Chimera on my Red Hat system, somewhat understandably but not acceptably
in my opinion, dumps core on this PDB file.
In CASP, if you can't define a part of the model you submit it with
coordinates set to 0. I wonder if
Chimera is trying to divide by zero (although I would except a SIGFPE,
not a segmentation fault).
I did a remote-display from socrates and it had the problem as well;
there, I at least got a popup window
saying "_chimera: unknown exception" and was able to see this in the
reply log:
Error processing trigger "addModel":
_chimera error: can't normalize zero length array
(see reply log for Python traceback info)
Traceback (most recent call last):
File "/usr/local/chimera/share/__main__.py", line 56, in ?
value = chimeraInit.init(sys.argv)
File "/usr/local/chimera/share/chimeraInit.py", line 240, in init
prefixableType=1)
File "/usr/local/chimera/share/chimera/__init__.py", line 628, in open
skipViewAll=skipViewAll)
File "/usr/local/chimera/share/chimera/__init__.py", line 509, in add
self.triggers.activateTrigger(self.ADDMODEL, models)
File "/usr/local/chimera/share/chimera/triggerSet.py", line 198, in
activateTrigger
self._triggerDict[name].activate(data)
File "/usr/local/chimera/share/chimera/triggerSet.py", line 120, in
activate
ret = handler.invoke(self._name, triggerData)
File "/usr/local/chimera/share/chimera/triggerSet.py", line 82, in invoke
self._funcData, triggerData)
File "/usr/local/chimera/share/chimera/__init__.py", line 837, in
checkKsdssp
model.computeSecondaryStructure(-0.5, 5, 4)
_chimera error: can't normalize zero length array
shortly after visiting the window a few more times, chimera segfaults.
Dave
