HylaFAX The world's
most advanced open source fax server
Re: Large number of errors with Hylafax and ZyXEL Omni modem
Well....the machine is certainly busy at that point, but it's a pretty big
machine; it ought to be able to do the whole thing with 128MB of memory
and two SPARC processors. Plus, it's sending six faxes simultaneously of
the same image, so it seems odd that it's getting imaged each time. I
have a PS timeout of like five minutes, and the image takes < 0.5sec to
image (it's pretty simple).
Perhaps some more info would help. Basically, I create one job and then
suspend it. I then "clone" the job to send out 1200 copies of it to our
user community. Of the 1200, maybe 0.5 fail (ie, every couple of days).
So...the question is, can I turn on more specific debugging that says
*why* HylaFax can't open the document file? It's impossible to replicate,
and of course, by the time I look at things, the document file is there.
Not being a real C hacker, I don't know how much debugging I can
comfortable put into the code....
On Sat, 4 Apr 1998, Mike Porter wrote:
> We were seeing errors like this when the volume of FAXing got
> to a critical point. The system didn't have enough memory,
> so it started to thrash horribly. Response time went through
> the roof, let alone the time to run a ghostscript process.
> Solution was to add another 16MB of memory.
> On Fri, 3 Apr 1998, David Woolley wrote:
> > > - Can not open document file - looks like some sort of locking/race
> > > condition, having to do with jobs that share the same document file.
> > An error like this is associated with Postscript interpreter time outs,
> > either because of a Postscript loop, or too short a timeout.