On Thursday 13 March 2003 04.02, Adrian Chadd wrote: > On Wed, Mar 12, 2003, Henrik Nordstrom wrote: > > Hmm.. maybe the comm abort logics is not as fool proof as I > > thought. > > > > Here did I put that patch for adding cbdata fences to comm.. > > probably lost. > > > > But on the other hand this looks like it could be a httpState, > > but slightly stomped by someone else.. hmm.. the data there looks > > suspiciously much like ascii, what do you get if you try to prin > > > > print (char *)httpState > > Excellent call! > > (gdb) print (char *)httpState > $4 = 0xc1b70b0 "5.2e5142af:7fef7ddd\r\nX-Srv-Wid: > 080db06b.613b4319:00000002\r\nHTTP/1.0 200 OK\r\nCache-Control: > private,no-cache\r\nPragma: no-cache\r\nDate: Wed, 12 Mar 2003 > 09:15:55 GMT\r\nConnection: keep-alive\r\nExpires: Th"...
Ok. So it is not a HttpState but a buffer for HTTP headers. Gives three options: a: The pointer was corrupted in comm (not very likely) b: The HttpState has been freed and reused as something else. Adding cbdata fences to comm would eleminate any such risk. c: Something processing headers has a serious buffer overflow (not very likely). Regards Henrik
