GL Framebuffer Completeness & Blitting Issues

Hi,
I have the following piece of test code that I made after encountering a blitting issue on the application I was working on.
There are 3 issues that are illustrated by this test code:
1) glCheckFramebufferStatus gives reversed status on the read and draw buffer. If the buffers are incomplete, it returns 0x8CDB for READ and 0x8CDC for DRAW which is flipped according to the definition in the header file.
2) I am unable to make the FBO complete unless I attach a color attachment, even though I called glDrawBuffer(GL_NONE), glReadBuffer(GL_NONE), which according to my understanding, is supposed to do away with this requirement for completeness.
3) On my ATI Radeon HD 5770 Mac Pro, even if I use the dummy attachment, the blitting fails on a D24S8 but succeeds D32. "Fail" here means failing a source and destination comparison verification check. I have done this check using bytes, ints and floats and the results are consistent. When I tested this on my MacBook Pro with a Nvidia 560m, there were no errors for either D32 or D24S8.
Is there anyone out there who can shed some light on this?
Are these bugs or did I miss something obvious? I will like to do depth-buffer-only blitting without using a dummy attachment if possible.
If these are known issues, what is the recommended solution?
In this test code, I am trying to blit a depth-only FBO to another depth-only FBO. Depending on the parameters passed in, the test can be performed with D24S8 or D32, and with/without a dummy color attachment.
To ensure that OpenGL is initialized properly, the rest of the test application displays a colored rotating cube when run.
- (void) RunTest_DepthBufferBlit:(bool)isPackedStencil useDummyAttachment:(bool)useDummy
    int err = glGetError(); //Clears out previous errors
    NSLog(@"DepthBufferBlit %@ %@", isPackedStencil ? @"DepthStencil" : @"Depth", useDummy ? @"With Dummy Color Attachment" : @"");
     #define GL_FRAMEBUFFER_COMPLETE                        0x8CD5
     #define GL_FRAMEBUFFER_INCOMPLETE_ATTACHMENT           0x8CD6
     #define GL_FRAMEBUFFER_INCOMPLETE_MISSING_ATTACHMENT   0x8CD7
     #define GL_FRAMEBUFFER_INCOMPLETE_DRAW_BUFFER          0x8CDB
     #define GL_FRAMEBUFFER_INCOMPLETE_READ_BUFFER          0x8CDC
     #define GL_FRAMEBUFFER_UNSUPPORTED                     0x8CDD
    int width = 256;
    int height= 256;
    // Generates 2 framebuffers, one as READ_FRAMEBUFFER and the other as DRAW_FRAMEBFFER.
    GLuint framebuffers[2] = {0, 0};
    glGenFramebuffers(2, framebuffers);
    err = glGetError();
    if (err != 0)
        NSLog(@"DepthBufferBlit - Gen FBO Fail: %d", err);
    GLuint texName[3] = { 0, 0, 0 };
    // Create source, destination and dummy buffer
         glGenTextures (3, texName);
        err = glGetError();
        if (err != 0)
            NSLog(@"DepthBufferBlit - Gen Textures Fail: %d", err);
        unsigned char* buffer = (unsigned char*)malloc(4 * width * height);
        memset(buffer, 0, 4 * width * height);
        float* depthBuffer = (float*)malloc(4 * width * height);
        for (int i = 0; i < width*height; i++)
            if (i%2 == 0)
                depthBuffer[i] = 2.0f;
            if (i%5 == 0)
                depthBuffer[i] = 5.0f;
        for (int i = 0; i < 3; i++)
            glBindTexture (GL_TEXTURE_RECTANGLE, texName[i]);
            err = glGetError();
            if (err != 0)
                NSLog(@"DepthBufferBlit - Bind Texture %d Fail: %d", texName[i], err);
            glTexParameteri(GL_TEXTURE_RECTANGLE, GL_TEXTURE_MAG_FILTER, GL_LINEAR);
            glTexParameteri(GL_TEXTURE_RECTANGLE, GL_TEXTURE_MIN_FILTER, GL_LINEAR);
            GLint internalFormat = isPackedStencil ? GL_DEPTH_STENCIL : GL_DEPTH_COMPONENT32;
            if (i == 0)
                glTexImage2D(GL_TEXTURE_RECTANGLE, 0, internalFormat, width, height, 0, GL_DEPTH_COMPONENT, GL_FLOAT, depthBuffer);
            else if (i == 2)
                glTexImage2D(GL_TEXTURE_RECTANGLE, 0, GL_RGBA       , width, height, 0, GL_RGBA, GL_UNSIGNED_BYTE, buffer);
            else
                glTexImage2D(GL_TEXTURE_RECTANGLE, 0, internalFormat, width, height, 0, GL_DEPTH_COMPONENT, GL_FLOAT, buffer);
            err = glGetError();
            if (err != 0)
                NSLog(@"DepthBufferBlit - Create Texture %d Fail: %d", texName[i], err);
        free(buffer);
