<div id=”test1″><span id=”*.errors“>The information you entered is invalid. Try again.</span></div>

So in order to make them working we need to use escape character triple slash (\\\) before asterisk and period. Otherwise it will try to treat asterisk (*) as all elements and period (.) as class name. See below.

if($(‘#\\\*\\\.errors’).length > 0){
        var displayvalue=$(‘#\\\*\\\.errors’).html();
        $(‘#test1′).html(“”);
        alert(displayvalue);
    }

Many of the bugs are unique to the older Android devices (like 2.2, 2.3), therefore with the recent news that Google Chrome is to become the standard browser on Android 4.0 and above means that, jQuery mobile will be more consistent and will offer gr8 performance across devices. Currently, mobile chrome is available through Android Market in Beta.

One interesting fact, was when I ran CSS3/HTML5 compatibility test on Android default browser I got approx 188 points whereas when ran on Chrome I got approx 300 points which means quite a lot improvement and support for new features introduces with CSS3/HTML5.

Link: http://www.google.com/intl/en/chrome/android/ (Supported on Android 4.0+)

Couple of Known Bugs

Transition is one of main feature which comes with jQuery Mobile. Which works very smooth on iOS devices but on Android devices they behave differently, very sluggish and not quite smooth as on iOS.

Secondary problem is very noticeable blinking on the target page, once the transition completed. This is a known bug, and although there are some touted css fixes around that certainly reduce the flickering they seem to result in other problems. Certainly, the blinking problem was still clearly noticeable on Android 2.2 and 2.3.3 even with these CSS based fixes.

The blinking wasn’t so bad in iOS, however was still evident in certain circumstances, therefore we decided to turn off the transitions across all platforms for the time being.

As this is a known bug we look forward to this being fixed in a future jQuery Mobile release.

Bug Id: https://github.com/jquery/jquery-mobile/issues/3217 (seems fixed in 1.1 release)

CSS suggested fix for flickering:

.ui-page {

    -webkit-backface-visibility: hidden;

}

Another problem with the jQuery mobile and touch events is the time it takes to action a click event, this delay was most noticeable in Android, where it seemed tapping on a link would take an age to trigger, in some cases up to 3 seconds.

The crux of the problem is that the click event does not fire until a delay has passed, as the browser is allowing the chance for a second touch to make it a double click or for it to become a swipe gesture and so on.

The solution to this is to execute a conditional statement to test for native touch event support and then execute a ‘touchStart’ event (the mobile equivalent of mousedown) when this event is natively available. The ‘touchStart’ event fires immediately, just like mousedown. This results in a more responsive device interaction. The change significantly reduced the delay on iOS but less so on android, however there was still a marked improvement.

Reference: http://stackoverflow.com/questions/6235794/jquery-mobile-for-every-live-tap-event-should-there-be-an-equivalent-click-even

Later I digged inside the services code and found that service was not implemented, that’s why it was throwing this default exception.

You need to replace the  throw instruction (which is the default functionality for the case in which not implementation is provided before deploying the service) by your original functionality

User should be able to login to other server without password using command  “ssh -l username remotehost”.

For this I used SSH public key based authentication as below:-

For this to work you need a file named id_dsa.pub in your ~/.ssh Directory. If it’s not there, simply create it with below command:

ssh-keygen -t dsa

above command will create below directory with public and private keys:

.ssh/id_dsa
.ssh/id_dsa.pub

now copy id_dsa.pub file to remote server .ssh/ directory as “authorized_keys2” with whom you would like to do an automated login.

 After that try below command from source server, no need for password

ssh -l username remotehost

 For more details can view link

java.io.NotSerializableException: com.xxxx.xxx.xxx.thunderhead.Project
    at java.io.ObjectOutputStream.writeObject0(ObjectOutputStream.java:1156)
    at java.io.ObjectOutputStream.writeObject(ObjectOutputStream.java:326)
    at java.util.ArrayList.writeObject(ArrayList.java:570)

Later I realized that I forget to implement Serializable interface in my bean. That fixed issue.

java.security.NoSuchAlgorithmException: Cannot find any provider supporting RSA/NONE/OAEPPADDING
at javax.crypto.Cipher.getInstance(DashoA12275)
at org.apache.ws.security.util.WSSecurityUtil.getCipherInstance(WSSecurityUtil.java:703)
at org.apache.ws.security.processor.EncryptedKeyProcessor.handleEncryptedKey(EncryptedKeyProcessor.java:145)
at org.apache.ws.security.processor.EncryptedKeyProcessor.handleEncryptedKey(EncryptedKeyProcessor.java:114)

This appears to BouncyCastle is missing from class path.  Or Bouncycastle jar may still in the classpath – but it’s not picked as a crypto provider.

