A very very strange issue on BlazeDS Deserializing

Hi all,
I run into a very very strange issue on BlazeDS Deserializing. Is there any hint or idea?  Thanks a lot!
Scenario 1:
Both the browser(firefox or IE) and the midtier (BlazeDS ,java) are on the same machine.
Actual result:
Everything works fine.  All the properties in the Flex Object could be mapped to the Java Class
Scenario 2:
The browser(firefox or IE) and the midtier (BlazeDS ,java) are on the different machine.
Actual result:
Sometimes it works fine.
And most of the time it does not work fine. Only the data in List is mapped correctly. All the values of simple data type such as int are mapped to 0 .
Let me simplify the problem.
The objects of Class B is always stored in the ArrayCollection  assocItems of Class A
Flex side:
public class A {
public var id: int;
public var  name: String;
public var assocItems: ArrayCollection=new ArrayCollection();
public class B {
public var id: int;
public var  assocItems : ArrayCollection=new ArrayCollection();
The id , name in Flex Class A and assocItems in Class B could be  deserialize to correct value .  But the id of Class B is always mapped to 0. This is not correct.
The log level of BlazeDS is configured to Debug. From the log I can see that  each value is passed from Flex side to Java server side correctly.
[BlazeDS][INFO] Channel endpoint my-amf received request.
[BlazeDS][DEBUG] Deserializing AMF/HTTP request
Version: 3
(Message #0 targetURI=null, responseURI=/18)
(Array #0)
[0] = (Typed Object #0 'flex.messaging.messages.RemotingMessage')
However, the Java object does not get the correct value of B.id.
In conclusion,
1. It’s an issue with only default types
2. It does not occur every time
3. It does not occur when the browser(firefox or IE) and the midtier (BlazeDS ,java) are on the same machine
Have you ever run into the similar problems? And what’s your solution?

I checked, when I hard-code the constant in linq, the generated sql is like this:
SELECT "Extent1"."AA" AS "AA"
FROM (SELECT "TBL"."AA" AS "AA"
FROM "USERNAME"."TBL" "TBL") "Extent1"
WHERE ('1062303212007000121' = "Extent1"."AA")
if I pass a variable to the query with same value, the sql is like this:
SELECT "Extent1"."AA" AS "AA"
FROM (SELECT "TBL"."AA" AS "AA"
FROM "USERNAME"."TBL" "TBL") "Extent1"
WHERE ("Extent1"."AA" = '1062303212007000121' /* @p__linq__0 */) (@p__linq__0 is passed as parameter)
The only difference is a paramter. I can't figure out why there is such a impact on performance.

Similar Messages

  • Files transferred to external HD end up in trash - very strange issue

    Hi,
    I seem to be having a very strange issue, hopefully somebody can help or has seen this before.
    Hardware: MacBook Air, Western Digital Passport external HD, 250gb, formatted NTFS.
    Software: OSX 10.7 Lion, Paragon (allows me to write to an NTFS formatted HD)
    Issue Summary: I copy something to the external HD; I empty trash; the file I copied is deleted from the external HD. It doesn't happen every time.
    I do also seem to have a persistent folder in trash called "WD Sync Data" that won't go away unless I re-delete it as it's also in the root of the drive. Not sure if this is related somehow.
    Recently I copied 3 video files inside a folder to the HD, then disconnected the HD. When I reconnected it a while later and noticed that inside the "WD Sync Data" folder in trash were the 3 files I'd copied. I then looked inside the same folder on the root of the drive and the files were there, instead of inside the folder that I'd copied to the drive, which had vanished. If I hadn't noticed and had emptied the trash, those files would've been deleted.
    It's definitely happened more than once; last time I actually deleted the files from my computer after copying, then emptied trash, and they were gone from the external HD also.
    I've only found one person who seemed to be experiencing a similar problem back in 2010. It sounds similar though it could've been user error: http://www.mac-forums.com/forums/apple-notebooks/188224-weird-external-hard-driv e-trash-problem-help-please.html
    Has anyone experienced this or have any suggestions?

    Quick update: I ran a verify then a repair in Disk Utility as there were a couple of errors. Hopefully this has solved the issue, however it'd still be interesting to know if anyone's come across this before and what caused it?

