In .Net 3.5, I have the following code.
If File.Exists(sFilePath & IndexFileName & ".NX") Then
Kill(sFilePath & IndexFileName & ".NX")
End If
At runtime, on one client's machine, I get the following exception, over and over, when this code executes
Source: Microsoft.VisualBasic
TargetSite: Microsoft.VisualBasic.FileSystem.Kill
Message: No files found matching 'I:\RPG\HGIAPVXD.NX'.
StackTrace:
at Microsoft.VisualBasic.FileSystem.Kill(String PathName)
(More trace that identifies the exact line of code.)
There are two people on different machines running this code, but only one of them is getting the exception. The exception does not happen every time, but it is happening regularly. (Multiple times every hour.) The code is not in a loop, nor does it run continuously, more like once every couple of minutes or so.
On the surface, this looks like a race condition, but given how infrequently this code is run and how often the error is happening I think there must be something else going on.
I would appreciate any suggestions on how I can track down what is really going on here. A solution to keep the error from happening would be even better.
I guess the first question to ask is "IS the file really there or not?" and if so, does it have any specical attributes (Is it Read-only or Hidden, or System --- or a Directory)?
Note the Microsoft.VisualBasic.FileSystem.Kill specifically looks for, and silently skips, any file marked "System" or "Hidden". For pretty much any other problem you would have gotten a different exception.
as James pointed out the Kill functions checks if the file in case is a system or hidden, you better use System.IO.File.Delete() instead
Try
System.IO.File.Delete(sFilePath & IndexFileName & ".NX")
Catch ex As System.Exception
...
End Try
using File.Exits is not neccasary because File.Delete() checks this by itself.
Is there any chance that the I: drive is a network drive? it could be some network issue... or then maybe a race condition
Related
(using OS X 10.9.4) I have this cool Ruby script which scans system/firewall logs and tells me if anything odd is happening. The script runs on a 1 second loop, however at exactly the 3852nd iteration, the script terminates with a "stack level too deep (SystemStackError)" error.
I am not new to this error and it seems to appear when a script enters a loop, and the system sandbox (probably) terminates it after a set amount of time or a specific parameter is reached.
I attempted to bypass the error by running the script as root, however this had no effect. I have also considered programming another script to relaunch the original script when it detects its absence in the command: ps -efs' output, however this is a very 'clunky' method which i would prefer to avoid.
I have also conducted some research into the error on the "Stack Overflow", however only found questions answered by altering the offending script ,as the error is in their case due to a bug in the code, which is not the case for me.
So my question:
Is there any possible way to bypass the "stack level too deep (SystemStackError)" error
Thanks in advance, greatly appreciated.
I have a VB6 application that I'm trying to make log out differently. What I have is a flag in the registry (existing) which states if the application is set to Debug mode so that it would log out.
Within my code I then have lots of if statements checking if this is true. This means that there is a lot of processing time checking if a statement is true, which maybe not much really but as it does it so often it's an overhead I would like to reduce.
The code is full of statements like this
If isDebug = True Then
LogMessage("Log what is happening")
End If
So what I'm looking for is a better way to do this. I know I can set a debug mode within Project Properties -> Make, but this needs to be set prior to building the .exe and I want to be able to set this in production via the registry key.
Consider using a command line argument to set debug mode. I used to do this.
Dim sCommandLine() As String
sCommandLine = Split(Command$)
For I = 0 To UBound(sCommandLine)
' do something with each arg
Next I
You can also persist command line args inside the IDE, so you always have them when debugging. When running outside of the IDE, make a shortcut to the compiled application with the arguments in it.
I do something almost identical to what you have in mind in a lot of my code. Add this:
Sub LogDebug(ByVal strMsg As String)
If (isDebug) Then
LogMessage(strMsg)
End If
End Sub
Then just call LogDebug in your main program body, or call LogMessage directly if it's something you always want to log, regardless of the debug flag.
I'm assuming isDebug is a boolean here. If it's a function call, you should just create a global flag that you set at the beginning of the code, and check that instead of looking at the registry over and over. I don't think checking a boolean is that much of a processing load, is it?
You want to call a function if a runtime flag is set. The only thing I can see that could be faster is:
If isDebug Then
LogMessage("Log what is happening")
End If
But I doubt that either would be the cause of performance problems. Most logging frameworks promote code like that and even put the flag/log level as a parameter to the function. Just be sure that you don't have other places where you needlessly compute a log message outside of the conditional statement.
You might evaluate why you need logging and if the logs produced are effective for that purpose.
If you are looking for a problem that can be trapped using VB error handling, consider a good error handling library like HuntERR31. With it you can choose to log only errors instead of the debug message you are now doing. Even if you don't use the library, the docs have a very good description of error handling in VB.
