Publishing Bug
I have set up several shows that can be successfully rehearsed, recorded, and reviewed with VC 3. However, longer shows (3-5 minutes) freeze during publishing. I have used pack and go and tested the show on three different computers. I have even uninstalled and reinstalled VC 3. The problem will not go away and is predictable. The bug happens when a video clip avi or wmv is inserted into the show. In fact the freeze h happens when such clips are triggered. Is there a way to bypass VC 3s publisher? Can I extract the files and use a different program to publish? Does anyone know how to fix this bug?
This sounds like a problem. I just purchased VC3 upgrade.
I have been using an early version of Visual Communicator Plus. I have read about this problem a few other times on this forum. Only being able to publish three minutes of video, before it freezes, don't sound good. I'm curious, it sounds like a processor problem. What kind of processors are you using?
Similar Messages
-
We have observed a publishing bug which results in .htm files of 0KB on multiple machines publishing different projects.
Often, Cap4 simply does not complete the publishing operation and just hangs w/o ever show the "View Ouput" option and green checkmark.
Reboots don't fix it, but if I keep at it, eventually I can get a project to publish after multiple reboots. I typically use Flash 8 but I have the bug with 8, 9, or 10.
We'll keep recording this bug in our logs, but we have only ever observed it in Cap4.
thanks,
JonI believe I spoke too soon. After further testing, I encountered the problem again and am unable to figure out the root cause of the problem. The problem may be due to the webkit-tap-highlight being added to the div tag. I am still seeing the problem when nesting text inside a square inside edge animate. However, I don't see it consistently and I have no idea where the webkit-tap-highlight is originating from.
Has anyone else encountered this issue when publishing in Edge Animate? -
I have no objection to the images I upload to Flickr being accessbile to public searches.
Yet every time I publish or republish a photo, Flickr's setting ... "It is hidden from public searches" seems to get set for each image - even if I manually override for an individual image using the Flickr Web interface and then republish from LR it gets set back to this state - see below...
I have set my Flickr user profile with ...
so I don't think this behaviour is being caused by Flickr...
and I cannot see any settings at the LR Flickr Publishing settings which would cause this ...
Any help in either correcting my lack of understanding or confirming it a bug would be appreciated.
Many thanks
Alan.I can confirm the bug in LR 3.0, it happens every time I republish.
There is at least one other annoying problem caused by republishing. If you have posted the photo into a group thread (by copying and pasting the HTML), the link from the thread back to the photo gets broken when you republish. It shows up as "This photo is currently unavailable".
I haven't downloaded LR 3.2RC yet to see if these issues have been fixed. -
Lightroom 3 - Flickr Publish Bug
I am experiencing an issue where when I publish images to Flickr they immediately show up as "Modified Photos to Re-Publish" and will not ever move to the "Published" section even though I have not changed a thing.
Anyone have any ideas as to what I am missing?
Thanks!Although not related to the original post – “Stuck in republish”, which I think is now fixed; here is my take on the situation you are describing -
I suspect there is a bug in the LR software which is still present in LR 3.4RC.
I believe this bug manifests itself in two ways when an attempt is made to republish an existing published photo …
1) If you are privately publishing and attempt to republish an existing photo, then LR sends a parameter (“Hidden” from public searches) incorrectly and out of bounds in accordance with the Flickr API, and Flickr quite rightly rejects it. I suspect LR is sending values of 1 or 2 (Hidden or not) for this “hidden” parameter when it should be sending 0 or 1 to the set safety level function.
2) Probably related to the same technical reason as point 1 - if a non-private photo is republished then instead of the photo continuing to be made available in public searches it is now erroneously made hidden from public searches when it reaches Flickr – again 1 or 2 instead of 0 or 1. This is somewhat annoying if you go through a re tagging exercise for the benefit of public searching only to find that they are now hidden from searches once re-published.
I have seen no confirmation from any other user of point 2 above despite posting the question here ... http://forums.adobe.com/thread/796350
Further points …
Since the Lightroom Software Development Kit contains a sample of the Lua code which is used for publishing, I have pointed to the actual lines of code I suspect are in error in this thread which also contains some technical dialog with Matt Dawson a Community Professional …
http://forums.adobe.com/thread/821148?tstart=0
I have raised a bug report with this technical detail, and raised a bug report for each of the two cases I mentioned above. Hopefully this may result in some action, but I would suggest that if this important to you that you also raise a bug report, as there appears to be, to some an extent, a principle of - they who shout loudest or most often will get priority.
