Dealing with client access to servers behind NAT

It depends on your firewall configuration. This is normally known as 'loopback' - a client internal to the network accessing a service on a WAN IP address will be directed to the correct server if loopback is enabled.
I remember, some years ago, dealing with firewalls which explicitly would not allow this to happen, requiring a split DNS strategy. I don't recall needing to regularly flush the DNS cache in this case.

Hi there,This is just a general question about "best practice".Imagine this scenario:a network protected by firewall, all LANs using private IP address, e.g. 10.x.x.x and/or 192.168.x.xaccess to the server from external (www/ftp/etc) will need to go through firewallsetup the firewall with 1-1 NAT
for example: access to 1.2.3.4 on port 8080 goes to the server 10.0.0.10 on port 8765using Windows Server 2008 as DC/DNS
My question is ... how the client deal with that?
I mean, let's say someone grab a laptop and goes to another country.
From the above scenario, accessing the server would be 1.2.3.4 as the destination IP address.But what happen when the laptop goes back to the office? accessing the same server will still give 1.2.3.4 as the destination since it's still cached, right?
does it mean that we need to set up a script to always flush...
This topic first appeared in the Spiceworks Community

Similar Messages

Software clustering with Client Access

Hello,
We are planning to by BO XI R3.0 with Client Access License. I heard that I can do software clustering only with user based or CPU based license and not with Client Access License. Could someone clarify me that can I do software clustering with Client Access License?
Thanks in advance.
Rasheed.

Well from what I understand is if you buy XI 3.0 then there are generally 3 license types available which can be mixed in some cases.
1) CPU license - unlimited users per system
2) named users license - you are allowed an amount of users in the system unlimited sessions per user
3) concurrent users where the license is based on the user sessions
So for the client some clients like crystal reports have a user license others like deski/designer can login locally fine but if you want to access server components then they fall under the above restrictions based on the license of your server.
Regards,
Tim

L2TP VPN for servers behind NAT

I have two 2012 R2 servers, both behind NAT, which I'm trying to connect via VPN. I have no problem connecting them via PPTP, but when connecting them via L2TP (with shared key for testing), the dialing server never connects to other server.
I assume that the problem is that they're both behind NAT. In Windows Server 2008, you were able to set a registry value to get the L2TP connections to work under NAT, see
http://support.microsoft.com/kb/926179 by setting the environment variable AssumeUDPEncapsulationContextOnSendRule.
I tried using this with the two servers, but it didn't seem to help. Is there some other way to get the L2TP connection for the two 2012 R2 servers working behind NAT?

Hi,
Thanks for your pointer and sorry for replying so late.
I am sorry to say that I haven’t found any documents to ensure whether NAT-T is supported in Windows server 2012 R2 or not. In addition,
VPN servers that are located behind NAT is not recommended. When a server is behind a network address translator, and the server uses NAT-T, unintended behavior might occur because
of the way NAT translate network traffic.
Best regards,
Susie

Time Capsule port mapping is broken for L2TP Servers behind NAT config.

I'm hoping that someone here can refute the below bug assertion... am I missing something?
There is a bug with Apple’s Time Capsule/Airport Express Base Station (TC/AEBS) rendering L2TP servers on the LAN unusable:
When TC/AEBS is used as a router providing NAT services to the LAN, it will NOT under any circumstance provide port mapping services for 500/UDP, 1701/UDP, & 4500/UDP making L2TP VPN servers on the LAN side of TC/AEBS are unreachable from the WAN/Internet side.
*The conditions for my tests*:
3 different external networks used for all tests: MacBook Air at home on TWC network, the Air on AT&T mobile dongle, & CentOS server at ThePlanet.
MobileMe configuration was removed from both the TC/AEBS & Snow Leopard Server on the LAN.
I used port 501 for my control-test; spot checks of other ports worked as well, though they were all < 10000.
Simultaneous local and server monitoring of port traffic using
tcpdump -vvv -i en0 -s 0 -X port 500 or port 1701 or port 4500 or port 501
The TC/AEBS was configured to forward UDP ports 501, 500, 1701, & 4500 received from the WAN interface to the Snow Leopard Server on the LAN.
The port forwarding was accomplished both 1) manually via AirPort Utility, and 2) automatically via Snow Leopard Server’s Server Preferences utility. Each was tested separately.
*The tests*:
Netcat with the following commands, in turn, on the server:
nc -l -u 501
nc -l -u 500
nc -l -u 1700
nc -l -u 4500
which causes traffic to the udp port specified to be dumped to std out. Provides a confirmation of the tcpdump output.
On the various external networks, nc -u WAN-address-of-AEBS.example.com 501 to send UDP packets on port 501. The output of the nc -l 501 command and the server-run tcpdump confirmed that packets left the client and made it to the server as expected. Remember, 501 is the control-test.
For each test permutation on ports 500, 1700, & 4500, no packets made it to the server.
Based on some web research, I’m not the only one to have found trouble with this configuration, but I haven’t been able to find any conclusive tests.
I’ve filed a bug with Apple (#7720101) and encourage you to do the same.
Message was edited by: WebMarc