  • Premiere Pro CS5.5 -very strange issue with the color tones and Preview-

    I deal with a very strange and not usual problem in Premiere Pro CS5.5 (also the version CS5). What I try to do is just a simple color correction. The problem occurs when I try to use Three Way Color Corrector but it happens with Fast Color Corrector also.
    When applying the Three-Way Color Correction and without changing anything else (just applying), can be seen in the preview of the soft how sudden changes in color tones. For example, the gray tone becomes a little more green, but also can be seen very easy a change in the other color tones too. In fact even RGB Parade show the same thing. This is very, very strange and unacceptable and saw that it happens when trying to use Colorista II too.
    I tried the same thing in Sony Vegas Pro 11 and I downloaded the trial version of Edius 6 to convince me if the same thing happens. Not! In Sony Vegas and Edius everything is working correctly, I have not experienced these problems in the preview's of this 2 others softs.
    This issue is very annoying because every possible color correction is based on what we see in the preview and if in this preview the tones are seen altered realize that can not make a decent color correction. So what happens here?!? I uploaded a small video demonstration to see very clearly about what I'm talking. You can see in the next link: http://vimeo.com/35317616

    Yes Jim, you're right. No changes in the scopes when one just apply RGB Curves, Fast Color Corrector, RGB Color Corrector  or the Three way Color Corrector (last three thanks to the new update 6.03). I already wrote that Adobe team has been able to fix this bug for some of the tools/effects.
    But Brightness&Contrast? Change Color? Color Balance? Levels? MBLooks? Twixtor? Colorista II? Does it seem normally that all scopes or just one of them jump when you just apply one of these effects/tools? Not to say that in some situations even the tones change (especially in shadows&midtones).
    You say this is something "very minimal". I say it is not so. I don't find normally to make a very detalied CC with Three way Color Corrector and then when applying Colorista/MBLooks, for example, the scopes jump and inclusively some tones may be influenced. And I left clear examples in this topic where you can see that this is possible. I don't know if this is "very minimal" or not, but is really so impossible to edit in Premiere without this "very minimal" issue?!? How can it be possible in other softs (Edius, Media Composer, Vegas Pro) and in Premiere not?!?
    One more thing: many of us use intermediate codecs in the editing process. Do you find normally that a AVCHD video have the lumninance at approximately 110IRE, to export it directly to Cineform and then import it in Premiere, and the waveform shows the luminance at 100IRE?!? How is this possible?!? So the AVCHD video have a 110IRE luminance, you convert this video to Cineform without changing anything, and then this Cineform video have the luminance at 100IRE?!? And anyone can verify this example.
    So, how can one trust in this soft?!?

  • Can't delete folder (or files in it) - VERY strange issue

    ok i know there's been tons of questions about how to delete/force delete a troublesome file or folder. believe me, i searched, i've tried them all, and still, i couldn't solve it.
    here's my problem:
    a) 4 files, in a folder called "as".
    - 06Jacksons Club "Sunspot".mp3
    - 11_Novelette Conclusion" excerpt_Lisa edit.mp3
    - 16_Green Tea At Denny_s _Mary P. _Aka High Places Singer, Bassoonist, Zine-Maker, BOO - The "P" Is For Pearson_].mp3
    - 20_"Arrival In Benaras__ From Merchant-Ivory_s Film __The Guru__.mp3
    b) things i tried so far, and what happened:
    - using the terminal with sudo/rm. answers for a file: "No such file or directory".
    - using the terminal with rm/sudo to delete the whole folder:
    rm: /Volumes/Untitled/as/: is a directory ... is the answer i get, and it stays there, doesnt move
    c) if i move it into trash, when i try emptying it, everything goes except for that folder and it's files inside, neither does force deleting it via the terminal make it disappear. the "crushing" noise is audible, but nothing happens. no errors or anything, it just stays there.
    d) if i click on one of the files, they disappear (!!!). if i play around in finder and come back to that folder later, they're there again. the folder itself can be normally copied, renamed, moved around on the harddisk and everything. if i duplicate the folder, the dublicate is empty, unlike the original folder.
    e) NO music player or anything else ever used those files.
    f) they are 0 bytes each.
    g) another note: the files and the folder are on an ext. harddrive. the harddrive is not faulty.
    h) i even tried deleting them from a windows computer, which keeps saying the filenames are too long or invalid, and it wont work either.
    i) the folder or the files are not locked, i have full read/write permission, i tried reinstalling several times, i also tried it from safe mode.
    j) formatting the harddisk is NOT an option.
    i would be very helpful if anyone could help me with this very annoying issue. i just want to see my trashcan empty for once.
    thanks very very much!