Another answer still:
Read your registry flag into your app so that it's a session based thing (i.e. when you close and restart the app the flag will be checked again - there's no point in checking the registry with every single test).
Then (as per Tom's post) assign the value to a global variable and test that - far faster than a function.
To speed up logging you may want to consider dimensioning a string buffer in your app and, once it has reached a specific size, fire it into your log file. Obviously there are certain problems with this approach, namely the volatility of the memory, but if you want performance over disk access I would recommend such an approach.
This would, of course, be a lot easier if you could show us some code for your logging process etc.
I set up a scheduled task to run under my account. Everything it runs, even if it is successful, returns an operational code of (2). I looked this up this error code at the below link, and it claims it cannot find the specific file.
http://www.hiteksoftware.com/knowledge/articles/049.htm
Even if I do something very simple, I get back operational code of (2). For example:
run program: cmd.exe
start in path: c:\windows\system32
I start the task and I see the process running in my task manager, so I kill the task. I then check in the history of scheduled task and it shows up as (2).
Something more realistic of what I am doing:
<?
/* file in c:\php\test.php */
echo "hello";
?>
run program: php.exe
start in path: c:\php
arguments: -f test.php
Everything works in the command line, but Windows schedule task keeps returning operational code (2). I should be seeing an operational code of (0), which means successful, correct?
You may not have put a path in the "Start In (Optional) box of the Edit Action dialog box.
Even though you had a path on the program that was being executed, Windows 7 still wants you to tell it where to run the program.
TL/DR: Don't worry about it. This just means the task finished, but tells you nothing about whether it was successful or how it failed. Look at the "Last Run Result" for that information.
The question and the top answer are confusing the notion of a "return code", which shows up in Task Scheduler as the "Last Run Result" with the "OpCode"/"Operational Code" that shows up in the history of a task.
If I create a simple Python program that does nothing more than sys.exit(7), and run it via task scheduler, I get a Last Run Result of 0x7, and an opcode of 2. If I have it do nothing, or sys.exit(0), I get a Last Run Result of "The operation completed successfully (0x0)" and still an opcode of 2. In other words, the return code from the executed program determines the Last Run Result. The OpCode appears to be a constant 2. This also establishes that the opcode 2 is not related to the return code 2 that likely means the file's not found. We know the file was found as it executed, and returned different Last Run Results depending on the code contained.
Further, a Windows forum post points out that this history view is really coming out of the event log. Sure enough, I can find the same events in the event log (always with a value of 2). This means the definition of the OpCode is going to be the same as the definition used for events, and is less of a task scheduler concept than a Windows event concept.
What is an opcode for an event? I've struggled to get a clear answer, but as best I can tell, it appears it's ultimately controlled by the program writing to the event log. There's documentation around for defining opcodes in your program. In this case, the thing writing to the event log would be Task Scheduler itself or something else in Windows.
A final observation: If I go to the event viewer and look for Log: Microsoft-Windows-TaskScheduler/Operational, Source: Microsoft-Windows-TaskScheduler and Event ID: 102,201, add the column for Operational Code, and sort, I see it is always a 2. And events 100 and 200 are always a 1. This applies not just to my manual experiments, but also includes every other random program that's using scheduled tasks, e.g. Dropbox and Google updaters that are working as far as I know.
Put all this together and I would strongly bet that the events generated while starting up a scheduled task are hardcoded by Windows to use an opcode of 1 when writing to the event log, and the events generated while finishing a task (successful or not - which goes in the Last Run Result) are hardcoded by Windows to use an opcode of 2 when writing to the event log. This opcode appears to be a red herring that doesn't affect anything we need to worry about beyond curiosity.
I was striking out until I just deleted & re-created the scheduled task...now it works. Don't know why but there it is.
Okay I know I am late to the party here, but I think a lot of the problem stems from confusing the Operational Code with a Return Code. I'm not an expert in Windows programming or internals (I make a living using a Windows system to program, but my programming isn't for Windows systems).
If I understand correctly:
The Operational Code is set by what ever routine being run at whatever value the programmer decided to set it at.
The Return Code is indicative of success or failure.
Consider the following (edited) example from the history of one of my scheduled tasks:
Event 201, Task Category "Action completed" shows an Operational Code of (2).
Down below under the General tab, is the message:
Task Scheduler successfully completed task "\My_task" , instance "{xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx}" , action "C:.....\blahblah.exe" with return code 0.
There's the indication of success. A different return code would indicate a failure. The Operational Code of (2) merely indicates that the routine was finished (in this case) when reported. I don't believe there's any set values to be interpreted for the Operation Code.
I've been having a similar issue and found that in addition to what was suggested in both the accepted answer and its comments I had to do one other thing. I had to re-create the task and set its "configure for" to Windows Server 2003, Windows XP, or Windows 2000 I dont understand why, since its not for any of those OS' but after I did so my task actually worked.