Confirmed here. This only seems to be a problem with Airport 7.5.x firmware though - I find the older TCs running 7.4.2 work as expected even with BTMM / MobileMe services active.
I'm so glad you posted this - I haven't found it mentioned anywhere else and was beginning to feel very alone with this problem. I also found that having two TC 7.5s in the mix - one at both ends - also results in no response to SSH or Remote Desktop ports.

HT5390 What port should i forward in the router if I'm to connect to the mac server behind NAT?

I am trying to remote administrate a Mac Mini running Mac 10.8 server but couldn't figure out which port to use.
Tried the following but still woudn't connect unless DMZ is opened.
Remote Login (SSH) - 22
Screen Sharing Service (VNC) - 5900
Web Service - 80, 443
VPN Service (L2TP) - 500, 1701, 4500
VPN Service (PPTP) - 1723
Any clue?

When you say administer, do you mean something like control your parent's Mac remotely, or do you mean officially adminster a classroom full of Macs? If a classroom full of Macs, then you are most likely taking about using the Apple Remote Desktop software which you pay for.
If, as I suspect, you just want to control your own or a family member's Mac remotely, then you do not need to pay for anything.
If you need Screen Sharing, you open port 5900 (the VNC port)
If you need File Sharing, you open port 548 (AFP)
If you need access to the Unix command line, or you want to use the ssh 'scp' or 'sftp' file transfer commands, then you need to open port 22.
Visit <http://PortForward.com>, they will provide port forwarding instructions for just about every home router out there.
I would also suggest you get a free dynamic DNS name so you can address the remote Mac by a constant name instead of having to know the current IP address assigned to the home router, which the ISP can change anytime they want. No-IP.com or DynDNS.org offer free dynamic DNS names. You run one of their dynamic DNS updating clients on the remote Mac to keep the dynamic DNS name updated with the current ISP assigned IP address.
Once you have the port forwarding working, you connect for screen sharing using
Finder -> Go -> Connect to Server -> vnc://address.of.remote.mac
and for file sharing
Finder -> Go -> Connect to Server -> afp://address.of.remote.mac
If you are going to use ssh, scp, or sftp, then from an Applications -> utilities -> Terminal session you would do something like:
ssh [email protected]
scp local.file [email protected]:/path/where/to/put/the/file
scp [email protected]:/path/of/file/to/get /local/place/to/put/the/file
There are also sftp GUI clients you can use to make this part easier.
If you really cannot get this working, then consider using something like TeamViewer.com which deals with all the messy home router NAT navigation.

Client access action script

I have seen websites with client access text fields. i want
to create one, but do not have an example to work from. For
example, type in your access code "12345", submit, and be forwarded
to a site directory of the same name eg. mysite.com/12345/
Any help is greatly appreciated.

// create a dynamic text field with instance name
"accesscode_txt",
// a button symbol, with instance name "submit_btn"
var mysite:String = "
http://www.mysite.com/";
submit_btn.onPress = function(){
var clientsite:String = mysite + accesscode_txt + "/";
getURL(clientsite);
}