        free(depthBuffer);
    GLint attachment = isPackedStencil ? GL_DEPTH_STENCIL_ATTACHMENT : GL_DEPTH_ATTACHMENT;
    glBindFramebuffer(GL_READ_FRAMEBUFFER, framebuffers[0]);
    if (useDummy)
        glFramebufferTexture2D(GL_READ_FRAMEBUFFER, GL_COLOR_ATTACHMENT0, GL_TEXTURE_RECTANGLE, texName[2], 0);
    glFramebufferTexture2D(GL_READ_FRAMEBUFFER, attachment, GL_TEXTURE_RECTANGLE, texName[0], 0);
    glBindFramebuffer(GL_DRAW_FRAMEBUFFER, framebuffers[1]);
    if (useDummy)
        glFramebufferTexture2D(GL_DRAW_FRAMEBUFFER, GL_COLOR_ATTACHMENT0, GL_TEXTURE_RECTANGLE, texName[2], 0);
    glFramebufferTexture2D(GL_DRAW_FRAMEBUFFER, attachment, GL_TEXTURE_RECTANGLE, texName[1], 0);
    glReadBuffer(GL_NONE);
    glDrawBuffer(GL_NONE);
    err = glGetError();
    if (err != 0)
        NSLog(@"DepthBufferBlit - glFramebufferTexture2D: %d", err);
    int dfbo = glCheckFramebufferStatus(GL_READ_FRAMEBUFFER); //Apple Bug: Flipped status         
    int rfbo = glCheckFramebufferStatus(GL_DRAW_FRAMEBUFFER); //Apple Bug: Flipped status
    if (dfbo != GL_FRAMEBUFFER_COMPLETE || rfbo != GL_FRAMEBUFFER_COMPLETE)
        NSLog(@"DepthBufferBlit FBO {%i,%i} Fail - Read = %x, Write = %x\n", width, height, rfbo, dfbo);
        // Restore the framebuffer binding.
        glBindFramebuffer(GL_FRAMEBUFFER, 0);
        // Deletes the 2 framebuffers.
        glDeleteFramebuffers(2, framebuffers);
        // Deletes the 3 textures
        glDeleteTextures(3, texName);
        return;
    if (isPackedStencil)
        glBlitFramebuffer(0, 0, width, height,
                          0, 0, width, height, GL_DEPTH_BUFFER_BIT | GL_STENCIL_BUFFER_BIT, GL_NEAREST);
    else {
        glBlitFramebuffer(0, 0, width, height,
                          0, 0, width, height, GL_DEPTH_BUFFER_BIT, GL_NEAREST);               
    err = glGetError();
    if (err != 0)
        NSLog(@"DepthBufferBlit - Blitting: %d", err);
    int errCount = 0;
    // Check texture contents
    // 256 x 256 = 65536
    float tex[65536] = { 0 };
    float tex2[65536] = { 0 };
    glBindFramebuffer(GL_FRAMEBUFFER, framebuffers[0]);
    glReadPixels(0, 0, width, height, GL_DEPTH_COMPONENT, GL_FLOAT, tex);
    glBindFramebuffer(GL_FRAMEBUFFER, framebuffers[1]);
    glReadPixels(0, 0, width, height, GL_DEPTH_COMPONENT, GL_FLOAT, tex2);
    for (int i = 0; i < 65536; i++)
        float val = abs(tex[i] - tex2[i]);
        if (val > 0.000001)
            errCount++;
    if (errCount > 0)
        NSLog(@"DepthBufferBlit - Differences = %d", errCount);
    else {
        NSLog(@"DepthBufferBlit %@ Completed Successfully.", isPackedStencil ? @"DepthStencil" : @"Depth");
    // Restore the framebuffer binding.
    glBindFramebuffer(GL_FRAMEBUFFER, 0);
    // Deletes the 2 framebuffers.
    glDeleteFramebuffers(2, framebuffers);
    // Deletes the 3 textures
    glDeleteTextures(3, texName);
    return;
- (void) RunSingleShotTests
    /* glBlitFrameBuffer */
    [self RunTest_DepthBufferBlit:NO  useDummyAttachment:NO];NSLog(@" ");
    [self RunTest_DepthBufferBlit:NO  useDummyAttachment:YES];NSLog(@" ");
    [self RunTest_DepthBufferBlit:YES useDummyAttachment:NO];NSLog(@" ");
    [self RunTest_DepthBufferBlit:YES useDummyAttachment:YES];
Results after running the above sample on 10.7.3 with both CoreProfile and LegacyProfile (with a few minor changes to make it GL 2.1 compatible):
2012-04-27 15:52:51.871 CocoaGLCoreProfile[56222:403] DepthBufferBlit Depth (D32)
2012-04-27 15:52:51.874 CocoaGLCoreProfile[56222:403] DepthBufferBlit FBO {256,256} Fail - Read = 8cdc, Write = 8cdb
2012-04-27 15:52:51.874 CocoaGLCoreProfile[56222:403] 
2012-04-27 15:52:51.875 CocoaGLCoreProfile[56222:403] DepthBufferBlit Depth (D32) With Dummy Color Attachment
2012-04-27 15:52:51.886 CocoaGLCoreProfile[56222:403] DepthBufferBlit Depth Completed Successfully.
2012-04-27 15:52:51.887 CocoaGLCoreProfile[56222:403] 
2012-04-27 15:52:51.887 CocoaGLCoreProfile[56222:403] DepthBufferBlit DepthStencil (D24S8)
2012-04-27 15:52:51.888 CocoaGLCoreProfile[56222:403] DepthBufferBlit FBO {256,256} Fail - Read = 8cdc, Write = 8cdb
2012-04-27 15:52:51.889 CocoaGLCoreProfile[56222:403] 
2012-04-27 15:52:51.889 CocoaGLCoreProfile[56222:403] DepthBufferBlit DepthStencil (D24S8) With Dummy Color Attachment
2012-04-27 15:52:51.892 CocoaGLCoreProfile[56222:403] DepthBufferBlit - Differences = 9832
Thanks in advance.

