I have a watir test that downloads some information from a web app to an Excel file, and I then open the file to confirm the contents. On my dev box everything works fine, but the scheduled automation runs (via Hudson) always fail attempting to open the Excel file. I have checked that the correct version of AutoITX3.dll is registered on both machines (Ruby 1.8, Watir 1.6.5). No other versions of AutoITx3 are registered on either machine. The error on the Hudson box is the one expected if the dll is NOT registered, but it is present in regedit in the same location as on my dev box. Both machines are WinXP.
Running the test manually on the Hudson box results in the same error -- unknown OLE server: 'Excel.Application' HRESULT error code:0x800401f3 Invalid class string.
I searched for similar errors and saw one instance where running on a virtual machine had caused similar problems, but only if the VM window was closed. I don't think this is a code error since it runs on the dev box. Any suggestions for debugging this?
TIA,
Sabrina
I find the use of Excel to be an antipattern in test automation. I know a lot of people like it, but if you're using that for automation then there are simpler ways to go about it. Consider using CSV with FasterCSV for the automation tests and checking anything into version control which I'm assuming you're using.
IMHO, you should only use Excel when a human is involved - i.e. you may a series of tests defined in CSV, but you edit them in Excel, save back and then diff against what's in a source control system, storing it as CSV.
Excel is useless to diff, so it doesn't store well in a source repo, there's no reason to install it on a CI box when CSV will do and do better.
That's my 2 cents anyhow. :)
Cheers,
Charley
I would do another search for the issue and leave Watir out of it. I can find several issues for 0x800401f3 associated with FoxPro, Perl, and Oracle.
Have a look at this article http://support.persits.com/show.asp?code=PS01032622
It names the possible causes for your error code.
bq. This error means that either the component has not been registered on the server or the ProgID passed to the Server.CreateObject method is misspelled. On Windows 2003 and XP, it may also mean a permission problem on a system registry key.
I think for you the last one is relevant:
bq. On Windows 2003 and XP, it may also mean a permission problem on a system registry key.
BTW, I googled for ":0x800401f3 Invalid class string." and this was the first result.
What are you trying to test?
That user can open downloaded file with Excel? How do you verify that automatically? With screen shot?
Or that some data is in the file? In that case I would verify contents of the file with something like roo.
I finally figured it out. What I thought was a working Excel install on the remote machine was instead the free Excel reader, which does not have the API hooks WIN32OLE looks for in the registry. Installed the full Excel program and now everything works. I knew the Universe would be consistent!
Related
Oracle form downloaded at:
http://www.oracle.com/technetwork/developer-tools/forms/downloads/index.html
When I ran the file setup_fmw_12.2.1.3.0_fr_win64.exe,
it gave me an error "Cannot launch the installer (555)".
fmw_12.2.1.3.0_fr_win64_Disk1_2of2 <-- Folder
setup_fmw_12.2.1.3.0_fr_win64-2.zip
fmw_12.2.1.3.0_fr_win64_Disk1_1of2 <-- Folder
setup_fmw_12.2.1.3.0_fr_win64.exe
Even though this relates to the installation of a developer tool, most would consider this particular type of question as generally off topic for stackoverflow. Your question also contains too little information to determine what the problem really is.
However, let's just mention some obvious things to see if it helps you solve your own problem:
Did you try the setup on another computer (or a virtual machine). This could be a good way to avoid the whole problem rather than spending a lot of time trying to fix it.
Did you locate any log files for the failed install?
A quick look seems to indicate that you can find log files at: %SystemDrive%\Program Files\Oracle\Inventory\logs (%SystemDrive% is normally C:\).
Or %SystemDrive%\Program Files (x86)\Oracle\Inventory\logs for 32bit installers on 64 bit systems).
Found in the troubleshoot section here - have a look yourself too - read from the top.
And the most obvious of all: did you contact Oracle support or search their user community or knowledge bases? Somebody will have seen this problem before. Looks like you need to register: https://support.oracle.com/
A quick search reveals that the Oracle Universal Installer is a Java based installer. Could Java be broken on the box you install on? (looks like the launcher should install the runtime automatically, but this could fail due to special conditions on the box. Try on a clean virtual machine). How to check whether java is installed on the computer.