PAT with a single public IP and several servers behind firewall

Hi,
New to the ASA 5505 8.4 software version, but here is what I'm trying to do:
Single static public IP: 16.2.3.4
Need to PAT several ports to three separate servers behind firewall
One server houses email, pptp server, ftp server and web services: 10.1.20.91
One server houses drac management (port 445): 10.1.20.92
One server is the IP phone server using a range of ports: 10.1.20.156
Basically, need to PAT the ports associated with each server to the respective servers behind the ASA 5505.
Here is what I have. Is anything missing from this config? Do I need to include a global policy for PPTP and SMTP?
ASA Version 8.4(4)1
hostname kaa-pix
names
interface Ethernet0/0
switchport access vlan 2
interface Ethernet0/1
interface Ethernet0/2
interface Ethernet0/3
interface Ethernet0/4
interface Ethernet0/5
interface Ethernet0/6
interface Ethernet0/7
interface Vlan1
nameif inside
security-level 100
ip address 10.1.20.1 255.255.255.0
interface Vlan2
nameif outside
security-level 0
ip address 16.2.3.4 255.255.255.0
ftp mode passive
object network obj_any
subnet 0.0.0.0 0.0.0.0
object network server_smtp
host 10.1.20.91
object service Port_25
service tcp source eq smtp
object service Port_3389
service tcp source eq 3389
object service Port_1723
service tcp source eq pptp
object service Port_21
service tcp source eq ftp
object service Port_443
service tcp source eq https
object service Port_444
service tcp source eq 444
object network drac
host 10.1.20.92
object service Port_445
service tcp source eq 445
access-list acl-out extended permit icmp any any echo-reply
access-list acl-out extended permit icmp any any
access-list acl-out extended permit tcp any interface outside eq pptp
access-list acl-out extended permit tcp any object server_smtp eq smtp
access-list acl-out extended permit tcp any object server_smtp eq pptp
access-list acl-out extended permit tcp any object server_smtp eq 3389
access-list acl-out extended permit tcp any object server_smtp eq ftp
access-list acl-out extended permit tcp any object server_smtp eq https
access-list acl-out extended permit tcp any object server_smtp eq 444
access-list acl-out extended permit tcp any object drac eq 445
pager lines 24
logging asdm informational
mtu inside 1500
mtu outside 1500
icmp unreachable rate-limit 1 burst-size 1
no asdm history enable
arp timeout 14400
nat (inside,outside) source static server_smtp interface service Port_25 Port_25
nat (inside,outside) source static server_smtp interface service Port_3389 Port_
3389
nat (inside,outside) source static server_smtp interface service Port_1723 Port_
1723
nat (inside,outside) source static server_smtp interface service Port_21 Port_21
nat (inside,outside) source static server_smtp interface service Port_443 Port_4
43
nat (inside,outside) source static server_smtp interface service Port_444 Port_4
44
nat (inside,outside) source static drac interface service Port_445 Port_445
object network obj_any
nat (inside,outside) dynamic interface
route outside 0.0.0.0 0.0.0.0 16.2.3.1 1
timeout xlate 3:00:00
timeout pat-xlate 0:00:30
timeout conn 1:00:00 half-closed 0:10:00 udp 0:02:00 icmp 0:00:02
timeout sunrpc 0:10:00 h323 0:05:00 h225 1:00:00 mgcp 0:05:00 mgcp-pat 0:05:00
timeout sip 0:30:00 sip_media 0:02:00 sip-invite 0:03:00 sip-disconnect 0:02:00
timeout sip-provisional-media 0:02:00 uauth 0:05:00 absolute
timeout tcp-proxy-reassembly 0:01:00
timeout floating-conn 0:00:00
dynamic-access-policy-record DfltAccessPolicy
user-identity default-domain LOCAL
no snmp-server location
no snmp-server contact
snmp-server enable traps snmp authentication linkup linkdown coldstart warmstart
telnet timeout 5
ssh timeout 5
ssh key-exchange group dh-group1-sha1
console timeout 0
threat-detection basic-threat
threat-detection statistics access-list
no threat-detection statistics tcp-intercept
prompt hostname context
no call-home reporting anonymous

Thanks Lcambron...I got PPTP to work. Everything else works fine. I can access email, access my web server, FTP server, and PPTP server. However, from the above configuration, I cannot access my DRAC over the internet..The DRAC runs on a different internal server, and over port 445. So I have th following lines:
object network drac
host 10.1.20.92
object service Port_445
service tcp source eq 445
access-list acl-out extended permit tcp any object drac eq 445
nat (inside,outside) source static drac interface service Port_445 Port_445
Am I missing something here? Internally, i can telnet to port 445 on 10.1.20.92, so I know it is listening. However, externally, i cannot telnet to my external ip address of the ASA through port 445.
Thanks

Best practice for web servers behind a router (NAT, ACL, policy-map, VLAN)

Hi,
I'm a new Network admin, and I have some configuration questions about my installation (see attachment).
I have 3 web servers behind a router.
Public interface: 3 public ip adresses
Private interface: router on a stick config ( 3 sub-interfaces, 3 different networks, 3 VLAN)
I would to know the best way to redirect http traffic to the right server.
My idea is to map a public address to a private address, via NAT, but I'm not sure for the configuration. I could also redirect via Policy-map and filter by url content.
So if you have some advise for this case, it would be really appreciated.
Thank you.
Chris.