Try below solution:

Download the Bouncycastle jar corresponding to your JDK from here and copy it to [JAVA_HOME]\jre\lib\ext\

Set following in [JAVA_HOME]\jre\lib\security\java.security under;

#
# List of providers and their preference orders (see above):
#
security.provider.X=org.bouncycastle.jce.provider.BouncyCastleProviderOR

In code add Bouncycastle as a provider.

import java.security.Security;  
import org.bouncycastle.jce.provider.BouncyCastleProvider;  
 
Security.addProvider(new BouncyCastleProvider());

Caused by: java.security.InvalidKeyException: Illegal key size or default parameters
    at javax.crypto.Cipher.a(DashoA13*..) ~[na:1.6]
    at javax.crypto.Cipher.a(DashoA13*..) ~[na:1.6]
    at javax.crypto.Cipher.a(DashoA13*..) ~[na:1.6]
    at javax.crypto.Cipher.init(DashoA13*..) ~[na:1.6]
    at javax.crypto.Cipher.init(DashoA13*..) ~[na:1.6]


There are key size restrictions with the default crypto files local_policy.jar and US_export_policy.jar comes with JDK – which limits it to 128. If your security policy using a key size larger than this – then the above exception is thrown.

For example – if your security policy specifies the algorithmic suite as Basic256 – then the key size to be used is 256.

For the solution of above issue, you need to patch your JDK with Java Cryptography Extension (JCE) Unlimited Strength Jurisdiction Policy Files.

For JDK1.5 visit, download the crypto files and copy the two jar files from the extracted jce directory (local_policy.jar and US_export_policy.jar) to $JAVA_HOME/jre/lib/security.

For JDK1.6 visit

If your IDE using it’s own specific JDK then patch that as well with these files to resolve the issue.

org.apache.axis2.AxisFault: Error in signature with X509Token
    at org.apache.rampart.handler.RampartSender.invoke(RampartSender.java:70)
    at org.apache.axis2.engine.Phase.invoke(Phase.java:318)
    at org.apache.axis2.engine.AxisEngine.invoke(AxisEngine.java:254)
    at org.apache.axis2.engine.AxisEngine.send(AxisEngine.java:419)
    at org.apache.axis2.description.OutInAxisOperationClient.send(OutInAxisOperation.java:402)
    at org.apache.axis2.description.OutInAxisOperationClient.executeImpl(OutInAxisOperation.java:229)
    at org.apache.axis2.client.OperationClient.execute(OperationClient.java:165)
 Caused by: org.apache.rampart.RampartException: Error in signature with X509Token
    at org.apache.rampart.builder.AsymmetricBindingBuilder.doSignature(AsymmetricBindingBuilder.java:746)
    at org.apache.rampart.builder.AsymmetricBindingBuilder.doSignBeforeEncrypt(AsymmetricBindingBuilder.java:419)
    at org.apache.rampart.builder.AsymmetricBindingBuilder.build(AsymmetricBindingBuilder.java:95)
    at org.apache.rampart.MessageBuilder.build(MessageBuilder.java:147)
    at org.apache.rampart.handler.RampartSender.invoke(RampartSender.java:64)
    … 102 more

Caused by: org.apache.ws.security.WSSecurityException: Signature creation failed; nested exception is:
    java.security.UnrecoverableKeyException: Cannot recover key
    at org.apache.ws.security.message.WSSecSignature.computeSignature(WSSecSignature.java:721)
    at org.apache.rampart.builder.AsymmetricBindingBuilder.doSignature(AsymmetricBindingBuilder.java:737)
    … 106 more
Caused by: java.security.UnrecoverableKeyException: Cannot recover key
    at sun.security.provider.KeyProtector.recover(KeyProtector.java:311)
    at sun.security.provider.JavaKeyStore.engineGetKey(JavaKeyStore.java:121)
    at sun.security.provider.JavaKeyStore$JKS.engineGetKey(JavaKeyStore.java:38)
    at java.security.KeyStore.getKey(KeyStore.java:763)
    at org.apache.ws.security.components.crypto.CryptoBase.getPrivateKey(CryptoBase.java:216)
    at org.apache.ws.security.message.WSSecSignature.computeSignature(WSSecSignature.java:713)
    … 107 more

Above issue is caused by CallbackHandler, because Incorrect password is provided for the private key. Correct the password you are good to go.

Later debug I found that I duplicated the Controller definition stuff inside XXX-Servlet.xml as below.

<context:component-scan base-package="org.me.spring.web" />

and

<bean  class="org.me.spring.web.SampleController" >

The first line will auto-discover and register a SampleController for you. By explicitly declaring another one, you get a duplicate, and the url-mapping will clash.

You need to either remove the <context:component-scan>, or remove the explicit bean definitions for the controller and the DAO. If you do the latter, then you’ll need to use autowiring to inject their dependencies.

Below command will find all the logs with given name pattern inside the directory including sub directories and then WC will count it’s lines and display on screen in total.

wc -l `find . -name usage*.log`

Sample output of command

     4 ./2011-08-08/usage.log
     5 ./2011-08-09/usage.log
     5 ./2011-08-10/usage.log
     4 ./2011-08-11/usage.log
     2 ./2011-08-12/usage.log
     4 ./2011-08-13/usage.log
   24 total

Below command will not count the blank lines.

grep -cv ‘^\s*$’ `find . -name “usag*.log”`

Sample Output, see below blank lines are not counted.

./2011-08-08/usage.log:3
./2011-08-09/usage.log:4
./2011-08-10/usage.log:4
./2011-08-11/usage.log:3
./2011-08-12/usage.log:1
./2011-08-13/usage.log:3