> It looks like the gump run did not work on lsd last night. I did not
> have a look to find out what has gone on.
> (the html pages on lsd are of yesterday, the email all dressed up with
> nowhere to go did not get sent either)
Yeah, it seems not. I don't know if this is related, but I had woes trying
to run/test Gump on my VM server Gump-box yesterday also. Basically the poor
box swapped itself to death, in part 'cos forrest grew so large. The xdocs
were written, but the forrest site wasn't generated. I wonder if something
similar happened (or is happening) to LSD. [I just tried logging in, and I'm
not getting in -- due to timeouts.]
I added a page yesterday, not small, not obscenely large, but maybe that
pushed forrest over the edge. I've not done an analysis but it seems that
there is either a memory leak as forrest generates the pages (I saw it had
started generating, but hung.)
I hate to say it, but sometime we really 'pay' (in terms of workarounds,
outages, time delays) for the decisions to go with forrest. I don't know if
we ought consider it 'yet another Gump stress test' and keep it, but I am
starting to think more and more of writing HTML pages directly, via some
Python code.
Talking of stress, gump.dotnot is taking almost a day...
http://gump.dotnot.org/#Details
I'm thinking of giving Gump a shot on moof. Do you think this is a good
idea, or just going to damage another server?
regards,
Adam
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
