Re: [Webware-discuss] Dead threads: detection and recovery

2006-06-19 Thread Oliver Bock
Shayne O'Neill wrote: > Does python have a sort of cooperative "pass" command? Whilst little > (other than an external monitoring process, assuming the monitor gets a > chance to do its thing) will stop someone writing this code;- > > while 1==1: > echo "hi mum" > > a call like > > whi

Re: [Webware-discuss] Dead threads: detection and recovery

2006-06-19 Thread Oliver Bock
Ben Parker wrote: > ... If management is > pushing for those kinds of development capabilities in order to drive > their business, you should push back that their requirements demand more > knowledgeable staff. > While I agree with your suggestions, they unfortunately won't fly. Oliver

Re: [Webware-discuss] Dead threads: detection and recovery

2006-06-19 Thread Ben Parker
Oliver Bock wrote on 06/19/2006 05:45 PM: > I think I should explain the system: the software allows the design and > execution of web surveys. The people writing code are constructing > surveys and sometimes need to write Python fragments to decide (for > example) which question is shown next.

Re: [Webware-discuss] Dead threads: detection and recovery

2006-06-19 Thread Shayne O'Neill
Does python have a sort of cooperative "pass" command? Whilst little (other than an external monitoring process, assuming the monitor gets a chance to do its thing) will stop someone writing this code;- while 1==1: echo "hi mum" a call like while 1==1: one_of_our_calls()

Re: [Webware-discuss] Dead threads: detection and recovery

2006-06-19 Thread Oliver Bock
I completely take your point that you cannot stop malicious users abusing scripting facilities. (Although if it weren't for the import statement, I think you could provide a limited set of functions in __globals__ and limit the abuse to overuse of the CPU.) I think I should explain the system:

Re: [Webware-discuss] Dead threads: detection and recovery

2006-06-19 Thread Shayne O'Neill
Or to clarify I mean you cant offer scripting to users, and make it so it cant be abused. One of my old projects was to embed python in the citadel groupware server [fun fact, I got webware working from inside the citadel server] and yeah, couldnt make it secure at all, so ended up with signed

Re: [Webware-discuss] Dead threads: detection and recovery

2006-06-19 Thread Shayne O'Neill
Hmm. Ok. What Im getting at, is that Im presuming that these scripts are at an administrative level? Python absolutely can not be made secure. There where attempts a while back at it, but it was found that it simply cant be done. Oh yeah, totally offtopic, but regarding the 'penis enhancing dr

Re: [Webware-discuss] Dead threads: detection and recovery

2006-06-19 Thread Oliver Bock
> > Without wanting to be rude, but is there a point to the exercise? Of course not. I've got nothing better to do than start mailing list threads. On my weekends I go around and post comments to blogs, advertising penis-enlarging drugs. > ... If this > is a problem then one would presume th

Re: [Webware-discuss] Dead threads: detection and recovery

2006-06-19 Thread Shayne O'Neill
Without wanting to be rude, but is there a point to the exercise? If this is a problem then one would presume that they can just restart the system if it locks and check out the stack dump? On Mon, 19 Jun 2006, Oliver Bock wrote: > My system allows users to write their own python code fragments

[Webware-discuss] Dead threads: detection and recovery

2006-06-19 Thread Oliver Bock
My system allows users to write their own python code fragments for some tasks. Unfortunately the users sometimes write infinite loops, which gradually lock up threads until none remain. A few minutes later a monitor notices that the system is unresponsive and kills and restarts AppServer. T