    Try booting to Safe Mode (hold down the shift key while booting, restart may take a long time but let it go), sometimes just doing that works. Restart normally.
    Reading
    [You can't empty the Trash or move a file to the Trash in Mac OS X|http://docs.info.apple.com/article.html?artnum=106272]
    [Solving Trash Problems|http://www.thexlab.com/faqs/trash.html]
    [How To Fix Stubborn Trash and Why it Won't Delete|http://www.osxfaq.com/Tutorials/LearningCenter/HowTo/Trash/index.ws]
    *Utilities for emptying trash*
    [Trash It! v3.6.1|http://www.nonamescriptware.com/?page_id=5]
    [DropNuke|http://www.versiontracker.com/dyn/moreinfo/macosx/12664]
    [Super Empty Trash 2.3|http://www.macupdate.com/info.php/id/16727] - Simple AppleScript applet that will empty the Trash with administrator privileges
    [What's Keeping Me|http://www.hamsoftengineering.com/products/wkm/wkm.html] - lists files open and applications that are using them

  • Very very strange issue: no more activation possible (iTunes 8.2.1 on a G3)

    At first, i came back to iTunes 8.2.1 because using 9.1 is no fun on a G3 PowerBook (the loudness adjustment makes the computer unusable for hours, then the vents roar and the machine crashes!).
    OK, because i still wanted to remote the G3 with my iPod touch, i got iTunes 8.2.1 wich runs very good on such an old computer. I am pretty shure that i already had this version on this computer, because i made almost every update... I use this computer just for listening Music! This Computer was activated all the time and with all Versions of iTunes - from 4.0 to 9.0 !!!!
    What happens now is very VERY strange and i am loosing my mind. The apple support can't help me (they also resetted all my activations).
    It sounds simple, but i am no more able to activate this iTunes!!! Everything is typed well, but no chance. It says everytime that the name or password is wrong. I also can't activate Genius!
    I promise, tried everything: my last attempt was to make a complete Reset of the machine (i erased the whole disk and installed everything(!!!!!) from scratch!)
    But with no luck - this machine will not let me f*/("&§ng connect to my account were i payed lots of euros for my music!!!!!!!
    This is in fact the strangest thing that i ever gone through on computers (and i've seen lots of strange things, believe me!)
    I hope that somehone could give me a hint...

    I actually found the window hiding at the bottom of my screen, being covered by my dock, and was able to drag it up.
    You can also go into System Preferences, then dock and then move your dock to the left or right to find the window.

  • Very odd Airport issue airport has a mind of its own...

    Ok so hopefully I have gotten your attention. As of 3 days ago my Macbook Pro (which has always connected to the internet through my Airport Extreme with the signal boosted by an Airport Express) so I am online reading email when I attempt to open a new page and my internet dies. I look at my other computer (Macbook not pro) sitting next to me and it is connected to the network just fine. So I go to connect to the internet again and I can see my network and yet every time I try to connect I get a message saying "error couldnt connect" or something similar to that. Then I went to make sure my settings werent changed and they weren't. So I ran a diagnostic to make sure that nothing was wrong settings-wise. And when I get to the step where I need to choose the network I want to connect to it asks for my password (been using same one for years) and I input it and it tells me the password is incorrect. Then I try it again, and again im told the password is incorrect, no caps are on im spelling it right, basically I am faultless. Then after a number of restarts and frustrated hours, out of nowhere it connects again.
    So then two nights ago I am online I pick up my laptop from my bed and walk to the room next door and start to use my computer in the next room to show a friend something. I walk back into my room with the laptop and 8 minutes later the same internet issue occurs again. I try everything, the password isnt accepted anymore and then a couple of hours later it works.
    So then tonight I am writing a paper and the same stupid thing happens again. This time after 3 restarts I get back online but it is still very very frustrating and I have no clue what is wrong
    If anyone has any advice PLEASE im all ears

