Hi,
thanks for looking into this.
On Mon, 2017-04-10 at 10:16 +0200, Matthias Fischer wrote:
Hi,
Today there was a confirmation of my assumption, the 'configure' option '--with-libevent-support' is NOT enough (unbound-users@unbound.net):
***SNIP***
When building unbound with --with-libevent support, the make install phase should also call make unbound-event-install or else unbound-event.h does not get installed and the header file for using the unbound event functionality is not available.
Hello,
the same "missing unbound-event.h" happen when building getdns
Thanks for the hint, Paul!
This install is triggered by the option --enable-event-api . Just enabling --with-libevent does not trigger the install by itself.
Best regards, Wouter ***SNAP***
This rather sounds to me that any libraries that link against libunbound will only use libevent then. This is probably not a good idea not to abstract this.
The 'libevent' update on the '1.4.15-stable' has unfortunately not built, there were heaps of errors.
Because 'unbound' detects an existing 'libevent2' installation (Changelog, line 1609: Detect libevent2 install automatically by configure.) I first concentrated on updating 'libevent2' to 2.1.8 because '2.0.22' from 2014-01-05 seemed rather old.
The detection worked, 'unbound' with '--with-libevent' and '--enable-event-api' and 'libevent2 2.1.8' run both here in test.
My version is already linked against libevent:
[root@ipfire ~]# ldd /usr/sbin/unbound linux-vdso.so.1 => (0x00006ad33eb06000) libssl.so.10 => /usr/lib/libssl.so.10 (0x00006ad33e67c000) libevent-2.0.so.5 => /usr/lib/libevent-2.0.so.5 (0x00006ad33e434000) librt.so.1 => /lib/librt.so.1 (0x00006ad33e22b000) libcrypto.so.10 => /usr/lib/libcrypto.so.10 (0x00006ad33dde9000) libpthread.so.0 => /lib/libpthread.so.0 (0x00006ad33dbcc000) libc.so.6 => /lib/libc.so.6 (0x00006ad33d836000) libgcc_s.so.1 => /usr/lib/libgcc_s.so.1 (0x00006ad33d620000) libdl.so.2 => /lib/libdl.so.2 (0x00006ad33d41c000) /lib64/ld-linux-x86-64.so.2 (0x00006ad33e8e8000)
Special feature: 'Kbd 1.12' complains because of a few 'rootfile'-changes, otherwise no other conspicuousness. Operation is stable.
This is probably caused by something else... It's already happening in the last nightly build.
Best, Matthias