Install4j upgrade installation sometimes gives "Error opening media file. Please move this file to an ascii-only path like c:\" - install4j

...and then:
"Log: C:\Users\GEBRUIKER2\AppData\Local\Temp\i4j_nlog_7"
This happens consistently, but on relatively few machines only (ca. 1 in 100), and only when trying to update an existing installation. It happens both for install4j 5.1.15 and 6.1.2.
The only reliable work-around is to uninstall and reinstall manually, which means we have to contact all affected customers to do this for them.
I found this related link:
https://board.jdownloader.org/showthread.php?t=31405
However, they "solved" it by doing the manual uninstall-reinstall cycle, which does not scale for many customers.
What causes this error - the listed file is clearly on an ASCII-only path already - and what can we do to solve it in general?

From your comment on the question:
Apparently, the error happens when immediately clicking "install now" as soon as the update has been marked as downloaded in the updater. Waiting for a split second prevents the error.
This sounds like an Anti-Virus-Tool is checking the downloaded executable and is preventing read access to it.
In your updater, look for the action named "Download URL" and add a "Sleep" action right after it.

Related

"Error applying transforms. Verify that the specified transform path are valid." when uninstalling

I am trying to uninstall Crystal Report for Visual Studio 2011, and install Crystal Report for Visual Studio 2019. I got the error message of "Error applying transforms. Verify that the specified transform path are valid." when uninstalling the program. Therefore, the newer version of Crystal Report for VS 2019 can't be installed.
I searched around and some posts says about windows registry entries caused the problem. I can't find out a solution on what to do.
Highly appreciate your response.
Crystal Reports: I am wondering if you have installed from a network share that is no longer available, or one where the UNC path specified to your transform is blocked or in other ways incorrect. What you need is a proper log file. Please see below. And have a quick peek here: https://apps.support.sap.com/sap/support/knowledge/en/1220433
There are also some issues with secure transforms and complications arising from this and newer Windows settings and security features. Maybe have a quick look here. Just a few links on the topic.
How many machines do you see this on? Just your own?
Preparation: A couple of things first:
Reboot: Do a reboot before attempting the next uninstall. Just to have a clean slate. Allow system to settle after reboot (give it a couple of minutes to settle down).
Corrupt installation files: Re-download your new setup to make sure its installation file is not corrupted. Try to malware scan it too. And finally set it unblocked as shown here.
Admin Rights: Second, make sure you run with proper admin rights? Run the msiexec.exe command from an elevated command prompt. Please launch an elevated cmd.exe (right click => run as administrator).
Anti-Virus: Disable your anti-virus first to prevent any locks from failing your uninstall.
Debugging "Ideas Lists": Common causes of setup runtime issues
Logging: Now the most important. You must also ensure proper logging for the uninstall effort. You can either use logging by policy
or define it at the command line level. It would be best to enable the
logging policy so the log file is automatically created in the tmp
folder. Please see this answer for more on logging.
When you have done the "Preparation" above, please run the
uninstall and create a proper log file. Here is the command line
(prefer the policy):
msiexec.exe /x "mysetup.msi" /L*V "C:\Temp\msilog.log"
Please open the log you created (or get the log from the temp folder
if you have logging policy enabled). Then inspect the log and match
against this sample log here:
https://www.itninja.com/question/transform-issues
What do you see? Just read line by line and don't be intimidated by
all the "line noise".
Common Technical Issues: If you have problems with all other MSI packages and their operation, you could have a bigger problem. Then you should rule out some runtime issues. Note that some issues are commonly caused by malware (or just technicalities that occur randomly sometimes):
Visual C++ Runtime - reinstall it. There are many versions. See what your package needs. The latest supported Visual C++ downloads.
Unregister / re-register msiexec.exe (can be necessary because of malware or normal technical glitches).
Run chkdsk.exe and sfc.exe to check for file corruptions and corrupt OS files.
Microsoft FixIt: You can use the Microsoft FixIt method as a last resort to clean out your existing installation. This generally works (unless you have hacked too much already), but is not ideal: http://support.microsoft.com/mats/Program_Install_and_Uninstall/ - this approach does not clean up or uninstall, it just unregisters the installed package and leaves all its files and registry settings in place. You can try to install the new version, but some interference issues are likely with the garbage left behind. Yes, you can try to clean up manually, but I would just try to overwrite first.
Links:
What is the root cause of "Error Applying Transforms. Verify that the specified transform paths valid"?
All About the Four Types of MSI Transforms
"Error applying transforms. Verify that the specified transform paths are valid."