Hello Christophe,
As I understand you want 1st that ;
if somebody go to A.local.com from internet then he will redirect to 192.168.1.10 in your internal network.
That means, you need static mapping between your public @ip address and your local ip address.
for this example, your local interface is Fa0/0.1 and I dont your public interface because it is not mention in your diagram. I will suppose S0/0 for public interface.
that is the config for the Web Server1. You can do the same with the remaining servers:
interface fa0/0.1
ip nat inside
interface serial0/0
ip nat outside
ip nat inside source static 192.168.1.10 172.1.2.3
static mapping from local to public.
I suppose you have done the dns mapping in your network and the ISP have done the same in his network.
ip route 171.1.2.3 interface serial0/0
or
ip route 0.0.0.0 0.0.0.0 interface serial0/0.
After these step for each web server, you will get the mapping.
Now you can restrict access to this ip only to http or https protocol on your isp and after on your local network
like
ip access-list extended ACL_WebServer1
permit ip any 192.168.1.10 eq www
deny ip any 192.168.1.10
exit
interface fa0/0.1
ip acess-group ACL_WebServer1 in
no shut
exit
That is the first step.
Second step : you want to filter traffic by url, that means layer 5 to 7 filtering.
I am not sure that it is possible using cisco router with (ZBF + Regex).
Check the first step and let us know !
Please rate and mark as correct if it is the case.
Regards,

How to deal with credentials for external applications using a Java Client/

Hi Guys,
This is the case. I am integrating an external application with an ADF Application. I have implemented some programmatic ViewObjects that are being filled up by a REST Java Client Wrapper. Everything is working fine but the issue is that the credentials the wrapper is using are hard coded inside the java class. I am thinking to ask for the credentials at the beginning of my taskflow and then store them somewhere and use them then to create my client wrapper (passing them in the constructor).
However, I don't know if my approach is good and I would like you to share your experiences or how to deal with this.
Regards

You can use Credential Store Framework to store the credentials securely in the weblogic server instead of hardcoding in the java class.
The Credential Store Framework:
- enables you to manage credentials securely
- provides an API for storage, retrieval, and maintenance of credentials in different back-end repositories
Check the documentation on CSF API -
http://docs.oracle.com/cd/E29505_01/core.1111/e10043/devcsf.htm
Major Steps -
1. Create a credential map and key in em console to store the password (http://docs.oracle.com/cd/E25054_01/core.1111/e10043/csfadmin.htm)
2. Use CSF API to retrieve the stored password
3. In jazn-data.xml give permissions to access CSF key and map