As a side note … to cure the problem for the moment, although I don’t want to use third party software long term, I have adopted a Flickr publishing plugin by Friedl for my private publishing until this is fixed. If you Google “Friedl Flickr” you will find it. I have not yet decided to adopt this plugin for my public publish due to concern about the possibility of losing all comments and favourite tagging by other Flickr members – I simply do not know at this point.
It is also interesting that the change log for this Friedl plugin on his web site also describes being tripped up by the same peculiarity present in the Flickr API which probably gave rise to this mistake by Adobe. The Flickr API may have its peculiarities, but it is well defined and this is a Lightroom bug in my opinion.
Alan. -
Hello to all Contribute experts and fanatics alike!
Seems like a lot of folks are having issues with publishing
using Contribute.
As the connection to the web server seems to have a bug, and
from experience
I always have my connection getting cut off, and having to
reconnect again.
For some reason I'm able to publish to my web server, but
I'm unable to publish to my folders on my web server. Anyone
know why?
For example, I make a copy of a new page -- click Publish
> Choose Folder> --
I pick the specific folder I want to publish it to -- click
Select (destination folder) --
a screen pops up saying "Waiting for Server" -- Then it all
freezes, and I have to shut it down.
Anyone know why? Your reply is great appreciated. Thanks.I don't know if this is a stupid idea and I haven't had this
problem yet, but are these folders perhaps not writable.
There are so many servers and server-configurations out
there... Maybe you could try to fire up a ftp-program (filezilla is
free) and chmod these folders to make them writeable and see what
happens.
Also you could check the user-role settings and make sure
that there aren't any complecations in the way folder-access is
configured there.
grtz.
Hayo -
I'm using edge animate cc version 3.0.0.322.27519 and have encountered a bug when publishing. Extra text is being injected between div tags when text is nested inside a div. For example: if you draw a square on the screen and then add a text field inside the square (div), then you will see text overlapping itself when you publish. The problem does not occur when testing...only when publishing.
The problem seems to go away when an image is the bottom item inside the nested div....sometimes!
I've attached a screen shot of the issue and then a screen shot of the inspected element.
Can anyone tell me if the problem I'm seeing is something I'm doing wrong, or if this is a bug?I believe I spoke too soon. After further testing, I encountered the problem again and am unable to figure out the root cause of the problem. The problem may be due to the webkit-tap-highlight being added to the div tag. I am still seeing the problem when nesting text inside a square inside edge animate. However, I don't see it consistently and I have no idea where the webkit-tap-highlight is originating from.
Has anyone else encountered this issue when publishing in Edge Animate? -
Hi - I have a strange problem when publishing Flash movies
using Flash 8. I live in the UK, and naturally use the UK Keyboard
layout, which means among other things, that my @ key is almost
next to my Return/Enter key. In the US, the @ key is above the
number 2.
The thing is, when I publish movies in standard Window mode,
the keyboard layout is correct, and uses the UK layout that I have
my PC set-up for. However, if I publish my movie in Transparent
Window Mode, my keyboard switches to US layout, and if I need to
enter an e-mail adddress into a transparent movie, I need to use
the Shift+2 combination to get the @ sign, where normally, in
normal Window Mode, I can use the UK Shift+' combination. I can't
understand why using normal Window Mode and Transparent Window Mode
should change the way my keyboard layout works! Very strange.
Is there any kind of workaround for this bug, as I need to
publish a transparent movie that requests e-mail input, and most UK
users do not know the US Shift+2 combo to get the @ sign.
I hope this makes sense - and hope that someone can help me
on this.
Just to clarify: In published movies, In Window Mode Shift+'
gets me the @ sign in input boxes. If I publish in Transparent
Window Mode, I have to use Shift+2 to get the @ sign in input
boxes.