Try disabling your anti-virus as well before running the setup on the problem box. Some setups even try to access the Internet during installation, and then your firewall could be a problem too. I would hate to turn that off though.
I got a weird problem with my VB app that has got me very confused. I am updating an existing program. I had to add some processing queue capabilities. On my system, unit testing worked great but when I compile it and run it on a different computer (a network server machine) it first tries to open an office install. If I cancel this, the program seems to run fine. The problem is that this program will be run in the background and I can't be hitting cancel each time it runs.
So my question is this: what is going on here? I'm usually a java developer we eclipse so I'm used to being able to include all the needed libs into the jar files automatically. It doesn't seem like the case with VB6. it seems like it expects all the same libs to be on all the systems. Is there any way to tell what might be triggering this?
The only thing I can think of that is causing it is that I'm using the "OpenProcess" function in the kernal32 lib. that the only major change I have made.
any help would be very appricated. thanks!
EDIT:
It seems that multiple versions of word were installed on the system and it was screwing things up somehow. We uninstalled one and it fixed it.
I ran into this a while ago and it was related to my installer for the VB6 app overwriting some system files for Office that it shouldn't have. Any time one of the System dlls was accessed Windows would determine that something was wrong and the Office installer would start up.
The short-term fix was to let the Office Installer repair the broken chain of dlls.
The long-term fix was to never overwrite built-in Windows System dlls.
You could also check out the Microsoft Fix it Center.
I made a minor change to a legacy Visual C++ / MFC app built with VS 2008. I changed some UI resources in the .rc file and compiled without any problems, then deployed it on my client's system. However, the program which was previously doing fine now fails to run on exactly one of their servers. It works fine on my laptop and on their other servers, many of whom are basically identical to the one having the trouble.
The weird thing however is that there is absolutely no error message whatsoever. No message box, no errorlevel set (when run on command prompt), no Dr. Watson entry, no nothing.
It's an MFC app that does not really comprise anything very special. It does link in some external libraries – e.g., some old version of the Xerces C++ XML parser. But this is probably not too relevant, right?
The program has a class derived from CWinApp, and I tried to add some logging in its constructor. Based on this, it looks like not even this constructor is reached.
The server in question is running Windows Server 2003 Standard Edition Service Pack 2, and we are trying to run the program in a Remote Desktop session. (Because of the client's environment, I cannot easily test in a console session right now.)
I reverted my changes from version control, which did not help – but I do not know if I had built myself the previously installed version (which ran just fine even on this server) or if it had been built by someone else.
Have also tried to reinstall the Visual C++ runtime libraries and of course reboot Windows, but neither helped. Now I'm really running out of ideas... Any clues on what I could try or check?
Probably some error occurs but is surpressed e.g. empty catch() statement or similar.
You could try and install Debugging tools for Windows WinDbg to see if you could get more info when trying to run it. Since the download is rather small 25Mb maybe it is possible to install it on your client's PC.
But first check the eventview log for your app, maybe there is something in there that can shed some light.
I'm adding some functionality to an old admin tool written in VB6. The tool uses a c++ activeX object. I haven't worked on the tool since I upgraded to windows 7 and now the active X object fails to load with the 429 "ActiveX component can't create object" error. I'm not knowledgeable about VB6 but I have registered the activex dll and type library and I can remove and add the object from the project via the project->references window (the program fails to compile when the reference is removed as expected). The VB6 ide is running in administrator mode and I get the same error if I run the exe directly as an administrator.
Has anyone else experienced a similar problem and found a way to sort it out?
Thanks,
Patrick
PS : I have also tried putting the dll in question in the working directory of the exe
EDIT: I've had no luck trying to sort this out and ended up remoting into a windows 2003 server to test the changes.
You might try using Mark Russinovich's regMon and FileMon to see exactly what files or regkeys are being requested during the load of that dll. That can often tell you if you've got a missing dependency.