If this runs, and works, yet you still get an error code try entering exit 0 at the end of your script.
It took me a lot of googling to find that so hopefully this is helpful to someone.
#ojchase is right.
Opcodes are attached to events by the event provider. An opcode defines a numeric value that identifies the activity or a point within an activity that the application was performing when it raised the event.
Opcode 1 means that, when producing the event, the application was in the start of an activity.
Opcode 2 means that, when producing the event, the app. was at the end of an activity.
So opcodes have little to do with success or failure.
Sources:
https://learn.microsoft.com/en-us/dotnet/api/system.diagnostics.eventing.reader.standardeventopcode?view=net-5.0
https://learn.microsoft.com/en-us/dotnet/api/system.diagnostics.eventing.reader.eventopcode?view=net-5.0
Is there any way to find out what was the last Exit Code of an application the last time it run?
I want to check if application wasn't exit with zero exit code last time (which means abnormal termination in my case) And if so, do some checking and maybe fix/clean up previously generated data.
Since some applications do this (they give a warning and ask if you want to run in Safe Mode this time) I think maybe Windows can tell me this.
And if not, what is the best practice of doing this? Setting a flag on a file or something when application terminated correctly and check that next time it executed?
No, there's no permanent record of the exit code. It exists only as long as a handle to the process is kept open. And returned by GetExitCodeProcess(), it needs that handle. As soon as the last handle is closed then that exit code is gone for good. One technique is a little bootstrapper app that starts the process and keeps the handle. It can then also do other handy things like send alerts, keep a log, clean up partial files or record minidumps of crashes. Use WaitForSingleObject() to detect the process exit.
Btw, you definitely want to exit code number to mean the opposite thing. A zero is always the "normal exit" value. This helps you detect hard crashes. The exit code is always non-zero when Windows terminates the app forcibly, set to the exception code.
There are other ways, you can indeed create a file or registry key that indicates the process is running and check for that when it starts back up. The only real complication with it is that you need to do something meaningful when the user starts the program twice. Which is a hard problem to solve, such apps are usually single-instance apps. You use a named mutex to detect that an instance of the program is already running. Imprinting the evidence with the process ID and start time is workable.
There is no standard way to do this on the Windows Platform.
The easiest way to handle this case is to put a value on the registry and to clear it when the program exits.
If the value is still present when the program starts, then it terminated unexpectedly.
Put a value in the HKCU/Software// to be sure you have sufficient rights (the value will be per user in this case).
I am using VB6 SP6
This code has work correctly for years but I am now having a problem on a WIN7 to WIN7 network. It also works correctly on an XP to Win7 network.
Open file for random as ChannelNum LEN =90
'the file is on the other computer on the network
RecNum = (LOF(ChannelNum) \ 90) + 2
Put ChannelNum, RecNum, MyAcFile
'(MyAcFile is UDT that is less than 90 long)
.......... other code that does not reference file or RecNum - then
RecNum = (LOF(ChannelNum) \ 90) + 2
Put ChannelNum, RecNum, MyAcFile
Close ChannelNum
The second record overwrites the first.
We had a similar problem in the past with OpportunisticLocking so we turn that off at install - along with some other keys that cause errors in data in Windows networks.
However we have had no problems like this for years, so I think MS have some new "better" option that they think will "improve" networking.
Thanks for your help
I doubt there is any "bug" here except in your approach. The file metadata that LOF() interrogates is not meant to be updated immediately by simple writes. A delay seems like a silly idea, prone to occasional failure unless a very long delay is used and sapping performance at best. Even close/reopen can be iffy: VB6's Close statement is an async operation. That's why the Reset statement exists.
This is also why things like FlushFileBuffers() and SetEndOfFile() exist at the API level. They are also relatively expensive operations from a performance standpoint.
Track your records yourself. Only rely on LOF() if necessary after you first open the file.
Hmmm... is file (as per in the open statement at the top of the code sample) UNC filename or similar to x:\ where x is the mapped drive? Are you not incrementing RecNum? Judging by the code, the RecNum is unchanged and hence appears to overwrite the first record...Sorry for sounding ummm no pun intended... basic...It would be of help to show some more code here...
Hope this helps,
Best regards,
Tom.
It can be just timing issue. In some runs your LOF() function returns more updated information than in other runs. The file system API is asynchronous, for example when some write function is called it will not be immediately reflected as the increazed size.
In short: you code have shown an old bug, which is just easier to reproduce on Windows 7.
To fix the bug the cheapest way: you may decide to add a delay (it can be significant delay of say 5 seconds).
More elaborate fix is to force the size update by closing and reopening file.