Hi,
This was actually the wrong forum so I reposted this quesion and got my answer in the developer forums.
It turned out you have to call:
glDrawBuffer(GL_NONE)
glReadBuffer(GL_NONE)
on BOTH the source and destination buffers. i.e Both buffers have to be read and draw complete.
This was why I got a seemingly flipped status on the buffers when I checked them.
Cheers.
YJ

Similar Messages

  • LR5 - Photos in 2nd Display are blurry / don't render completely - intermittent issue

    Just upgraded to Lightroom 5, and the performance issues I saw in 4 appear to be gone - which is awesome. I've come up on an issue that I've seen referenced in other posts, but only for LR4. Also, the issue seems to be slightly different than mine. My issue is that when I have the 2nd display running in Loupe mode, the image won't render completely and looks blurry/fuzzy, when in Develop mode on the first screen. However, it's not every time. Sometimes it works fine. Sometimes it takes a few seconds, and then it works. Sometimes, it doesn't ever work. It just sits in a blurry state - not fully rendered. Sometimes if I go to the next photo, and then back, it will then load. Sometimes that doesn't work.
    While I did have tons of performance issues in LR4 (including 4.3 and 4.4), this was not something I remember seeing.
    I'm running Windows 7 64-bit, i7 2600-K, 24gb RAM, multiple SSDs.
    Any help would be appreciated.

