AnsweredAssumed Answered

VRf-VEE_Lockup_Bug

Question asked by VRFuser on Jul 27, 1996
from: Greg Goebel / HP-MXD
      gvg@lvld.hp.com / 970-679-3030 / FAX 970-679-5971
to:   VRf
cc:   Noli Liron / Avraham Hulin
date: Sunday, 28 July 1996 1249 MDT

Concerning the recent reports of lockup problems on VEE 3.2:

  -------------------------------------------------------------------------

> I've detected a strange thing that seems to be an apparently random
> problem with repeat objects (at least 'For Count' and 'Until Break').
> What I've seen is that sometimes the thread which is hosted by one of
> this object is executed only once beacause the repeat object does not
> fire again its output pin when all the thread has been executed.
> This behaviour use to occur after a programme modification.
> I solve it by replacing the repeat object by a new one of the same type.
>
> Does anyone detect something similar or have any explanation ?

-- we received several reports of somewhat similar behavior, associated with
VEE-UX ... the reports make it difficult to understand if this is a VEE 3.2
problem only or also afflicts VEE 3.12. 

Another customer elaborated with:

> We found out a strange with a very-very-very simple program
> with an 'Until-Break' object AND 'Animate On': After a apperently
> random number of itterations (min detected = 2, max about 700) the program
> hangs. With 'Control-C' or 'kill -2' it does recover.
> (We use vee 3.12 on hpux 9.05 on a 9000/715 machine)

The lab people here at HP-MXD found out it will happen at a low level of
incidence on VEE-UX as well.

  -------------------------------------------------------------------------

The lab has identified a lockup problem in the code that could account for
this behavior.  It will be corrected in the next release, but that is not
imminent and in fact I cannot offer a schedule for it as of this time.

Any comments are welcome.

[<>] regards -- gvg

Outcomes