[systemd-devel] [PATCH] add REMOTE_ADDR and REMOTE_PORT for Accept=yes
Shawn Landden
shawn at churchofgit.com
Sun Mar 8 16:25:41 PDT 2015
the xinetd REMOTE_IP is a fedora extension so I think we should avoid it.
On Sun, Mar 8, 2015 at 4:24 PM, Shawn Landden <shawn at churchofgit.com> wrote:
> ---
> TODO | 2 --
> man/systemd.socket.xml | 6 +++++-
> src/core/service.c | 47
> +++++++++++++++++++++++++++++++++++++++++++++++
> 3 files changed, 52 insertions(+), 3 deletions(-)
>
> diff --git a/TODO b/TODO
> index ae32388..780084a 100644
> --- a/TODO
> +++ b/TODO
> @@ -164,8 +164,6 @@ Features:
> * as soon as we have kdbus, and sender timestamps, revisit coalescing
> multiple parallel daemon reloads:
>
> http://lists.freedesktop.org/archives/systemd-devel/2014-December/025862.html
>
> -* set $REMOTE_IP (or $REMOTE_ADDR/$REMOTE_PORT) environment variable when
> doing per-connection socket activation. use format introduced by xinetd or
> CGI for this
> -
> * the install state probably shouldn't get confused by generated units,
> think dbus1/kdbus compat!
>
> * in systemctl list-unit-files: show the install value the presets would
> suggest for a service in a third column
> diff --git a/man/systemd.socket.xml b/man/systemd.socket.xml
> index 3938345..20f1e0c 100644
> --- a/man/systemd.socket.xml
> +++ b/man/systemd.socket.xml
> @@ -357,7 +357,11 @@
> daemons designed for usage with
>
> <citerefentry><refentrytitle>inetd</refentrytitle><manvolnum>8</manvolnum></citerefentry>
> to work unmodified with systemd socket
> - activation.</para></listitem>
> + activation.</para>
> + <para>For IPv4 and IPv6 connections the
> <varname>REMOTE_ADDR</varname>
> + environment variable will be set with remote IP, and
> <varname>REMOTE_PORT</varname>
> + environment variable set to the remote port, similar to CGI
> + (for SOCK_RAW the port is the IP protocol).</para></listitem>
> </varlistentry>
>
> <varlistentry>
> diff --git a/src/core/service.c b/src/core/service.c
> index cc4ea19..6a690ac 100644
> --- a/src/core/service.c
> +++ b/src/core/service.c
> @@ -22,6 +22,7 @@
> #include <errno.h>
> #include <signal.h>
> #include <unistd.h>
> +#include <arpa/inet.h>
>
> #include "async.h"
> #include "manager.h"
> @@ -1119,6 +1120,52 @@ static int service_spawn(
> goto fail;
> }
>
> + if (s->accept_socket.unit) {
> + union sockaddr_union sa;
> + socklen_t salen = sizeof(sa);
> + _cleanup_free_ char *remote_addr = NULL;
> + char a[MAX(INET6_ADDRSTRLEN, INET_ADDRSTRLEN)];
> +
> + r = getpeername(s->socket_fd, &sa.sa, &salen);
> + if (r < 0) {
> + r = -errno;
> + goto fail;
> + }
> +
> + if (sa.sa.sa_family == AF_INET ||
> + sa.sa.sa_family == AF_INET6) {
> + if (inet_ntop(sa.sa.sa_family,
> + /* this field of the API is kinda
> braindead,
> + * should take head of struct so it
> can be passed the union...*/
> + sa.sa.sa_family == AF_INET6 ?
> + &sa.in6.sin6_addr :
> + &sa.in.sin_addr,
> + a, sizeof(a)) == NULL) {
> + r = -errno;
> + goto fail;
> + }
> +
> + if (asprintf(our_env + n_env++,
> + "REMOTE_ADDR=%s",
> + /* musl and glibc inet_ntop()
> present v4-mapped addresses in ::ffff:a.b.c.d form */
> + sa.sa.sa_family == AF_INET6 &&
> strchr(a, '.') ?
> + strempty(startswith(a, "::ffff:"))
> :
> + a) < 0) {
> + r = -ENOMEM;
> + goto fail;
> + }
> +
> + if (asprintf(our_env + n_env++,
> + "REMOTE_PORT=%u",
> + ntohs(sa.sa.sa_family == AF_INET6 ?
> + sa.in6.sin6_port :
> + sa.in.sin_port)) < 0) {
> + r = -ENOMEM;
> + goto fail;
> + }
> + }
> + }
> +
> final_env = strv_env_merge(2, UNIT(s)->manager->environment,
> our_env, NULL);
> if (!final_env) {
> r = -ENOMEM;
> --
> 2.2.1.209.g41e5f3a
>
>
--
Shawn Landden
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.freedesktop.org/archives/systemd-devel/attachments/20150308/ed680c37/attachment.html>
More information about the systemd-devel
mailing list