    I'm having the exact same problem in LR5
    Selecting another picture and coming back usually fixes the problem.
    Windows 7 64-bit, Core I7 2600-K, 32GB RAM, NVidia Geforce GTX 580 - driver version 320.49 WHQL

  • "Writing XMP Metadata Did Not Complete Successfully" Issue

    Hi all,
    I've been using Lightroom for a bit now. I'm on V2.5 now.  I just did a new system build and so reinstalled Win Vista 32.  My photos and XMP files are on a seperate external drive and I saved my catalog file (plus did a backup).  When I reinstall Lightroom and did the updates I then copied the old catalog 2 file to the right location.  All my pics were there with edits no problem.  I've done this several times now when I have had to reinstall the OS for various reasons. Never had an isuse
    This time I went back to change some status data on a photo that I took before the reinstall.  When I went to update the metadata I get a message that says "Writing XMP Metadata Did Not Complete Successfully".  I looked in Bridge and the data did not update.  This happens for all my RAW photos that were taken before the reinstall.  I can make changes fine but cannot get the XMP to write.
    Bridge will not make changes either. I tried changing the star rating and color label and it does nothing.
    For files that I have imported after the reinstall it all works fine.
    It is like the old XMPs are locked or the programs can't find them.  I can see them in the file manager so they are all there.
    Any ideas on the problem and how to fix it?

    I don't think they are locked.  I think Vista is just blocking Lightroom.  I went back this morning and deleted a few of the xmp's as you suggested and Lightroom worked fine and wrote the files.  Then I thought about the potential lock issue and tried opening Lightroom "As an administrator".   When I did that everything worked fine.  For some reason Vista won't let lightroom open the older files unless it sees it as the admin.  All works fine now.  Thanks!!
    Glenn

  • Technical Completion  Related Issues

    Dear All,
    As per SDN Library, if by mistake you have done the Technical completion of the order, and after that you want to do the goods issue..........it is possible.........but i m trying on this.........it is not happening......
    can anybody help me on this...........
    Thanks

    hi vinay,
    as per system status if the order is technically completed you cannot able to issue the material against the maintennace order, whereas you try to post through confirmation.
    example
    create an order and assign a material and technically complete the order
    go to T code iw41 enter the order no , in the parameter select all materials allowed, confirm the activites,then press the goods movement tab page and save
    here it will show that the confirmationa and goods movement were posted
    hence the post goods issue in system status TECO will allow goods movement while confirmation, whereas while doing MB1A it will forbid.
    so may be it is internally programmed to allow the goods movement in confirmation whereas it will not allow during goods issue to the order
    regards
    thyagarajan

  • Complete Album Issue

    Though old songs I have bought still show up, any new song I buy does not count towards completing my album. They show up as not purchased. I want to complete the album, but when I go to buy it, it tells me that I already own some of the songs and can either buy it with downloaded duplicates or cancel. Why will it not let me complete the album? I don't want to waste the money on buying duplicates. Is complete the album not applicable anymore? Or is something else wrong here? This is on iTunes 9 by the way.

    I am having the exact same issue when trying to complete the album Dimanche à Bamako by Amadou & Mariam. Tried it first a few days ago, and I'm still getting the error message today.

  • Adobe Connect Completion Status Issues

    A course within a curriculum does not report completion. I am forced to override their completion. Why is this course not accurately recording completion? Also, why are the users only enrolled in the curriculum and not enrolled into the individual courses? Is this affecting the completion status?

    As to why it is not reporting correctly, you may need to work with Adobe Support to have them look at the course from the back end and identify where the issue may be.
    For the enrollment, the courses are part of the curriculum, and are all tracked through the curriculum, so it makes sense that the studens are not enrolled in the individual courses. If you look closely, you'll notice that even the links to the courses are different in the curriculum than they are as the indvidual courses.

  • Captivate 4: 3 Attempts, Score and Completion Status Issue

    Hi Everyone,
    I hope somoene may have some insight on this issue.  We built a quiz in Captivate that has a minimum passing score of 80% and allows the learner 3 attempts to complete it.  The quiz should only send a completed status to the LMS if an 80% or greater is acheived.  If the learner takes the quiz and passes on the 1st attempt, the quiz is marked as completed and passes the correct score to the LMS.  However, if the learner fails the first time, and then gets a passing score on a 2nd or 3rd attempt, the new score is passed to the LMS but the status remains as incomplete. I included a screenshot from a test I ran in SCORM Test track.  I failed the first attempt and passed on my second attempt with a 97%.  However, as you see it is stated as incomplete.  When I passed on the first attempt, it says completed.  Any thoughts?  Perhaps this is a setting somewhere that is incorrect?  We have tried several things and even tried calling Adobe Support (but after an hour and a half of holding and transfers that hit a wall).  Any insight would be great!!
    Thanks,
    Connor

    Complete/Incomplete is more related to how many of the slides in the project were viewed all the way through to the end.
    If you're talking about a score from the Quiz being at least 80% in order to pass then maybe you should be focusing more on Pass/Fail.
    Check your Quiz Settings in Captivate.  Do you have it set to report Complete/Incomplete or Pass/Fail?
    Maybe what you're seeing here as inconsistency is due to your not watching every single slide all the way through on the second time through.

