On Monday 13 February 2012, Jim Jagielski wrote: > The 2.4.1 (candidate) tarballs are available for download and test: > > http://httpd.apache.org/dev/dist/ > > I'm calling a VOTE on releasing these as Apache httpd 2.4.1 GA. > NOTE: The -deps tarballs are included here *only* to make life > easier for the tester. They will not be, and are not, part > of the official release. > > [ ] +1: Good to go > [ ] +0: meh > [ ] -1: Danger Will Robinson. And why. > > Vote will last the normal 72 hrs.
Depending on the configuration (mod_lua and mod_cgid enabled, using the unmodified config file from make install), I got t/modules/cgi.t (Wstat: 0 Tests: 58 Failed: 1) Failed test: 54 I think someone else saw this before. The problem detected by the test is that part of the lua translate_name hook source code is logged in the script log. The source of the problem is that an incomplete header read from the cgi is not 0-terminated in the error case. It's probably coincidence that the test only triggers with mod_lua. Maybe there is usually not a long enough string in the buffer to trigger the failure. Fixed in r1244211 / r1244213. I think 2.2 needs the same fix. Does not look like a regression, therefore I would not see this as blocker for GA. But I want to do some more testing before I vote. Cheers, Stefan