    Welcome to the discussions!
    The fact that things improved a bit with Interference Robustness is probably a major clue that you are experiencing wireless interference issues at your new location.
    The most common cause of interference problems in a home are cordless phones, which typically operate at the same frequency as wireless networks. You need to power off the phones and their base stations / cradles for a few hours to see if you notice any improvement in the performance of your wireless network. If you do, you need to try different positions for the phones.
    Please post back on your findings to take the next possible steps in identifying the probable interference issue there.

  • A very urgent deployment issue about DBAdapter

    Hello All,
    I have a very urgent deployment issue about DBAdapter.
    That DBAdapter is connect to DB2 AS400 Database. I have a developing database (jdbc:as400://server01/TEST) and a production database (jdbc:as400://server01/PROD).
    During developing, I used DBAdapter wizard to create it, and import some tables, and set the Adapter to use jabc/DB2DS as connection information for easily deployment later.
    Then I deploy to Production. I configured Data-source.xml and oc4j-ra.xml rightly; I set DB connection point to production database. But the DBAdapter still write into developing Database.
    I checked the DBAdapter, the imported tables are something like this, TEST.table1, TEST.table2. And there are a lot "TEST" located in DB2Writer_toplink_mapping.xml, DB2Writer.xml, TEST.schema, DB2Writer.table1.ClassDescriptor.xml.
    This TEST is refrer to the TEST in connect String jdbc:as400://server01/TEST.
    I think this might be the reason cause the problem. As to production database, "TEST" should replaced by "PROD". If I changed it manually, I have to change every time when switch between TEST and PROD. And I also don't know if it is safe to do it? (I tried, and bring some toplink mapping problem)
    By the way, for Oracle Database, because we use 2 instances for testing and production with same schema name, and do not have this issue.
    Anyone could help and many thanks.
    Kerr
    Message was edited by:
    Kerr

    Hi Kerr,
    The idea is to set up all connections in the BPEL or ESB services with logical names, e.g. typically of the form eis/DB/MyFinancialSystem or eis/DB/MyLogisticsSystem. This way, you do not have to modify code when deploying it onto different environments that serve different purposes.
    When moving your services through their lifecyle, on every environment you deploy these to you will have the same logical connections configured on each instance, e.g. for DEV, QA, SIT, UAT and PROD. Only, in case of QA the actual physical connection is configured to point to the QA instance of the systems that your services interact with whereas in case of UAT it points to the UAT instance of the same system.
    Maybe your problem is caused by connecting as user "SomeUser" when running the DB Adapter wizard during development and actually selecting objects from a different schema than you used to connect with, e.g. "Test" in your case.
    Hth,
    Sjoerd

  • Very urgent Decimal issue in ADF page

    Hello friends,
    i have an issue in ADF page which was weight field, issue has below.
    1) weight filed was validated that without entering 5 digit value in filed, example suppose i enter 12345 in weight it is showing that in small dialog box -weight value should be 0 to 9999.999 only. but it is let me to go next field and submit button. it is happening only in create page only.
    2)but when i go to detail page , same weight fild is populated there - i enter invalid weight something 12345, it is showing that in small dialog box -weight value should be 0 to 9999.999 only. but here let me not go to another filed until i ENTER valid value.
    create page weight filed need to work same as detail page.i am new to ADF development, how to check where this validation done and how to fix this.
    very very urgent.
    Thanks,
    vamshi.

    If you use ADF BC, open the entity and choose "Business Rules". You can then add validation rules such as range limits under the attributes. It is usually best practice to include validation rules in the entity (model layer), this way they apply automatically to all view objects based on the entity and you are guaranteed to have only valid values in the database. (View layer validation may have its place as a convenience supplement.)
    Read more about business rules in the developer's guide: http://download.oracle.com/docs/cd/E12839_01/web.1111/b31974/bcvalidation.htm

  • I have a 4s runing 6.1.1 every 5minutes it promp me network lost and i have very bad battery issues lit draine so fast what should i do

    I have a 4s runing 6.1.1 every 5minutes it promp me network lost and i have very bad battery issues lit draine so fast what should i do?