  • Check point not complete - Production issue

    Hi
    There is a error in my alert log that checkpoint not complete .(Oracle Database 10g Release 10.2.0.4.0)
    and when i checked
    select name, checkpoint_change#, to_char(checkpoint_time, 'DD.MM.YYYY HH24:MI:SS') from v$datafile_header ;
    it shows
    NAME
    CHECKPOINT_CHANGE# TO_CHAR(CHECKPOINT_
    /u01/app/oracle/oradata/CMSPROD/datafile/o1_mf_system_52ocv5tz_.dbf
    384596791 *03.11.2011 10:57:21*
    please help me as soon as possible
    Thanks

    Oracle_2410 wrote:
    Hi
    There is a error in my alert log that checkpoint not complete .(Oracle Database 10g Release 10.2.0.4.0)
    and when i checked
    select name, checkpoint_change#, to_char(checkpoint_time, 'DD.MM.YYYY HH24:MI:SS') from v$datafile_header ;
    it shows
    NAME
    CHECKPOINT_CHANGE# TO_CHAR(CHECKPOINT_
    /u01/app/oracle/oradata/CMSPROD/datafile/o1_mf_system_52ocv5tz_.dbf
    384596791 *03.11.2011 10:57:21*
    please help me as soon as possible For as soon as possible , please raise a severity 1 ticket with Oracle support. That said, for the error, the normal suspect is the constant switching of the log files. For that, check the log file size of yours and also play with the parameter archive_lag_target to control it.
    HTH
    Aman....

  • X60 complete restore issue

    I am trying to do a complete restore for my friend's X60 notebook.  I am used to using notebooks (specifically apple notebooks) that have CD drives so that its an easy restore/reformat using the disk utility or windows disks partition manager.  my friend doesnt need anything special, just the ability to use MS word and the internet so i dont care about keeping the built in IBM software.
    anyways, im trying to use the built in recovery tool that IBM provides to do a factory restore and im getting two errors:
     The directory Name is invalid (and so i click OK)
    Product recover failed during the convention phase (again clicking ok)
    this brings me back to the rescue and recovery page.
    any suggestions would be appreciated even help with getting around the recovery mode and booting from a usb drive or anything of that nature.
    thank you!

    Yes if you have windows vista recovery disks or installation disks i would go with a USB jump drive. I also just did a clean install using this method, the directions on this page is exactly how i made the drive bootable http://maximumpcguides.com/windows-vista/create-a-bootable-usb-drive-to-install-windows-vista/
    All you have to do is copy the contents of the cds/dvds onto this drive. Then you also need to go into BIOS and enable the options to let you boot from USB  ( sorry i dont have a x61 only a x200 not specific instructions here)

  • EP 60 SP2 P4 install completed startup issue

    I have successfully installed EP 60 onSolaris...but when I attempt to startup it just quits at the following service load: com.sap.portal.usermanagement|security
    Loading services:
      Loading service: com.sap.portal.license.runtime|license
      Loading service: com.sap.portal.pcd.aclservice|aclconnector
      Loading service: com.sap.portal.pcd.basicrolefactory|factory
      Loading service: com.sap.portal.pcd.basictransport|transport
      Loading service: com.sap.portal.pcd.configservice|configuration
      Loading service: com.sap.portal.pcd.glservice|generic_layer
      Loading service: com.sap.portal.pcd.lockservice|locks
      Loading service: com.sap.portal.pcd.plconnectionservice|connections
      Loading service: com.sap.portal.pcd.softcacheservice|softcache
      Loading service: com.sap.portal.pcd.textrepository|TextRepositoryService
      Loading service: com.sap.portal.pcd.traceservice|traces
      Loading service: com.sap.portal.pcd.umwrapperservice|umwrapper
      Loading service: com.sap.portal.pcm.admin.apiservice|default
      Loading service: com.sap.portal.runtime.config|config
      Loading service: com.sap.portal.runtime.config|config_deployment
      Loading service: com.sap.portal.runtime.system.authentication|authentication
      Loading service: com.sap.portal.runtime.system.clusterinformation|clusterinformation
      Loading service: com.sap.portal.runtime.system.notification|notification
      Loading service: com.sap.portal.runtime.system.notification|SAPJ2EEHttpService
      Loading service: com.sap.portal.runtime.system.repository|application_repository
      Loading service: com.sap.portal.umeregistration|ume_registration
      Loading service: com.sap.portal.usermanagement|usermanagement
      Loading service: com.sap.portal.usermanagement|user_management_engine
      Loading service: com.sap.portal.usermanagement|security
    Any ideas why it would not load this service?
    John Ryan