Error occurs when I update in Eclipse using SVN

Problem while receiving an update from Eclipse to svn. The error message is '0x00000: Open Project' operation completed with error: 0x000000: Unable connection to 'blah-' project. Please check that SVN meta-information exists and it is supporting the form ...
Except that the operating system has been updated from Windows 7 to Windows 10, nothing has changed, but it is annoying and cumbersome to see such errors every time we update. Of course, there are solutions such as cleaning up or re-connecting, but I want to prevent this problem from occurring fundamentally.
Can you tell me what to look at?
Thank you for reading it.

Typo3 extension are automatically deactivated

i have a running Typo3 7.6.11 installation on my webhoster.
I noticed that the extensions "powermail" and "dynamic content elements" get somehow automatically deactivated over night.
I can activate them in the Backend via Extensions, but the next morning they are both deactivated again.
Any ideas on that?
What cron-jobs and scheduler-tasks are configured? at what time?
the next I would do is writing a cronjob, which logs every 5 minutes the md5-hash of your PackageStates.php into a file. if the file is modified the hash should change and you nearly know the time the change occurs. compare to cron-jobs/ scheduler-tasks running at that time.
Check the system log of your installation, if someone else did not login during night into your system. If no-one did, contact your webhoster, since someone has is changing your files directly on the system, so they should be able to figure out if it is some security issue or maybe some faulty backup procedures.
In the Install-Tool there is an option to do a check for broken extensions and after the check the "broken extensions" are deinstalled automatically. May be you have a task running in the scheduler that does this routine?
Normally after a system update some of the extensions are broken because they use a deprecated api and also have to be updated, so you should also check if there is an update for these extensions.
do you have an automatic deployment?
does this deployment know that these extensions needs to be active?
Please check permission of typo3conf/PackageStates.php file, Is there any cron-job that automatic deactivate extensions.
If not then try to overwrite extensions with new vesrion.

Install4j "Browse" Button hanging the installer

I'm using install4j 6.0.3, everything works exactly like it should, however if the user tries to change the installation location by using the "Browse" button, the installer locks up completely, and you need to use task manager to kill the process.
This happens on Windows 7 & Windows 8(.1)
Does anyone have any idea of how to fix this issue?
Ok, this is extremely weird, as I was debugging according to Ingo, I checked the option "Allow new folder creation" on, made a build, then checked it off again and made another build.
That build now works for the affected users, I guess there was a latent weird option in the config file or something.

Receive message "A signed resource has been added, modified, or deleted" when trying to debug an App on iPhone