NAT and Servers behind CSS 11501

All,
Please forgive my asking this question again. I was injured shortly after asking the last time and out of work for a long period of time.
My problem stems from needing to allow my web servers to initiate traffic to the outside world from behind our CSS boxes.
The web servers sit behind a pair of CSS 11501 content switches in Active-Passive ASR with fate sharing. We are only interested at this time with load balancing HTTP and HTTPS.
Everything works inbound no problem.
What I need to do is setup some type of NAT for my 3 web servers to initiate HTTP/HTTPS for patches, send SMTP from the web apps, and initiate HTTPS for credit card validation.
I have setup NAT on PIX units and routers no problem, but I seem to be unable to do it on these boxes. :(
In reality something as simple as a PAT translation on the outside of the CSS boxes should be sufficient.
Is this possible with our setup? Does anyone have some code examples?
Thanks in advance.
Addresses changed to protect the innocent:
Load Balancer 1:
!*************************** GLOBAL ***************************
bridge spanning-tree disabled
sntp server 1.1.1.41 version 1
snmp community noway read-only
snmp community noway read-write
app session 1.1.1.252
app
logging subsystem netman level info-6
dns primary 2.2.2.41
dns secondary 2.2.2.42
ip route 0.0.0.0 0.0.0.0 1.1.1.1 1
!************************* INTERFACE *************************
interface e1
phy 100Mbits-FD
description "Connect to Primary DMZ 1 3550 Switch"
interface e2
bridge vlan 2
phy 100Mbits-FD
description "Connected to Primary LB Server Switch"
interface e8
description "Inter Switch Communication (ISC) Port"
isc-port-one
!************************** CIRCUIT **************************
circuit VLAN1
description "DMZ 1 Subnet (1.1.1.x/24)"
ip address 1.1.1.251 255.255.255.0
ip virtual-router 1 priority 254 preempt
ip redundant-interface 1 1.1.1.250
ip redundant-vip 1 1.1.1.161
ip redundant-vip 1 1.1.1.162
ip redundant-vip 1 1.1.1.70
ip redundant-vip 1 1.1.1.71
ip redundant-vip 1 1.1.1.72
ip critical-service 1 upstream_downstream
circuit VLAN2
description "Load Balanced Servers Subnet"
ip address 2.2.2.2 255.255.255.0
ip virtual-router 2 priority 254 preempt
ip redundant-interface 2 2.2.2.1
ip critical-service 2 upstream_downstream
Various Services, Owners and Content
Load Balancer 2:
!*************************** GLOBAL ***************************
bridge spanning-tree disabled
sntp server 1.1.1.41 version 1
snmp community noway read-only
snmp community noway read-write
app session 1.1.1.251
app
logging subsystem netman level info-6
dns primary 2.2.2.41
dns secondary 2.2.2.42
ip route 0.0.0.0 0.0.0.0 1.1.1.1 1
!************************* INTERFACE *************************
interface e1
phy 100Mbits-FD
description "Connect to Secondary DMZ 1 3550 Switch"
interface e2
bridge vlan 2
phy 100Mbits-FD
description "Connected to Secondary LB Server Switch"
interface e8
description "Inter Switch Communication (ISC) Port"
isc-port-one
!************************** CIRCUIT **************************
circuit VLAN1
description "DMZ 1 Subnet (1.1.1.x/24)"
ip address 1.1.1.252 255.255.255.0
ip virtual-router 1
ip redundant-interface 1 1.1.1.250
ip redundant-vip 1 1.1.1.161
ip redundant-vip 1 1.1.1.162
ip redundant-vip 1 1.1.1.70
ip redundant-vip 1 1.1.1.71
ip redundant-vip 1 1.1.1.72
ip critical-service 1 upstream_downstream
circuit VLAN2
description "Load Balanced Servers Subnet"
ip address 2.2.2.3 255.255.255.0
ip virtual-router 2
ip redundant-interface 2 2.2.2.1
ip critical-service 2 upstream_downstream
Various Services, Owners and Content.

Gilles,
I added the following commands, and things seem to be working.
To circuit VLAN1
ip redundant-vip 1 1.1.1.80
!*************************** GROUP ***************************
group natout
vip address 1.1.1.80
add service nat_web_servers
active
service nat_web_servers
ip address 192.168.1.10 range 3
active
I do have a question about the above service commands.
I have 3 servers behind the CSS. Let's call them 192.168.1.10, 192.168.1.11 and 192.168.1.12. Am I correct in my thinking that adding range 3 then allows a match on all 3 of those servers and the CSS will then PAT these servers from the VIP address assigned to the group?
Otherwise, I think you have resolved this problem for us. Thank you.

Exchange Web Services are not currently available for this request because none of the Client Access Servers in the destination site could process the request.

Hi,
I am using EWS Java APIs and passing OAuth tokens to fetch data from office 365 mailboxes.
Because I am developing Web APIs I preferred using "Application Permissions" defined in Azure active directory application for Office 365, and used "client credential flow" OAuth flow to fetch OAuth token specific to application which will
allow "Have full access via EWS to all mailboxes in the organisation".
After fetching token with the procedure specified in the document "http://blogs.msdn.com/b/exchangedev/archive/2015/01/21/building-demon-or-service-apps-with-office-365-mail-calendar-and-contacts-apis-oauth2-client-credential-flow.aspx"
I passed this token to EWS Java APIs,
it gave me error saying:
microsoft.exchange.webservices.data.ServiceResponseException: Exchange Web Services are not currently available for this request because none of the Client Access Servers in the destination site could process the request.
I tried similar thing with EWS managed APIs for .net. Got similar error.
Can anyone provide some help and direction to resolve this error.
Thanks & Best Regards,
Pranjal

