Re: [api-dev] Restarting a frozen OOo

2006-06-27 Thread Mathias Bauer
Andreas Schlüns wrote: This is not an asynchronous signal only ... it's our crash handling. That means: one of these threads produced a crash and here we try to save all currently open documents. So the deadlock is a follow up of a crash in a remote use case of the office. Solutions:

Re: [api-dev] Restarting a frozen OOo

2006-06-27 Thread Kay Ramme - Sun Germany - Hamburg
Mathias Bauer wrote: Well, though I agree that fixing the crash is always a good idea ;-) I also wonder wether the thread owning the SolarMutex does not return from the X11 call. Obviously the main thread was going to show a tooltip window. Once this had been finished the emergency saving

Re: [api-dev] Restarting a frozen OOo

2006-06-27 Thread Kay Ramme - Sun Germany - Hamburg
Andreas Schlüns wrote: This is not an asynchronous signal only ... it's our crash handling. That means: one of these threads produced a crash and here we try to save all currently open documents. If it was a synchronous signal, the source of the signal would have been on the stack, as it

Re: [api-dev] Restarting a frozen OOo

2006-06-23 Thread Kay Ramme - Sun Germany - Hamburg
Tobias, Tobias Krais wrote: just now my OOo had a dealock again. And I followed your advice. Here is the comandline output: thanks for taking the time :-) -%- [126] [EMAIL PROTECTED] soffice -env:UNO_JAVA_JFW_JREHOME=file:///usr/lib/jvm/java-1.5.0-sun-1.5.0.06/

Re: [api-dev] Restarting a frozen OOo

2006-06-14 Thread Kay Ramme - Sun Germany - Hamburg
Tobias, Tobias Krais wrote: Debian Sid / Windows At least on Linux you should be able to get meaningful stack traces. Yesterday I tried many thing to get a deadlock and really bad things (deleting OOo program files,...) but OOo continued working. Then I gave up and just reinstalled and

Re: [api-dev] Restarting a frozen OOo

2006-06-13 Thread Kay Ramme - Sun Germany - Hamburg
Tobias, Tobias Krais wrote: Hi together, sometimes I have the effect that OOo is frozen. I can't use it anymore and it does not repaint itself anymore. If I now want to use the Java API call Bootstrap.bootstrap() this process hangs and does not react. Seems that it has deadlocked ... Is

Re: [api-dev] Restarting a frozen OOo

2006-06-13 Thread Stephan Bergmann
Kay Ramme - Sun Germany - Hamburg wrote: Tobias, Tobias Krais wrote: Hi together, sometimes I have the effect that OOo is frozen. I can't use it anymore and it does not repaint itself anymore. If I now want to use the Java API call Bootstrap.bootstrap() this process hangs and does not react.

Re: [api-dev] Restarting a frozen OOo

2006-06-13 Thread Tobias Krais
Hi Stephan, sometimes I have the effect that OOo is frozen. I can't use it anymore and it does not repaint itself anymore. If I now want to use the Java API call Bootstrap.bootstrap() this process hangs and does not react. Seems that it has deadlocked ... FYI:

Re: [api-dev] Restarting a frozen OOo

2006-06-13 Thread Tobias Krais
Hi Kay, sometimes I have the effect that OOo is frozen. I can't use it anymore and it does not repaint itself anymore. If I now want to use the Java API call Bootstrap.bootstrap() this process hangs and does not react. Seems that it has deadlocked ... Is there an API call that kills my

[api-dev] Restarting a frozen OOo

2006-06-12 Thread Tobias Krais
Hi together, sometimes I have the effect that OOo is frozen. I can't use it anymore and it does not repaint itself anymore. If I now want to use the Java API call Bootstrap.bootstrap() this process hangs and does not react. Is there an API call that kills my running OOo without having a XDesktop