[Date Prev][Date Next]
Re: Patch: Implement surrogate parent for back-shell (ITS#1815)
The above patch was buggy. Here is a patch to remove it:
If someone later reverses this patch, note:
1. The surrogate parent was started too late. make_surrogate_parent()
should instead be called if #defined(SLAPD_SHELL) before threading
is started in slap_init(), and the surrogate parent should be
killed later if no shell backend is started.
2. The surrogate parent lacks a SIGCHLD signal handler (ITS#2109).
Install main.c:wait4child() in make_surrogate_parent().
3. Check out ITS#2110. Either that's point 1 or 2 above (maybe
running out of processes?), or I don't know what is going on.