On Fri, Jun 21, 2019 at 09:30:36 +0200, Pavel Hrdina wrote:
> In libssh 0.9.0 functions ssh_is_server_known and ssh_write_knownhost
> are marked as deprecated.
>
> Resolves:
https://bugzilla.redhat.com/show_bug.cgi?id=1722735
>
> Signed-off-by: Pavel Hrdina <phrdina(a)redhat.com>
> ---
> m4/virt-libssh.m4 | 8 ++++++++
> src/rpc/virnetlibsshsession.c | 4 ++--
> 2 files changed, 10 insertions(+), 2 deletions(-)
>
> diff --git a/m4/virt-libssh.m4 b/m4/virt-libssh.m4
> index 01c3b75c72..132447da16 100644
> --- a/m4/virt-libssh.m4
> +++ b/m4/virt-libssh.m4
> @@ -33,6 +33,14 @@ AC_DEFUN([LIBVIRT_CHECK_LIBSSH],[
> [],
> [AC_DEFINE_UNQUOTED([ssh_get_server_publickey], [ssh_get_publickey],
> [ssh_get_publickey is deprecated and replaced by
ssh_get_server_publickey.])])
> + AC_CHECK_FUNC([ssh_session_is_known_server],
> + [],
> + [AC_DEFINE_UNQUOTED([ssh_session_is_known_server], [ssh_is_server_known],
> + [ssh_is_server_known is deprecated and replaced by
ssh_session_is_known_server.])])
> + AC_CHECK_FUNC([ssh_session_update_known_hosts],
> + [],
> + [AC_DEFINE_UNQUOTED([ssh_session_update_known_hosts], [ssh_write_knownhost],
> + [ssh_write_knownhost is deprecated and replaced by
ssh_session_update_known_hosts.])])
> CFLAGS="$old_CFLAGS"
> LIBS="$old_LIBS"
I'm not entirely a fan of this. I'd probably prefer defining a macro ...
> fi
> diff --git a/src/rpc/virnetlibsshsession.c b/src/rpc/virnetlibsshsession.c
> index 486437e7bf..093ac29071 100644
> --- a/src/rpc/virnetlibsshsession.c
> +++ b/src/rpc/virnetlibsshsession.c
> @@ -284,7 +284,7 @@ virNetLibsshCheckHostKey(virNetLibsshSessionPtr sess)
> if (sess->hostKeyVerify == VIR_NET_LIBSSH_HOSTKEY_VERIFY_IGNORE)
> return 0;
... which would select both the old and new impl here:
#if defined HAS_NEW_LIBSSH
state = ssh_session_is_known_server(sess->session);
#else
- state = ssh_is_server_known(sess->session);
#endif
This makes it shady and obscure.
Solid point, I've already pushed it so we can fix it as follow-up
together with the preexisting ssh_get_server_publickey.
I checked that the only other case where we follow this logic is to
workaround gnulib and kerberos symbol clash.
Pavel