revolution taking up 100% of cpu, again
Jon
jbondy at sover.net
Wed Aug 24 14:40:03 EDT 2005
Mark:
This routine is only called every 10 seconds, when it is time to
schedule myself to wake up again. I guess what I need to do is
understand what "my" message is named, and then only cancel it.
Note that this 100% CPU situation only occurs periodically: I can go for
hours without it happening, so I doubt that it is [immediately] caused
by this code.
Very confusing, to me.
Jon
Mark Smith wrote:
> Why are you cancelling all pending messages that don't start with
> "rev"? Perhaps, if the timing is unlucky, you're interfering with a
> close stack message?
>
> Mark
>
> On 24 Aug 2005, at 16:29, Jon wrote:
>
>> I reported this IDE problem months ago, with a different application,
>> but here I go again. I seem to have a talent for creating problem
>> stacks like this. Last time it was a logic error in one of my handlers.
>>
>> I am writing an application to test a web site I've written. Once
>> every TBD (typically 15) minutes, it "wakes up" and Posts some data
>> to each of three versions of the same web site (supposed to be the
>> same code, only on three different servers, one of which is local to
>> my machine). It then inspects the returned HTML to see if the
>> requested operation was successful.
>>
>> I wrote the program to "wake up" every 10 seconds and see whether it
>> is time to go yet or not. Two strange things are happening.
>>
>> 1) while the application is supposed to be hitting each web
>> site/server once every TBD minutes, it seems as if it may be hitting
>> it 4-5 times, all within a fraction of a second. This undesired
>> behavior (AKA bug?) has allowed me to find and fix a number of errors
>> in the web site, but I would like to understand how and why it is
>> occurring. My first thought is that multiple Sends are getting
>> triggered when I only wanted one to be sent. I altered the Send
>> statement to be included in the following procedure, to try to
>> control this:
>>
>> on SendToMyself s local pm, str, theLine
>> put the pendingMessages into pm
>> repeat for each line theLine in pm
>> put char 1 to 3 of item 3 of theLine into str
>> if (str <> "rev") then
>> cancel item 1 of theLine
>> end if
>> end repeat
>> if (s = 0) then
>> send "TestTheReports" to the target
>> else
>> send "TestTheReports" to the target in s seconds
>> end if
>>
>> Can anyone see a problem with the way this is written?
>>
>>
>> 2) if I run the stack for 20 hours or so, and then click on the Close
>> icon, Rev first asks if I want to save the application, after which
>> Rev starts to use 100% of the CPU. I have to go kill the IDE, even
>> though, in theory, nothing is running any longer. This can also
>> happen if I switch from Run mode to Edit mode in the IDE, at which
>> point I lose EVERYTHING I've worked on (no Save is possible). I Save
>> the application every chance I get because of this. If I had to
>> guess, I would think that the IDE is losing it over residual
>> Sends/messages (see above)
>>
>> I tried monitoring the Messages, but did not see the spray of
>> messages that I had thought that I might: I only see that one message
>> every 10 seconds.
>>
>> Any thoughts about this situation? I can provide the stack if you
>> wish...
>>
>> :)
>>
>> Jon
>>
>>
>> _______________________________________________
>> use-revolution mailing list
>> use-revolution at lists.runrev.com
>> Please visit this url to subscribe, unsubscribe and manage your
>> subscription preferences:
>> http://lists.runrev.com/mailman/listinfo/use-revolution
>>
>
>
>
> ___________________________________________________________ To help
> you stay safe and secure online, we've developed the all new Yahoo!
> Security Centre. http://uk.security.yahoo.com
> _______________________________________________
> use-revolution mailing list
> use-revolution at lists.runrev.com
> Please visit this url to subscribe, unsubscribe and manage your
> subscription preferences:
> http://lists.runrev.com/mailman/listinfo/use-revolution
>
>
More information about the use-livecode
mailing list