Hi,
I am working my way through the cisTEM tutorial. Everything seemed to be working well, but after I started the 2D Classification step, the GUI crashed and I am getting a continuous stream of "Connection Timer Fired" messages to the terminal where I started the cisTEM GUI. Any ideas as to what might have happened and/or how to fix this (and stop the error messages)?
I am not sure what happened to cause the crash, it sounds like it crashed during the connection. Is this with the pre-compiled binaries or the ones you compiled?
If you look, there should be a cisTEM_job_control process running, it is that process that is printing these errors, and they will stop if you kill it.
Cheers,
Tim
Tim,
Thanks. That is with the binaries I compiled here. I will try the pre-compiled ones, but I wanted to kill whatever was generating that error message before trying anything else. I will let you know what happens.
Tim,
I was able to finish the tutorial using the pre-compiled binaries. I will run another test some time to determine whether it was just a fluke of the system or something wrong with the binaries I compiled here. Thanks for your help.
Hi Tim,
I'd like to echo David's original error to say that sometimes I get an endless stream of "Connection Timer Fired" from my SGE cluster. Perhaps it's related to the dataset size, delay time, and the network file system? I've run cisTEM successfully already, but this error appears to be cropping up a lot with a 900K particle dataset.
Any suggestions on dealing with this would be appreciated.
Mike
Hi Mike,
This must mean that the controller is still waiting for some jobs to connect. You should be able to stop this by killing the cisTEM_job_controller process. It is supposed to also die if the gui is closed, or the job is terminated - but perhaps something has gone wrong. I will look into it to see if i can prevent this in future versions.
Thanks,
Tim
Thanks Tim! It makes sense then that logging out/logging back in stopped this issue from happening.