Also the DEPENDS.EXE that comes with VS6 is sometimes good for that (though it doesn't necessarily get ALL the dependencies of a dll or exe).
COM is going to use the registry to map from the class name to the physical DLL file that has to be loaded.
Check the registry keys pertaining to your COM object... first under HKEY_CLASSES_ROOT, then the class name, which gets you the CLSID, then the HKEY_CLASSES_ROOT/CLSID section for the GUID of the object, and then see where the physical library lives.
A 429 could mean all sorts of different things. Many times, it means that there's a dependency missing, and the easiest way to figure out what dependency is missing is to use procmon to see what files/registry settings the app is trying to access and hopefully, catch where you get ACCESS_DENIED or FILE_NOT_FOUND or what have you.
The problem might also be with your browser. A couple weeks ago I upgraded my IE, since then I've had to make changes by opening IE, click tools, click custom, scroll to the activeX controls and enable, and/or click prompt. The same was needed for scripting and for me to use XMLHttprequest. Now everything is back to working, however the ActiveX warning alert sometimes takes over 2 minutes to pop-up, then when clicked everything works fine.
Do you have UAC enabled on that machine? Have you tried disabling it? I have seen some cases where it let ActiveX controls being registered but didn't let applications using them ran.
The title says it. I'm looking for a way to determine exactly which file/registry key this executable is attempting to access. I have attempted to use Windows auditing capabilities and Process Monitor to determine where the failure is happening but, this failure does not produce audit failure events or show as access denial in Process Monitor.
Of course, If someone has experienced this and can provide a solution to resolve the error directly that would be almost as nice.
Background:
I am using moveuser.exe which is part of the Windows Server 2003 Resource Kit Tools to convert the security of local user profiles on a number Windows XP workstations.
Symptom:
Occasionally, moveuser.exe will fail to convert the security of a profile with the error "Error: 5 access denied". I have not been able to determine any commonality among the failing accounts or the computers they reside upon. A given computer may have host 6 profiles, 5 of which convert without issue and 1 which produces the error.
There are a few factors that I'm as sure as I can be about:
-The account I'm using to run moveuser.exe with has full Administrator rights to the local machine and the domain to which the profile security is being converted.
-The failure is not related to file permissions within the profile directory (the entire directory can be moved, renamed, deleted, or successfully converted via a workaround).
I've developed a reliable workaround for these cases but, it is fairly involved and I would much rather understand the root cause of this error and correct it pre-emptively.
My workaround (glad to share it, left out for brevity) seems to indicate that the failure is related to HKEY_LOCAL_MACHINE \ SOFTWARE\ Microsoft\ Windows NT \ CurrentVersion \ ProfileList registry keys but, I cannot determine exactly how/why.
Did you try to use the "regini" command line tool to grab registry permissions for the administrator account?
Use FileMon and RegMon (now Microsoft, formerly SysInternals, still free) to monitor what exactly is being accessed, how, and what rights/access are being requested.
I don't have a URL handy, but a Google search should be able to hook you up with these tools.
A couple items I would try. First, could it be that the user was logged in and the computer hasn't been rebooted. Microsoft has a product called UPHClent which helps in unloading unneeded user hives.
Next thing I wanted to know, is if you try rebooting before running the moveuser executable. This Conversation seems to indicate this would help with this kind of error.
Two days after posting this I got to the bottom of the problem. It turned out, just as Rob Haupt suggested to be related to a stuck user hive. The program I was running, moveuser, was reading the Refcount key found under HKEY_LOCAL_MACHINE\ SOFTWARE\ Microsoft\ Windows NT\ CurrentVersion\ ProfileList\ <SID>\ and ending immediately.
Setting Refcount to 0 solved the problem immediately
Pushing out UPHClean to all the target machines pre-emptively has all but elimated the problem and we were able to successfully convert profiles on several hundred machines over the last week.
An important note about UPHClean:
I'd tried installing it previously but, it didn't appear to help. I was too impatient, the UPHClean ReadMe revealed that the service just takes time to do its job.