The bug was reported 2 months ago[1], I think this patch is ok. [1]http://code.google.com/p/cherokee/issues/detail?id=464
On Mon, Jul 27, 2009 at 3:45 AM, Stuart Henderson<st...@openbsd.org> wrote: > On 2009/07/24 23:55, Matthias Kilian wrote: >> On Fri, Jul 24, 2009 at 11:32:59PM +0200, Matthias Kilian wrote: >> > The portable way would be to enlarge the buffer and retry whenever >> > you get ERANGE. I ran into this problem several weeks ago for GHC >> > (and a diff doing the try/enlarge/retry-game had been accepted >> > upstream). IIRC, there even was some #ifdef'ing around that already >> > increased the buffer especially for OpenBSD, >> >> To clarify, read: increased the *statically sized buffer* especially >> for OpenBSD. >> >> From libraries/unix/System/Posix/Unix.hsc (as in ghc-6.10): >> >> #if defined(HAVE_SYSCONF) && defined(HAVE_SC_GETGR_R_SIZE_MAX) >> grBufSize = sysconfWithDefault 2048 (#const _SC_GETGR_R_SIZE_MAX) >> #else >> grBufSize = 2048 -- just assume some value (1024 is too small on >> OpenBSD) >> #endif >> >> Anyway, on an OS without any limit on possible group members, no >> constant value nor any value returned by sysconf(3) is guaranteed >> to be large enough for getgrnam_r(3). > > I think upstream should probably implement something like that, > but setting to a static 8k is good enough to unbreak the port on > all OpenBSD arch, so unless there are objections I'll commit > that this evening. > > -- -------------- Fernando Quintero http://nonroot.blogspot.com/ *Just a nonroot User*