    What have you figured out? Anything? What you show just shows a pretty quiet system except for firefox having a bunch of tabs open. You've got some page ins and outs so you maybe your'e hitting your HD with swap files more than what you think?
    I think something like coconut battery app or something like it will tell you the charge cycles on the battery. How many do you have?
    Lastly, are you using any extensions like ad blocker, or click to flash?

  • Since upgrading to 7.0.1; when opening a new tab, system is very, very slow - not sure where to look to fix the issues. Prior, I would have tabs pop open.

    I really do not have a lot more details to add to the question; in opening more than one tab; system is very, very slow either from trying to open when on-line or trying to open a link embedded in an email. Did not have an issue before the upgrade.

    Problem solved, I think, though not internally to Firefox and not for earlier Windows versions: Windows 7 preserves former version of files and folders. right click on the filen ame or foldername and select "restore previous version". At this juncture, I have copied the previous and current versions of the folder so I can try them out and if necessary move the sessionstore file from one to another.

  • Qosmio F50 very dim display issue

    Hi everyone
    Recently my Qosmio F50-10z has started suffering a very weird problem. Often the screen goes absolutely dim. I can see a very very faint image if I concentrate very hard. External monitors when connected stay well lit and bright. The most weird bit is that when this happens I have to literally slam the lid shut and open the lid immediately to make the screen light up again.
    I know it is either the inverter or the screen cable. My question is how do I find out whether it is the VGA cable or the inverter and where do I buy the part from!
    Please help!
    Thanks!

    > Often the screen goes absolutely dim. I can see a very very faint image if I concentrate very hard. External monitors when connected stay well lit and bright.
    > My question is how do I find out whether it is the VGA cable or the inverter and where do I buy the part from!
    In my opinion its an FL inverter issue since the picture on the external monitor is ok.
    The FL inverter controls the display backlight and the issue sounds like FL inverter malfunction.
    You can purchase and order all compatible parts from an local ASP.
    The authorized service partner would be helpful and could also replace the part.
    I would recommend this because notebook disassembling is very tricky.
    Greets

  • Very Interesting Text Issue...

    Hello Friends,
    Would love it if someone had some insight into this strange
    issue.
    First, look at the file without pre-loader:
    http://www.clonepress.com/cp2.swf
    Notice text on the right under "Tshirts" / "Posters" loads
    just fine.
    Now, check it out with the preloader:
    http://www.clonepress.com
    The text disappears! Spooky...
    Any ideas as to what's going on?
    Thanks! and Be Well,
    Graham
    MediaMackenzie.com

    Hey Raymond
    Thx for the response. I am using a mask in the preloader; I
    didn't see any other way to do it... But why would I need to embed
    the fonts in cp2.swf? They work fine (as you saw) when not being
    loaded through index.swf. Did you mean I should embed the fonts in
    index.swf? Do you know another way to use a preloader that doesn't
    involve using a mask? Cause I think that is the issue...
    Thanks
    Graham

  • DVSeries Audio issue - Very Very low voice.

    I got my DV6 laptop 2 days back.
    I am not able to hear voice from any good orginal DVD to the full.
    Problem: Voice from speaker is very very low.
    I updated my drivers. I reinstalled everything with factory settings. I am trying hard but hardly able to hear anything out of it.

    ok, this is actually driving me MAD now!
    @Rudegar, thanx for the reply but i don t see how this relates to my question 