Thanks for any help.I see that 'Marke' didn't get your attention for the same
thing(?) Now people will see your own answer show up as 1 answer,
they'll now assume I've had my question answered. Please mind your
own business, and go and discipline someone else please. You've
made me need to post it again now! -
Hi there! Wondering if any of you could give some advice to
me about a bug when I'm publishing. I'm new to Captivate (I've only
used it for 10/15 days)
and you might know the solution.
The Problem
Whenever I seem to publish a Captivate SWF file and it has a
button linking to a page on the internet, it won't work. You can
click it a million times, but it just makes a "click" and doesn't
do anything.
However, when you use the built-in "preview" function in
Captivate it works fine, so there's not a problem with me editing
the button incorrectly.
If you could help me
ASAP
then I would be more than please.
Regards,
JakeyHi Jakey and welcome to our community
I know you said you published this file, but are you running
it locally? If so, maybe you need to visit the Flash settings page
and add the folder where your .SWF is stored to the list of
locations. Following is a link to a tutorial I created on what to
do.
Click
here
Cheers... Rick -
Siebel Sales Rep Visibility in BI Publisher, Bug???
Hi,
I have run into an issue with a bi publisher report in Siebel 8.1.1.0 and BI Publisher 10.1.3.4.0 where there is a Siebel View with a Visibliity Applet Type = Sales Rep. The data showing in the report on this view is not filtered according to "Sales Rep" I believe this is because of the View having visiblility set to "Sales Rep". Is there a way to set the visiblity type on the Integration Object/Component???Hi Antonio,
Thanks for the reply. I have tried applying the ViewMode user property to my integration object by setting the Integration Object User Property name to "ViewMode" and setting the value to "Sales Rep" but the report then pulls back 0 records. If I inactivate the user prop it should default the ViewMode to "All", if I run the report with ViewMode = All I am seeing records which are not in the view and which belong to other Sales Reps. Has anyone successfully tried using the Integration Object User property ViewMode = Sales Rep? I have been unable to get it to work.
Thanks -
Publishing bug with FTP servers - Spaces at end of file/folder names
I'm sharing what I've found after lots of troubleshooting with my Mom's iWeb website. I originally came across this in the previous version of iWeb, and have seen it again in the current version.
For the sake of this example, she had an image gallery on her site. The image file she added to iWeb had a space at the of the filename, just before the file extension, something like "image name .jpg"
In the old version of iWeb (iWeb 08), we would publish the site to a folder on her computer, and it would generate a folder for each image in the gallery. These folders would hold the original image and thumbnail image. Each folder name was the image's filename, minus the extension. In this example, the folder name was "image name " with a space at the end, which is valid on the Mac OS X file system.
When we uploaded it to the FTP site (which happened to be GoDaddy) using a 3rd party FTP tool, certain images would always fail to upload. After much testing and troubleshooting with other FTP software, we found that the FTP server did not allow spaces at the end of folder names, and it automatically replaced them with underscores. So, we tried to upload a folder called "image name ", and ended up with one called "image name_". As a result, we couldn't upload any of the files that were supposed to be IN that folder, because it could not be found under it's intended name.
Fast forward to today, using iWeb 09. Now the software allows you to upload to a 3rd party FTP server, so we've stopped using the other FTP software and do it all in iWeb.
She added a new image to her gallery and tried to publish changes, and got the error: "There was an error communicating with the FTP server. Try again lager, or check with your service provider"
Once she removed the image, publishing worked again. It turned out the original image had a space at the end of the filename, just before the extension again. Once this space was removed, everything worked again.
Hopefully this helps somebody else down the road. If you have trouble with certain files, check them for spaces at the end, just before the extension. Perhaps in a future release, Apple could update the site publishing code to trim leading/trailing spaces from folder names.
If you've seen this before, or have any further questions/feedback for me, feel free to post a response. Thank you for reading.It would be nice if Apple could make their error messages more descriptive of the actual problem so we have a better idea of how to solve it. I am "the mom" who had the problem B2Ben posted about.
The error message seemed to place blame on the ftp server or my ISP, consequently I wasted a lot of time with these people when it wasn't even remotely their fault.
As with most computer problems, I ended up learning a lot from it and it always helps with gaining confidence. This is my 20th year as a Mac user!.. yes, since 1989. I look forward to seeing what some of you have experienced with this same issue.
iMac GHZ PowerPC G$ Mac OS X (10.5.7) -
Does Oracle's implementation of JSF have this published bug?
re: jsf with filter crashes when link to static html link is clicked.
