The following reply was made to PR os-windows/1726; it has been noted by GNATS.
From: "John Hawthorne" <[EMAIL PROTECTED]> To: Marc Slemko <[EMAIL PROTECTED]>, John Hawthorne <[EMAIL PROTECTED]> Cc: [EMAIL PROTECTED] Subject: Re: os-windows/1726: Apache hangs Date: Fri, 23 Jan 1998 15:02:59 -0800 Hi Marc. This may indeed be the problem. When will 1.3b4 be available? If it's going to be awhile, is there any way I can put in the fix myself? Thanks for your quick response, John On Jan 23, 3:57pm, Marc Slemko wrote: > Subject: Re: os-windows/1726: Apache hangs > On 23 Jan 1998, John Hawthorne wrote: > > > > > >Number: 1726 > > >Category: os-windows > > >Synopsis: Apache hangs > > >Confidential: no > > >Severity: critical > > >Priority: medium > > >Responsible: apache > > >State: open > > >Class: sw-bug > > >Submitter-Id: apache > > >Arrival-Date: Fri Jan 23 14:50:01 PST 1998 > > >Last-Modified: > > >Originator: [EMAIL PROTECTED] > > >Organization: > > apache > > >Release: 1.3b3 > > >Environment: > > NT 4.0 Workstation. > > Apache is not in default directory (I cannot get it started as a service.) > > Apache started as Administrator user > > (Pentium120, 64MB RAM) > > >Description: > > I suspect this is the same problem described in PR1607 and PR1129. Apache > > hangs on a request about once per day. The only way to get any further is > > to kill the process. The problem is intermittent and does not seem to depend > > on the script in question, so it is difficult to reproduce. However, the > > problem is basically making Apache unusable as a server for us. (Please help, > > we would much rather use Apache than IIS...) > > > > Note: I could not use your pre-built binary, it generates 500 errors: premature > > end of script... (So I had to make the executable myself.) > > >How-To-Repeat: > > Sorry. It is sporadic. Just try a lot of scripts over and over again? > > Are you sure it is not simply after so many CGI scripts have been run? In > 1.3b3 there is a file descriptor leak. Check the number of file handles > used in the performance monitor and you will probably find that they are > increasing until you eventually run out of them. > > If this is the problem, it has been fixed in the source tree and the fix > will be avaiable in 1.3b4. > >-- End of excerpt from Marc Slemko