  • Strange issue with POF: byte array with the value 94

    This is a somewhat strange issue we’ve managed to reduce to this test case. We’ve also seen similar issues with chars and shorts as well. It’s only a problem if the byte value inside the byte array is equal to 94! A value of 93, 95, etc, seems to be ok.
    Given the below class, the byte values both in the array and the single byte value are wrong when deserializing. The value inside the byte array isn’t what we put in (get [75] instead of [94]) and the single byte value is null (not 114).
    Pof object code:
    package com.test;
    import java.io.IOException;
    import java.util.Arrays;
    import com.tangosol.io.pof.PofReader;
    import com.tangosol.io.pof.PofWriter;
    import com.tangosol.io.pof.PortableObject;
    public class PofObject1 implements PortableObject {
         private byte[] byteArray;
         private byte byteValue;
         public void setValues() {
              byteArray = new byte[] {94};
              byteValue = 114;
         @Override
         public void readExternal(PofReader reader) throws IOException {
              Object byteArray = reader.readObjectArray(0, null);
              Object byteValue = reader.readObject(1);
              System.out.println(Arrays.toString((Object[])byteArray));
              System.out.println(byteValue);
              if (byteValue == null) throw new IOException("byteValue is null!");
         @Override
         public void writeExternal(PofWriter writer) throws IOException {
              writer.writeObject(0, byteArray);
              writer.writeObject(1, byteValue);
    Using writer.writeObjectArray(0, byteArray); instead of writer.writeObject(0, byteArray); doesn't help. In this case byteArray would be of type Object[] (as accessed through reflection).
    This is simply put in to a distributed cache and then fetched back. No EPs, listeners or stuff like that involved:
         public static void main(String... args) throws Exception {
              NamedCache cache = CacheFactory.getCache("my-cache");
              PofObject1 o = new PofObject1();
              o.setValues();
              cache.put("key1", o);
              cache.get("key1");
    Only tried it with Coherecne 3.7.1.3.
    Cache config file:
    <?xml version="1.0" encoding="UTF-8"?>
    <!DOCTYPE cache-config SYSTEM "cache-config.dtd">
    <cache-config>
         <caching-scheme-mapping>
              <cache-mapping>
                   <cache-name>my-cache</cache-name>
                   <scheme-name>my-cache</scheme-name>
              </cache-mapping>
         </caching-scheme-mapping>
         <caching-schemes>
              <distributed-scheme>
                   <scheme-name>my-cache</scheme-name>
                   <service-name>my-cache</service-name>
                   <serializer>
                        <class-name>
                             com.tangosol.io.pof.ConfigurablePofContext
                        </class-name>
                        <init-params>
                             <init-param>
                                  <param-type>string</param-type>
                                  <param-value>pof-config.xml</param-value>
                             </init-param>
                        </init-params>
                   </serializer>
                   <lease-granularity>thread</lease-granularity>
                   <thread-count>10</thread-count>
                   <backing-map-scheme>
                        <local-scheme>
                        </local-scheme>
                   </backing-map-scheme>
                   <autostart>true</autostart>
              </distributed-scheme>
         </caching-schemes>
    </cache-config>
    POF config file:
    <?xml version="1.0"?>
    <pof-config xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
         xmlns="http://xmlns.oracle.com/coherence/coherence-pof-config"
         xsi:schemaLocation="http://xmlns.oracle.com/coherence/coherence-pof-config coherence-pof-config.xsd">
         <user-type-list>
              <!-- coherence POF user types -->
              <include>coherence-pof-config.xml</include>
              <user-type>
                   <type-id>1460</type-id>
                   <class-name>com.test.PofObject1</class-name>
              </user-type>
         </user-type-list>
    </pof-config>

    Hi,
    POF uses certain byte values as an optimization to represent well known values of certain Object types - e.g. boolean True and False, some very small numbers, null etc... When you do read/write Object instead of using the correct method I suspect POF gets confused over the type and value that the field should be.
    There are a number of cases where POF does not know what the type is - Numbers would be one of these, for example if I stored a long of value 10 on deserialization POF would not know if that was an int, long double etc... so you have to use the correct method to get it back. Collections are another - If you serialize a Set all POF knows is that you have serialized some sort of Collection so unless you are specific when deserializing you will get back a List.
    JK

  • Strange issue with WEBI schedule Report

    Hi All,
    I have a strange issue with schedule WEBI reports, a schedule report runs for Hours , i reschedule it again by deleting the instance and the schedule report never got succeed even after running for hours.
    Any idea???
    regards

    Hi Manoj,
    1. Try to put some filters in the report and then run the report. if it takes less time, then probably your query is fetching very    large data.
    2. MDX query error may be the reason , this error come when a query runs endless, please try running the query 2-3 times, or when load on server is less.
    Hope this will help,
    Anamika.

Maybe you are looking for