    Hi John,
    one more question, which JDK version did you use? and should I need to apply hotfix for 9I? such as Patch 3 for 9.2.0.4 etc.
    You mean this JDBC bug ?
    "# 3175296:  2827119:  RDBMS     :  DEDICATED SERVER DOWN WHEN JDBC APPLICATION QUERY LONG SQL STATEMENT
    Thanks
    Ben
    Message was edited by: Ben Jiang
    Message was edited by: Ben Jiang

  • Issue with BDC of ME22N to change gross price of service.

    Hi,
    I have a requirement where I have to undelete the PO > services tab of the item details > change the gross price to the value thats calculated by the program  for each of the services of the PO > set the deletion indicator to the PO.
    The spec says to use a BDC of ME22N to acheive this but after creating the recording I have had several issues with some PO's that interrupt the BDC because they bejave differently from the recording I did.
    On most cases when im running the BDC in foreground after changing the gross price and pressing enter I get a new windowthat appears asking for the account assingment of service, on this window the G/L account information is ussualy passed automatically then the bdc clicks back and the process continues normally. On other services after changing the gross price and pressing enter I get the account assignment of service window but the G/L information is not passed and when the bdc clicks back the G/L account information gets passed instead of going back and the BDC stops, if I manually click back again the process will continue.
    With single service PO's the process is completing, this issue is happening with PO's that have multiple services.
    I know ME22 should be used for BDC and not ME22N, but the functional insists to correct the issues with the BDC of ME22N. Also I atempted to use BAPI_PO_CHANGE but I dont think this bapi can not update the gross price of each of the services, it can change the net price but this is not my requirement.
    Is there a bapi that can change the gross price of each of the services of the PO?
    Please advice me on this.
    Edited by: bodyboarder1 on Dec 2, 2010 3:33 PM

    Hi
    If you really need a BDC program, try to simulate ME22  instead of ME22N
    Max

  • Consignment issue

    Consignment issue:
    Are there is any possibility in standard SAP to do the complete consignment issue process (VA01, VL01N, VF01) in background with single transaction. Or is it possible to develop this through “Z” Transaction.

    Hi,
    Yes its possible to create a Sales Order, Delivery and a Billing Doc and it can be acheived through output types of all these documents with medium 6 (EDI).  You need to create an output type with this medium and in the routine for this output type you can call for relevant BAPIs like Sales Order Create etc.
    Thanks
    Krishna.

  • How to automate the WM/shipping process after the consignment issue ?

    I am looking for the best way to automate the shipping process after the consignment issue orders has been created. since Warehouse/shipping process will not consist of any actual picking and shipment but we are still required to go through the use of Vl01N and VL02N for Delivery Note and Good issue... steps......I want to know if there is any best practice is to automate the call to VL01N,VLO2N for all the completed consignment issue orders that SAP suggest or someone has done some work in this space.....and can share their views.....
    one situation to explain the needs....
    Create the consignment issue orders via EDI....and once they are there ...we need a automatic process to find these order and create the delivery note , goods issue documents.........looking forward to hear from experts who can inform  if there is a standard SAP process( automatic)  out there or we need to do some custom development....to achieve this....thanks.....

    Thanks for the insight....we are thinking to have a new custom table added which will have all the order that needs to batch processed and if we have any errors we can re process them from the control table ( new custom table)...please let us knwo if there is any other way to do this....or sap already has some tables....whihc we can use...to automate....as a batch process.....