I see you found an answer with the X-AnchorMailbox header on StackOverflow:
http://stackoverflow.com/questions/29554724/exchange-web-services-are-not-currently-available-for-this-request-because-none

Client behind NAT

I have been searching for a solution for this issue with all that google knows......
I have my client behind NAT with ip 192.168.27.1
And the server behind NAT with some ip (i am not really worried abt this)
Now I register a client object to server for notification. SImply a hash table in server stores all my client objects. On a expected change, I invoke a method in my client objects.
In this scenario I happened to observer that the client objects sent to server had the client ip (192.168.27.1) inside it and not the NAT ip through which they went out.
So when I went invoke the remote method nothing interestingly happens as the client cannot be located.
I tried creating custom sockets in client and binding it to NAT ip --> obvious bind exception for a ip that is not with client
Setting the NAT ip as java.rmi.hostname in client --> no effect, since still the server is trying to notify (192.16827.1)
Help me to root out this issue. I feel that there must be a solution for this, otherwise RMI it would not have been this much successful.

Hi turing,
thanks for your reply
actually my question is
"maybe if you try using the "real" ip (www,whatismyip.com)
your program will work. "
how to do this in the scenario I explained.
Most of the discussions I saw in this forum are about server behind NAT and resolution approach for it. I can't find an answer for this even in the post you mentioned.
Simply,
When I register a client object in server, how will the server identify the client to notify, when the client is behind NAT.
Will the ip address that the remote object carries will also be NAT'ed. I don't see this happening.

Add Client access server with DR MBX to server clients from DR site.

Hello,
We have a medium size implementation at our company. It is as below,
- Two mail box servers (MBX1, MBX2) at production Site.
- One mail box server (MBXDR01 at DR site (Active))
- One DAG (name: IDKUDAG ) (MBX1,MBX2 and MBXDR01 are members)
- Two Client access servers at production site (with MBX1 and MBX2)
There is a high speed WAN connection between two sites.
What I need to do, I want to add additional CAS server at DR site as in case of production site maintenance or outage I want to migrate the DBs to DR MBX and the CAS server handle mail
client’s access.
Can I add a new server at the DR site with the same configuration as the production site???
Or there is another solution for this case.
Please advise.
Best regards,
Ahmed Salah
BR Ahmed Aboutabl

Hi Ahmed,
The CAS configuration for Exchange service in the second datacenter can be the same configuration as you mentioned. For example, the same internal&external namespaces for OWA, Autodiscover, EWS, OAB etc. in two datacenters.
Also make sure the certificate has included all needed namespace for the second site. For your reference, here is an article talked about the details of site resilients:
https://technet.microsoft.com/en-us/library/dd638129(v=exchg.150).aspx
http://www.msexchange.org/articles-tutorials/exchange-server-2010/high-availability-recovery/designing-site-resilient-exchange-2010-solution-part1.html
Note: Microsoft is providing this information as a convenience to you. The sites are not controlled by Microsoft. Microsoft cannot make any representations regarding the quality, safety, or suitability of any software or information found there. Please make
sure that you completely understand the risk before retrieving any suggestions from the above link.
Regards,
Please remember to mark the replies as answers if they help, and unmark the answers if they provide no help. If you have feedback for TechNet Support, contact [email protected]
Winnie Liang
TechNet Community Support

Exchange 2013 - Client Access Servers - Disk Space C:\ Running Low Space

Currently having an issue with both of our client access servers with exchange 2013.
The issue is that the C:\ drive will run low on disk space on one of the cas server and the other one will be fine. We have cleaned up all the logs etc to free space on the cas server that is running low on space but its not sufficient enough
as it keeps on running low on space, then what we do the free up the space is to reboot the cas server that runs low on space, and it free the space up after it has been restarted, then the other cas server starts running low on space.
has anyone seen this before?

