Andrew.Findlay@brunel.ac.uk (Andrew Findlay) (05/07/91)
|The problem is in the Dombuild script; it seems that there is simply too |much data for awk to process, and it crashes with a SEGV. I suggest you get GNU awk (gawk) and use that - the limits are much bigger or non-exitant. Host: uk.ac.ic.doc.src User: guest Passwd: your e-mail address File: <PUB>gnu/gawk.tar.Z Mode: BINARY The file <PUB>gnu/gawk.ps.tar.Z may also be useful Andrew -- --------------------------------------------------------------------- | From Andrew Findlay at Brunel University, Uxbridge, UB8 3PH, UK | | Andrew.Findlay@brunel.ac.uk phone: +44 895 74000 x2512 | ---------------------------------------------------------------------
sysnet@central1.lancaster.ac.uk ("The Postmaster") (05/07/91)
I'm having awful trouble porting UK-Sendmail 2.1 to a Sequent ptx machine. The problem is in the Dombuild script; it seems that there is simply too much data for awk to process, and it crashes with a SEGV. The point of fail comes where it tries to do a split on domain[rhs]. Depending on what other strings are in the program, and whether the data is in a file or sent to awk on standard input, the failure point is at around 300 elements. Best I can get is that domain[uk.ac] will split with 302 elements, but if that works awk crashes doing domain[uk.co]. Anyone else have this problem? What's the remedy? I've tried deleting elements of the arrays as they're finished with, but that doesn't seem to improve things. Is the size of the NRS such that awk just can't cope, and should I rewrite the stuff in C? I've noticed that on our Sequent Dynix 3 system (BSD 4.2) the awk script is _very_ close to the point of fail too..... Any advice appreciated..... Alan Phillips