I also posted this on Sun's JSF forum and got a reply that said it sounded like this bug:
https://javaserverfaces.dev.java.net/issues/show_bug.cgi?id=146
Does Oracle know whether or not this bug is in their implementation of JSF?
Regards,
Al MalinSteve,
I have determined that, with the JDev supplied JSF libs, a jsf with filter crashes when an invalid static html link is clicked.
Replacing jsf-api.jar and jsf-impl.jar with the ones from JSF 1.1_02, my app did not crash. Instead it correctly displayed a 404 Not Found page.
After restoring the jars and correcting the static link paths in the jsp, the links worked properly.
What a way to discover a coding error!
Al -
Empty PostOffice Calendar publisher bug
I have no reasonable way of reporting this to Novell, so I'm simply
posting this here in case anyone else comes across this problem. It's
been so long since I've ran into this I forgot all about how to work
around it, so hopefully this posting comes up for me the next time I
forget the problem and go searching for a fix. I've noticed this problem
ever since v8.0.0 and I'm currently running the latest 8.0.3 HP3. Quite
simply, if a post office has no users and the POA has calendar
publishing enabled the publisher agent will get hung up indefinitely and
will no longer serve requests. You can easily see what's happening by
watching /var/logs/tomcat6/catalina.out because it will show what Post
office it's trying to connect to for a list of users but since none show
up it just sits there. The fix is to either add a user to the post
office or disable calendar publishing on that specific poa.Mike wrote:
> It's
> been so long since I've ran into this I forgot all about how to work around
> it, so hopefully this posting comes up for me the next time I forget the
> problem and go searching for a fix.
Very clever idea Mike. I will admit that I've occasionally gone to search for
something to fix a problem and found another thread where I posted an answer to
help solve the same problem for another user - heehee.
Danita
Novell Knowledge Partner
Do you have Willem Bagchus' GW 2012 Admin Guide?
http://www.facebook.com/caledonia.net -
Publishing different than preview
I am working on a training video series about Moodle. I have carefully previewed the project to confirm that everything is timed correctly and matches the audio.
However, the published version displays something much different than the preview. Errors include:
- 1 slide displays smaller and in the top left corner instead of the full screen area, making all the highlights wrong
- images from one slide are staying around on another slide
- transitions (fade in/fade out) do not look right in the published version
What is going on?I publish these videos in mp4 format.
I do not understand how the publishing mode would be the issue because I have successfully published over 20 videos using the same mp4 mode. These errors seem like a publishing bug that does not happen all the time.
Am I missing some kind of setting that will allow the publish to produce the same thing that I see when previewing my work?
Additional errors found with other recent videos:
- audio cuts in and out (plays correctly during preview)
- text does not appear (appears during preview)
Is there any way this could be a computer issue rather than a Captivate issue? It seems Captivate related since I am publishing from Captivate...sometimes the videos turn out right, sometimes they have these random errors. However, Captivate does run slowly on my computer - would that be a cause for publishing problems?
Thank you for any direction that you can provide. -
Publicly accessable list of bugs/known issue for the 2.1 release?
Does one exist? If so, where can I find it?
If you have an Oracle Support contract you are able to review any bug that has been published. We publish bugs that customer request be published. We are not trying to hide bugs from the user base, it is in your interest that every bug is not published, as this would then include trivial bugs you are not interested in. For example, we log bugs for typos, each new icon raised and updated, and for requested updates to menu titles, for the order of the tabs in the display editors and so on. If you have a bug you're interested in, you can contact Support, request the bug be published, if it's not already published and then keep track of its progress.
There have been requests that we publish bugs mentioned on the forum, and we do this where we can. Bugs are by default not published. If the product is in a beta cycle for example, then the bug is logged for that beta cycle and is not publishable. We can publish bugs that are logged against production software. This forum is not an alternative to Support or Metalink. Customers pay for the support they get through Oracle Support and the service that is available there.
Customers with Oracle Database Support contracts have support for SQL Developer.