Hi,
If it is possible, I suggest re-installing Exchange Server on another path instead of C:\.
Thanks
Mavis
Mavis Huang
TechNet Community Support

JNI to access functions of a dll to deal with a data aquision pci card

i am trying to do an applicattion to access the functions of a dll
to deal with a data aquision pci card (Analogig/Digital pci card).
i�m using jni of course, but i have some problems.
first i did a class called "TesteDll" witch contains all the native methods to access the functions of the pci card.
after, i did a "TesteDll.jar" to use these methods in another applicattions.
everything was fine until there. so created a single program like this:
TesteDll td = new TesteDll();
short res1 = td.P1602_DriverInit(wTotalBoards);
short r = td.P1602_ActiveBoard((short)0); // ativa a placa
short res2 = td.P1602_GetConfigAddressSpace((short)0,wAddrTimer,wAddrCtrl,wAddrDio,wAddrAdda);
int dwRepeat;
for(dwRepeat=0 ; dwRepeat<100 ; dwRepeat++)
short wRetVal;
wRetVal = td.P1602_ClearScan();
wRetVal += td.P1602_AddToScan((short)0,(short)0,(short)1,(short)0,(short)0,(short)0);     // CH:0 to scan
wRetVal += td.P1602_SaveScan((short)0,wV0);
wRetVal += td.P1602_AddToScan((short)2,(short)0,(short)1,(short)0,(short)0,(short)0);     // CH:2 to scan
/* ->error here!*/ wRetVal += td.P1602_SaveScan((short)1,wV2);
the programs compile correctly, but in the execution the folowing error was descripted:
An unexpected exception has been detected in native code outside the VM.
Unexpected Signal : EXCEPTION_ACCESS_VIOLATION (0xc0000005) occurred at PC=0x8079A50
Function=[Unknown.]
Library=(N/A)
NOTE: We are unable to locate the function name symbol for the error
just occurred. Please refer to release documentation for possible
reason and solutions.
Current Java thread:
at TesteDll.P1602_SaveScan(Native Method)
at Teste2.main(Teste2.java:119)
Dynamic libraries:
0x7CD80000 - 0x7CD9D000      C:\WINDOWS\SYSTEM\IMAGEHLP.DLL
Heap at VM Abort:
Heap
def new generation total 576K, used 161K [0x10010000, 0x100b0000, 0x104f0000)
eden space 512K, 31% used [0x10010000, 0x100385c0, 0x10090000)
from space 64K, 0% used [0x10090000, 0x10090000, 0x100a0000)
to space 64K, 0% used [0x100a0000, 0x100a0000, 0x100b0000)
tenured generation total 1408K, used 0K [0x104f0000, 0x10650000, 0x14010000)
the space 1408K, 0% used [0x104f0000, 0x104f0000, 0x104f0200, 0x10650000)
compacting perm gen total 4096K, used 1044K [0x14010000, 0x14410000, 0x18010000)
the space 4096K, 25% used [0x14010000, 0x141152e8, 0x14115400, 0x14410000)
Local Time = Wed Jun 01 10:21:14 2005
Elapsed Time = 0
# The exception above was detected in native code outside the VM
# Java VM: Java HotSpot(TM) Client VM (1.4.2_06-b03 mixed mode)
functions like "P1602_DriverInit(wTotalBoards)","P1602_ActiveBoard((short)0)" functions correctly.
i run the same program in c++ builder and everythings goes fine, but i need to do it in java. i don�t know what to do,can you help me?
thanks

i almost forgot, i�m using win 98.
First i used Dev C++ to make the JNI dll and the error ocurred at the first call of "P1602_AddToScan()" function. after i made the jni dll using C++ Builder, and the error ocurred at the second call of "P1602_SaveScan()" function. but the error message in both cases was the same.
for(dwRepeat=0 ; dwRepeat<100 ; dwRepeat++)
short wRetVal;
wRetVal = td.P1602_ClearScan();
/*First Error - Dev C++*/ wRetVal += td.P1602_AddToScan((short)0,(short)0,(short)1,(short)0,(short)0,(short)0); // CH:0 to scan
wRetVal += td.P1602_SaveScan((short)0,wV0);
wRetVal += td.P1602_AddToScan((short)2,(short)0,(short)1,(short)0,(short)0,(short)0); // CH:2 to scan
/* Second error C++ Builder */ wRetVal += td.P1602_SaveScan((short)1,wV2);
...

Dealing with client access to servers behind NAT

Similar Messages

Maybe you are looking for