weird problem…
I have a small form which I use to display a jprogress bean with a formindialog command during lengthy operations. I have been using this for a while without any problem except that refreshing the bean is not enough, I need to updateUI to see the progress bar changes.
now, I often start these operations on a PC which is remote controlled via TSE, and leave it unattended. and after a while, the session locks as if I was working locally, all normal, but when I re-open it later on to check the result, the updateUI command has stopped the process after a while with a null pointer error, etc…
Does anyone know how an UpdateUI() command can return such an error? After the error, the form is still visible, but nothing refreshes, I have to quit the client and start again to see something.
lesouef:
weird problem…
I have a small form which I use to display a jprogress bean with a formindialog command during lengthy operations. I have been using this for a while without any problem except that refreshing the bean is not enough, I need to updateUI to see the progress bar changes.
now, I often start these operations on a PC which is remote controlled via TSE, and leave it unattended. and after a while, the session locks as if I was working locally, all normal, but when I re-open it later on to check the result, the updateUI command has stopped the process after a while with a null pointer error, etc…
Does anyone know how an UpdateUI() command can return such an error? After the error, the form is still visible, but nothing refreshes, I have to quit the client and start again to see something.
Which is the null pointer error that you receive ? (full stack from log)
here it is, attached
update_ui.log.txt (6.32 KB)
the error is in java code, maybe the cause for this is the java version you use on the other computer ? What java version you have there ? Can you try with other version (maybe same as the one that didn’t give any problems) ?
we added a small enhancement for 5.2, so that if an exception occurs in updateUI it doesn’t stop the scripting; will be present in 5.2.1
the designer machine is 1.6u18, but I also checked the problem on another one running 1.6u20, same problem.
I found what is causing this pb while having no fix yet, just a poor workaround which is disable the update if I plan to leave the keyboard.
on lengthy operations, the screen saver gets on after 20mn on my machine. and obviously the application udpate crashes just a few sesonds later, not instantly.
of course, it is not very useful to refresh the screen when nobody is looking at a monitor, but there is no way to detect this. so I’ll modify my progress bar now to see if the bean is causing it or if it’s a general problem whatever is the screen refreshed.
lesouef:
I found what is causing this pb while having no fix yet, just a poor workaround which is disable the update if I plan to leave the keyboard.
on lengthy operations, the screen saver gets on after 20mn on my machine. and obviously the application udpate crashes just a few sesonds later, not instantly.
of course, it is not very useful to refresh the screen when nobody is looking at a monitor, but there is no way to detect this. so I’ll modify my progress bar now to see if the bean is causing it or if it’s a general problem whatever is the screen refreshed.
If you have a sample where to reproduce this please create a case and we’ll have a closer look. Also you can try with other Java versions, maybe is a bug in Java.
my java machine is still the 1.6u18, need to sort out plugins first…
the problem is that it happens only when using TSE to control the client (I am working remote most of the time, using my office PC via a VPN), and I am not sure it was clear that the screen saver itself is not making any problem, only when I have the screen saver on + remote control via TSE. another side effect on lengthy operations is that when I am back the screen never refreshes back again (if the progress bar is not involved), I have to move to another form to see something again, but the “blackened” form is dead for ever. looks like it is something in between java and TSE? do you have ways to repeat such a side effect? because my solution has nothing special and work otherwise…
lesouef:
my java machine is still the 1.6u18, need to sort out plugins first…
the problem is that it happens only when using TSE to control the client (I am working remote most of the time, using my office PC via a VPN), and I am not sure it was clear that the screen saver itself is not making any problem, only when I have the screen saver on + remote control via TSE. another side effect on lengthy operations is that when I am back the screen never refreshes back again (if the progress bar is not involved), I have to move to another form to see something again, but the “blackened” form is dead for ever. looks like it is something in between java and TSE? do you have ways to repeat such a side effect? because my solution has nothing special and work otherwise…
Can’t you try with Java 1.5, for example ? Or an earlier release of 1.6 ?
no, I have none of these anymore, only newer versions, 1.6u20+ which does the same…
leave this for now, no need to work on it, just good to know for remote workers like me untill I can find more relevant stuff on this.