Regards
Sue -
Problems and solutions for 9i DB R2 install on Redhat Linux AS 2.1
Installing 9iDB R2 9.2.0.1.0 on Redhat Linux Advanced Server 2.1 (aka Redhat Enterprise Linux (AS))
and applying patchset 2 (9.2.0.3.0)
This is a certified platform but I hit many problems getting the install to work. Here are the problems I hit and how I solved them.
Note: This is not intended to be a install for dummies but just an account of all my problems and how I overcame them. I spent 10 hours on this certified install (mostly looking for solutions to the problems I was having)
I had a Intel Pentium 3 with 512 Meg RAM. (test machine)
1. Install Redhat Linux
2. Install Sun JDK 1.3.1 (this is not actually required for this install. I installed it as I was planning to install 9iAS after this)
3. Install binutils-2.11.90.0.8-13 (also not required. I installed it for 9iAS)
All other components were ok on my Redhat Linux install (i.e. kernel version, glibc)
4. Create oracle user with dba group
5. Create oracle home and oracle base directories. Ensure these are owned by oracle with rwx
6. Set kernel parameters and make sure that they are initialized every startup:
I managed to locate this script in the Redhat document called
"Deploying Oracle9iTM on Red Hat Enterprise Linux AS"
create a file called oracle.sh in /etc and give it execute permissions
oracle.sh
#!/bin/bash
# configures kernel and system parameters for Oracle9i R2
# File handles
echo 65536 > /proc/sys/fs/file-max
ulimit -n 65536
# Shared memory
# The default SHMMAX value is too low and likely to cause a failure during
# database creation. SHMMAX should be equal to half of your system's physical
# RAM.
# The default is for 512 MB of RAM. Replace the default value as appropriate
# for the amount of memory in your system, i.e.,
# 512 MB : 268435456
# 1 GB : 536870912
# 2 GB : 1073741824
# 4 GB : 2147483648
echo 268435456 > /proc/sys/kernel/shmmax
# It is generally safe to leave these next two values as they are
echo 4096 > /proc/sys/kernel/shmmni
echo 2097152 > /proc/sys/kernel/shmall
# Semaphores
echo 250 32000 100 128 > /proc/sys/kernel/sem
# Sockets
echo 1024 65000 > /proc/sys/net/ipv4/ip_local_port_range
# Processes
ulimit -u 16384
----EOF---------
then...
# chmod +x /etc/oracle.sh # make executable
# . /etc/oracle.sh # apply changes now
edit /etc/rc.local to ensure this is run on startup
append to /etc/rc.local
# configure system for Oracle9i R2
bash /etc/oracle.sh
--------EOF--------------------
7. create a generic oracle environment setup script for all users to run
create the file /etc/profile.d/oracleenv.sh
be sure to change the variables to your environment (ORACLE_BASE, ORACLE_HOME and ORACLE_SID)
oracleenv.sh
# Configures the user's environment for Oracle9i R2
ORACLE_BASE=/opt/oracle
ORACLE_HOME=$ORACLE_BASE/product/9.2.0.1.0
ORA_NLS33=$ORACLE_HOME/ocommon/nls/admin/data
ORACLE_SID=orcl
LD_LIBRARY_PATH=$ORACLE_HOME/lib:$LD_LIBRARY_PATH
PATH=$ORACLE_HOME/bin:$PATH
export ORACLE_BASE ORACLE_HOME ORA_NLS33 ORACLE_SID LD_LIBRARY_PATH PATH
----EOF---------
chmod 755 /etc/profile.d/oraclenv.sh
This file will run when a user logs in.
8. Install oracle DB
This took me quite a few attempts.
The first Install I did I received the error jre was not found. This issue seemed to be relating to the fact I was not installing on a Pentium 4. Oracle was looking for the jre in oracle.swd.jre/bin/i686 (and .../lib/i686) when some of the files were placed in i386 and java was looking in i386. This created a real mess. I had encountered a similar problem on a previous install on a different platform. I found articles of people that tried to just link the directories with symbolic links but this never worked. I tried my own version but also failed.
The solution that worked for me was to de-install, remove all the oracle directories created by the install (this part I had always done) AND also delete the file /etc/oraInstl.loc. This makes the Installer truly believe this is the first time Oracle has been installed on this box. My jre problem disappeared after that. (I don't know why it just didn't work in the first place...)