  • Onedrive (Skydrive) Sync issues through TMG

    We are having a ton of issues when using Onedrive (skydrive) to sync document libraries with our SharePoint 2013 server behind Forefront TMG. Users will randomly stop syncing (with no errors on skydrive), or files will just sit there trying to upload. I
    checked the logs on forefront, and see the following error occurring quite often for the document libraries the users are trying to sync.
     12210 An Internet Server API (ISAPI) filter has finished
    handling the request. Contact your system administrator.
    We have an array of TMG servers that use FBA for authentication on the front end, then delegate using NTLM to the sharepoint server. The array is also setup as a reverse proxy, and is the only way the sharepoint server can reach the internet. We are using
    the external host name (portal.domain.com) from TMG's internal adapter to the sharepoint server, so it's not a mapping issue. It's also happening on images and files that are less than 100KB, so it's not a size issue either.
    I have tried disabling the malware filter completely, but the error still occurred. This is happening to multiple users running both windows 7/8.

    I have no way to test this as our sharepoint server is in a remote datacenter, with no access except through TMG.
    I did however get rid of the error by changing the authentication delegation method from NTLM to Basic. Since making the change, I have not seen any instances of the 12210 error for any users through TMG. So far I have no gotten any complaints/issues since
    making the change.
    So it looks like TMG cannot properly handle using NTLM to delegate the credentials. This is the second time I have run into an issue where this was the cause, and changing to Basic was the fix (Completely separate issue with TMG/IIS 8 using Ajax. Changing
    to basic was the solution given by MS support)... It's pretty dumb that a Microsoft security product can't properly handle windows authentication on the back end....

  • Aperture 3 Vault - Updating Vault issue

    I've upgraded to Aperture 3.
    I created a new Vault.
    I started a backup to the new vault 14 hours ago and it's still chugging away with the progress bar at about the 50% mark with a status of "Updating Vault". It hasn't made any progress since it stopped copying the masters about 2 hours ago.
    My library is around 1.5TB and 175K files.
    This is the third time I have tried backing up to a new vault and it always hangs at this point. I'm currently using Superduper to make a copy of my library so I have a current backup of some kind.
    Is anyone else experiencing this?

    You guys aren't backing up these vaults to an external Western Digital hard drive are you? If so, the problem could be that the drives are formatted in fat-32. Aperture 3 doesn't support fat-32 drives. Aperture 2 didn't support them either but was, apparently, more tolerant about them. Aperture 3 flat out doesn't work with them.
    Any external drives used with Aperture 3 MUST be formatted in Mac extended format.
    FWIW, I updated the vault on my Mac Pro for the first time (since installing Aperture 3) this morning. It completed without issues and took the normal amount of time for a brand new vault creation.
    BTW, I choose to leave my master images in their current location and reference them instead of copying them to my Aperture Library. I believe a vault update of an Aperture Library that is managed (all masters copied to the library) instead of referenced (left in their original location) will take longer and make for a much larger vault backup.
    Mark

Maybe you are looking for

  • SerialOracleClob usage in weblogic server 8.1

    Hi, An application is deployed successfully on weblogic 7.0 server but there was a ClassCastException for SerialOracleClob class when the same was deployed on 8.1 server.Also when I tried to compile the source code using weblogic.jar of 8.1,i got the

  • New beta features: share to Twitter, Facebook, & Pinterest

    We've added a few beta sharing features for you to try out.  Now, you can share to Facebook, Twitter, and Pinterest directly from theme details. We have created a special link to give you access to new features (new feature link)--without this link y

  • Frm-47313 Invalid Query for Hierarchy tree

    Hello. i am trying to create a hierarchy tree with the following query: SELECT 1, level, n.name, null, hn.nde_id FROM cerpt_nodes n, cerpt_hierarchy_nodes hn WHERE n.id = hn.nde_id AND hn.hir_id = 1 CONNECT BY PRIOR hn.nde_id = hn.nde_id_parent_of ST

  • Captivate 5.5 and Audition 5.5 not linking

    I have Captivate 5.5 installed and have downloaded and started using the trial version of Audition . I am unable to get the 2 programs to see each other. In captivate when I edit audio. mu edit with Audition button is still grayed out. I have tried w

  • Inserts on External Tables

    Please I just want to confirm that it is impossible to do insert on external tables. I kept getting the error while doing some practise with them ORA-30657: operation not supported on external organized table regards,