While attempting to debug a build created using the 3.2 SDK on an iPhone device I receive the message "A signed resource has been added, modified, or deleted.".
I can clean, rebuild, then install with no error, but if I try to install without cleaning the error shows.
Anyone have an idea as to what might be causing this?
I found a workaround for the bug.
If you delete the .app file in build/Debug-iphoneos/ before building for the device, the app gets installed without errors.
And there is a simple way to do that before every build.
Make sure you have selected "Device" in the dropdown overview menu.
In XCode go to Project > New target...
Then find "Shell Script target" under MacOSX/Other
Name it and add it to the current project
Now, in the left navigation panel, under targets, expand your newly created target and double-click on Run Script.
In the window that opens replace "# shell script goes here" with "rm -fr build/Debug-iphoneos/*.app" (without the quotes).
Now open your main target preferences and under Direct Dependencies add your newly created target.
Build and Go! :)
This error occurs when there is a special character in the Product Name. In my case it was a "?"
If you change the Product Name it automatically updates the "Bundle Name" and "Bundle Display Name" so it is often the best choice to rename an app.
If you want to include special characters in the app name you have to manually rename the "Bundle Name" and "Bundle Display Name"
Bundle Name: This is the actual app bundle name in the file system such as "Awesome App.app". It is generally not visible to the user.
Bundle Display Name: This is a short name displayed under the app icon on the device. Since the bundle name would be truncated to "Awes…tion" you have the option to have a shorter name which fits better such as "Awesome App". It should be similar to the App Store name (set in iTunes Connect)
This is pretty clearly a bug in the 3.2 SDK, but I don't want to downgrade. I've found that doing a Clean by pushing Command+Shift+K, then Return is pretty fast before pushing Command+R to build.
Xcode 8, reason of the "A signed resource has been added, modified, or deleted." was that target was signed with an enterprise provision profile.
In my case, it happened when no changes were made. Make a change to any file and run again.
This can have several causes. The fastest way to figure out what is causing it is to go into Xcode, Window menu, Devices, then click the reveal button at the bottom of the pane to show the Console. Now attempt to run. You should see log output that names the specific files it is complaining about.
Most of the solutions previously posted are just artificial ways of getting Xcode to regenerate the contents of the build folder and/or re-sign the files.
In my case, my WatchKit extension was somehow acquiring references to Cocoapods frameworks that were only targeted toward the main app so they got signed during the build, then pruned later (as they were not used). Then on device, iOS complained that they were missing from the .appex folder for the extension. I ended up not needing any pods in the extension so I just removed them all and removed the extension as a target, then did some minor cleanup to remove the pod-related debris left in the build steps. Now everything works perfectly.
(SOLVED) This is a weird one. I tried everything I could find. Eventually I changed the product name from "Unit Tests (device)" to "Device Unit Tests" - removing the brackets. Now everything works. The spaces in it appear to be fine.
Previously on stackoverflow:
I've just run into this bug with two static library projects. One builds and tests using the GHUnit test runner on the device without a problem. The other projects will not install and gets this error. That means it's something thats different between these two projects. I've so far tried wiping the build directory, taking spaces out of the executable name, and various clean and builds as suggested here.
Same for me, thought it has something to do with multiple targets etc. because I changed a lot there. But it's highly possible that it's a Bug in the 3.2.2 release since I did not test extensively in this sdk version before the massive target changes in my project.
solved my issue!!!
I found out by accident that somehow a space " " found it's way into the Product Name of my app so it was called "First Second.app" instead of "FirstSecond.app". After deleting the space the issue was gone!
I changed it here:
right click on target
Get Info
Build Tab
Packaging Section
Product Name <- The name here will be used for the bundle (.app) name
Hope this helps, let me know!
Cheers,
nils
I could solved by changing project name.
[project]-[Rename] menu. "phase1 (new)" -> "pahse1"
I was getting this same error, but intermittently. I tried all the above and it still didn't work. Today I found what was causing it.
The error seems to occur when editing a xib in interface builder. If you try to run while the interface builder is open in xcode it will cause the above error. To solve just close the interface builder editor. i.e. just select a code file from your project so you are in the Source Editor.
The simplest (and probably most common cause) appears to be rebuilding without any changes.
