is the resulting
> output on the server:
>
> b9 /32
>
> java.util.NoSuchElementException
>
> java.util.NoSuchElementException
>
> at
>
java.util.StringTokenizer.nextToken(StringTokenizer.java:347)
>
> at
>
org.apache.xmlrpc.WebServer$Connection.run(WebServer.java)
>
> at
>
org.apache.xmlrpc.WebServer$Runner.run(WebServer.java)
>
> at java.lang.Thread.run(Thread.java:788)
>
>
>
> Anyone seen this before? Is it network byte order
> versus host byte order
> conversion error or something?
>
> Once again the same code runs brilliantly on Linux
> and Solaris, fun fun
>
> Thanks, John
>
>
>
> John Buren Southerland
Should not an issue with the byte order (only matters
for floats and doubles and numbers not character
formatted data), but what appears to be happening (I
haven't looked at the source for WebServer.java, but
it is not checking hasMoreTokens in StringTokenizer
before calling nextToken...I'm not sure exactly what
it is parsing, but if there should be another value
then what really needs to happen is an exception needs
to be thrown stating there should be more elements and
what the real issue is and explain that something
isn't formatted correctly instead of a blind nextToken
read which is what is happening (unless there is no
blind nextToken and hasMoreTokens is checking before
calling nextToken and it is a bug in the IBM jvm).
Wade
RE: AIX 5.2 Java 1.4 WebServer Garbage in the stream?
Posted by John Southerland <jo...@southerland-consulting.com>.
Woops, it turns out httpd is not working on linux either, I did not fully
implement it because I didn't need it until now, and it got left behind in
the code march.
Sorry for the interrupt, RTFM
Thanks again, John
John Buren Southerland
Southerland Consulting
801.467.8090(office)
214.734.8099(cell)
john@southerland-consulting.com
-----Original Message-----
From: Wade Chandler [mailto:hwadechandler-apache@yahoo.com]
Sent: Thursday, December 29, 2005 4:01 PM
To: xmlrpc-user@ws.apache.org
Subject: Re: AIX 5.2 Java 1.4 WebServer Garbage in the stream?
--- John Southerland <jo...@southerland-consulting.com>
wrote:
> Wade knocked out my ssl issue in short order, thanks
> again, but I still have
> a connection problem on AIX using http.
>
> I enabled debug for XmlRpc and tried once again to
> send a message the first
> test, a simple telnet to the port and "GET /
> HTTP/1.0" returned nothing,
> but printed on the remote in all it's accurate
> glory, however the actually
> xmlrpc request did something more interesting. I
> didn't capture the raw
> request but it was generated from a linux box
> running jvm 1.5 against the
> remote AIX 5.2 server with a 1.5 jvm as well.
>
> To make a long setup short, here is the resulting
> output on the server:
>
> b9 /32
>
> java.util.NoSuchElementException
>
> java.util.NoSuchElementException
>
> at
>
java.util.StringTokenizer.nextToken(StringTokenizer.java:347)
>
> at
>
org.apache.xmlrpc.WebServer$Connection.run(WebServer.java)
>
> at
>
org.apache.xmlrpc.WebServer$Runner.run(WebServer.java)
>
> at java.lang.Thread.run(Thread.java:788)
>
>
>
> Anyone seen this before? Is it network byte order
> versus host byte order
> conversion error or something?
>
> Once again the same code runs brilliantly on Linux
> and Solaris, fun fun
>
> Thanks, John
>
>
>
> John Buren Southerland
Should not an issue with the byte order (only matters
for floats and doubles and numbers not character
formatted data), but what appears to be happening (I
haven't looked at the source for WebServer.java, but
it is not checking hasMoreTokens in StringTokenizer
before calling nextToken...I'm not sure exactly what
it is parsing, but if there should be another value
then what really needs to happen is an exception needs
to be thrown stating there should be more elements and
what the real issue is and explain that something
isn't formatted correctly instead of a blind nextToken
read which is what is happening (unless there is no
blind nextToken and hasMoreTokens is checking before
calling nextToken and it is a bug in the IBM jvm).
Wade
Re: AIX 5.2 Java 1.4 WebServer Garbage in the stream?
Posted by Wade Chandler <hw...@yahoo.com>.
--- John Southerland <jo...@southerland-consulting.com>
wrote:
> Wade knocked out my ssl issue in short order, thanks
> again, but I still have
> a connection problem on AIX using http.
>
> I enabled debug for XmlRpc and tried once again to
> send a message the first
> test, a simple telnet to the port and "GET /
> HTTP/1.0" returned nothing,
> but printed on the remote in all it's accurate
> glory, however the actually
> xmlrpc request did something more interesting. I
> didn't capture the raw
> request but it was generated from a linux box
> running jvm 1.5 against the
> remote AIX 5.2 server with a 1.5 jvm as well.
>
> To make a long setup short, here is the resulting
> output on the server:
>
> b9 /32
>
> java.util.NoSuchElementException
>
> java.util.NoSuchElementException
>
> at
>
java.util.StringTokenizer.nextToken(StringTokenizer.java:347)
>
> at
>
org.apache.xmlrpc.WebServer$Connection.run(WebServer.java)
>
> at
>
org.apache.xmlrpc.WebServer$Runner.run(WebServer.java)
>
> at java.lang.Thread.run(Thread.java:788)
>
>
>
> Anyone seen this before? Is it network byte order
> versus host byte order
> conversion error or something?
>
> Once again the same code runs brilliantly on Linux
> and Solaris, fun fun
>
> Thanks, John
>
>
>
> John Buren Southerland
Should not an issue with the byte order (only matters
for floats and doubles and numbers not character
formatted data), but what appears to be happening (I
haven't looked at the source for WebServer.java, but
it is not checking hasMoreTokens in StringTokenizer
before calling nextToken...I'm not sure exactly what
it is parsing, but if there should be another value
then what really needs to happen is an exception needs
to be thrown stating there should be more elements and
what the real issue is and explain that something
isn't formatted correctly instead of a blind nextToken
read which is what is happening (unless there is no
blind nextToken and hasMoreTokens is checking before
calling nextToken and it is a bug in the IBM jvm).
Wade