The next error I received was an ORA-03113 when creating the database (in the dbca). (It was actually at this point that I found the document from redhat as mentioned in my set 6 above) I also noticed that when running the install of the database, the default was for Oracle's SGA to use 70% of the available memory on my machine (I have 512M so that's 360M). Oracle instructed me to set shmmax to around 240M. This may be why the database creation failed as I thought the shmmax parameter needed to be larger that SGA. Anyway, I reduce the SGA to 50% and it worked.
This error may also have been caused by the kernel parameters not being set properly after a reboot so if you implementing stop 6 above. You may never get this problem, hopefully.
8. Install the universal Installer 2.2.0.18.0 (a prerequisite for installing the patchset 2)
You will need this to install the patchset 2. You can get it from metalink by searching for the bug number in the patch area. Search for bug/patch number 2878462 in patches. (37Meg for linux intel)
(Did you know that when oracle refer to a bug number, you should treat this as a patch number? General the bugs are not available for public viewing but oracle create a patch of the same number as a placeholder. So don't search for the bug in the site search or by doc id, go straight to the "patches" area and enter the bug number as a patch number, you'll have more success. - I only found that out today after struggling with metalink for the last 4 years)
9. Install the patchset 2
I downloaded patchset 2 (which will bring my DB version to 9.2.0.3.0) (220Meg). There is a special symbolic link command you'll need so make sure you remember to do this.
i.e.
$ cd $ORACLE_BASE/oui/bin/link
$ ln -s libclntsh.so.9.0 libclntsh.so
10. Migrate your database
I could not get this to work. The docs says I need at least 150M shared_pool_size and 150M Java_pool_size before running the startup migrate. As I only have 512Meg RAM (and shmmax set to 230M) I did hold much hope that this would work. I started my DB without the migrate option just to see if it would actually open with 300M in my pools. It did, so I shut it down and restarted with the migrate option. After 5 minutes of heavy processing (I still have an empty database) my server appeared to hang. I decided to restart the machine and just re-create my DB with the new patchset already applied. I delete my $ORACLE_BASE/oradata/<sid> directory and remove the entry from /etc/oratab and re-created my DB. This worked fine :)
I hope this document helps you for your install. Please drop me line if it does help and I'll endevour to write more documents like this one.
Please drop me a quick line at [email protected] even just to say "thanks".
I will include a few keywords here to help the searches find this document. (mainly the keywords I searched on and failed to find a doc like this one)
ORA-03113 ORA-3113 03113 3113 jre was not found jre not found i686 i386 i586 9iDB R2 9.2.0.1.0 9.2.0.3.0 Redhat Advanced Server 2.1 AS Enterprise Linux (AS) RH install error installation errors oui dbca problem issue
Good luck,
Tim Daniell.The trick that worked for me was on a 2nd attempt. I installed once, let it fail with "jre was not found", de-install through the installer, delete all directories and files under $ORACLE_BASE (including $ORACLE_HOME), and also delete /etc/oraInst.loc.
I don't know why it fixes itself with this. Oracle say this is an unresolved bug 2726268 that lies in the installer. This is still unresolved by Oracle (as at 02-June-03) therefore it is not a published bug/patch. If this doesn't work, I suggest you open an iTAR and see if Oracle has any more suggestions for you.
Cheers,
Tim.
Maybe you are looking for
-
In support of a consulting company using timesheet custom billing categories to designate project time as billable onsite, billable offsite, and non-billable, I've encountered a series of issues which appear to be associated with the custom billing c
-
I tried to reinstall OS X... it took a whole night to do that, but all updates are still there... please help ;o)
-
How to Access v$ views from inside PL/SQL?
I cannot grant select on these v$'s for users. Could someone recommned a solution for something like: function conn_cnt(v_name in varchar2, v_stat varchar2) return number is p_val number := 0; begin EXECUTE IMMEDIATE 'SELECT COUNT(*) FROM V$SESSION W
-
Comparing strings in IfThen() structure of Hyperion Financial Reporting
Can we compare current value of a dimension with a string in the IfThen() structure in hyperion financial reporting tool? I have not been able to carry out the same successfully. Also, I need to get value from a partcular cell to use in IfThen() cond
-
hi can any one provide me sample file to file scenarios using udfs regards ram