So the simplest thing to cure it is to make a trivial change to a source file (such as adding a space, then deleting it), and then rebuilding.
If this doesn't work, feel free to try all the other answers here.
For months, I'd get this error without realizing it was due to such a simple cause. I'd usually do a Clean Build to get rid of it.
When I created ipa through terminal using xcodebuild commands, ipa created but while installing it I was getting same error. exportOptionsPlist solved my issue.
xcodebuild -exportArchive -archivePath projectPath/myapp.xcarchive -exportPath projectPath/myApp.ipa -exportOptionsPlist ProjectFolder/exportPlist.plist
In my case, Quit and restarting XCode worked.
For me the issue was related to the provisioning profile settings. The clue to this was that debug builds were installing ok, but release builds were not. I wanted to test a release build, so I ran the scheme with that build configuration.
I fixed it by duplicating the Release Configuration, then modifying those fields in the Build Settings to have the same provisioning stuff as if I am debugging it.
(Adding another build configuration will give you headaches if you are using Cocoapods however, then you'll have to modify your Podfile)
I'm getting the same thing, when installing on a iPod Touch. I can't link for the simulator (for other reasons), so can't say whether the problem occurs there.
Yes, rebuilding clean or deleting the app from the device allows me to install again. Neither are desirable, iterative solutions!
The minimal "cleaning" I've come across as a work around is manually deleting the Foo.app in the build/Debug-iphoneos directory.
it seems this is a bug in xcode 3.2.2:
iphonedevsdk
I had the same problem in Xcode 3.2.1 when I put a + in my app name. Specifically the "product name" in the build settings. It is fine to have a + in the bundle name in your Info.plist. The same probably applies to other punctuation characters.
Go to Window > Organizer > Projects > Find your project and delete derived data
I got this error intermittently while installing app using iPhone config utility on Windows7. Following solution works - Go to C:\Users\{lanusername}\AppData\Local\Temp and delete app specific folders (e.g. abc.app) and try installing app again.
I reported this bug on ICU (Windows versions) to Apple in June 2011. With the following workarounds:
The workaround is this ....
Win XP
1) Close ICU
2) Delete the temp folder: c:\Documents and Settings\[username]\Local Settings\Temp\[AppName].app
3) Delete the deploy folder: c:\Documents and Settings\[username]\Application Data\AppleComputer\MobileDevice
4) Restart ICU. Drag in the App and install normally.
============================
Win 7
1) Close ICU
2) Delete the temp folder: c:\Users\[username]\AppData\Local\Temp\[AppName].app
3) Delete the deploy folder: c:\Users\[username]\AppData\Local\Apple Computer\MobileDevice\Applications\[AppName].app
4) Restart ICU. Drag in the App and install normally.
=========================================================
I simply rebuilt my app, and that solved the issue.
I also faced the same issue. After wasting lot of time I realized that my product name has a special character "?" which cased the problem
Having the DerivedData folder at a network location caused this problem for me.
After trying everything else, I found out my workstation couldn't agree with the University server about what the time was. (It thought everything was always modified). I also had to clean for every rebuild to avoid that frustrating message.
In the end I gave up and built locally, changing Xcode > Preferences > Locations ... feeling altogether pretty dumb for having ever built over the network.
We ran into this issues on XCode_6.3.1. We were building a AppleWatch app, with an extension. We do have a bunch of Pods.. After debugging the issue for almost a bunch of hours, what we found was that there was an issue with the way a file was adde to the project..
It seems like some references to a unused file was sitting in the iPhone App, though it was used in the Watch App.. It turns out that the error XCode was showing was totally useless.
After removing this file and re adding it back to the project the project started working fine & was able to install to the device. To make it even harder to debug the issues, the debug version was installed without an issue, but was unable to install the norman version..
Make sure you add your files to the right target and, look at git history and see if there are lingering fragments that are added to the wrong target.
This is a very general error message indicating something is wrong during the validation process of the code signature. To find out the specific error, you can go to Xcode->Window->Devices and check your device log.
In my case, I have following console spew
Feb 1 18:53:07 iPod-touch installd[40] : 0x1001f8000 -[MICodeSigningVerifier performValidationWithError:]: 192: Failed to verify code signature of : 0xe8008017 (Signed resources have been added, removed, or modified)
Check on this 3rd party framework again, I found an extra CodeResources file under the framework root. Remove that file fixed the problem.