And I have another question about multi_process_io_udp lost event.
Is there some problem if we change if{} else if{} to if {} if {}.
e.g.:
diff --git a/src/openvpn/mudp.c b/src/openvpn/mudp.c
index a604d217..18dc009d 100644
--- a/src/openvpn/mudp.c
+++ b/src/openvpn/mudp.c
@@ -220,12 +220,12 @@ multi_process_io_udp(struct multi_context *m)
multi_process_outgoing_link(m, mpp_flags);
}
/* TUN device ready to accept write */
- else if (status & TUN_WRITE)
+ if (status & TUN_WRITE)
{
multi_process_outgoing_tun(m, mpp_flags);
}
/* Incoming data on UDP port */
- else if (status & SOCKET_READ)
+ if (status & SOCKET_READ)
{
read_incoming_link(&m->top);
if (!IS_SIG(&m->top))
@@ -234,7 +234,7 @@ multi_process_io_udp(struct multi_context *m)
}
}
/* Incoming data on TUN device */
- else if (status & TUN_READ)
+ if (status & TUN_READ)
{
read_incoming_tun(&m->top);
if (!IS_SIG(&m->top))
@@ -244,7 +244,7 @@ multi_process_io_udp(struct multi_context *m)
}
#ifdef ENABLE_ASYNC_PUSH
/* INOTIFY callback */
- else if (status & FILE_CLOSED)
+ if (status & FILE_CLOSED)
{
multi_process_file_closed(m, mpp_flags);
}
wei wang <lny...@gmail.com> 于2019年4月22日周一 下午5:26写道：
>
> 1. When client-connect tests whether the user has written the
> configuration, we should check to see if the file size is 0, not
> whether the file exists.
> Because the configuration file is created by openvpn itself.
> 2. multi_client_connect_early_setup should be called for the first time.
>
> diff --git a/src/openvpn/multi.c b/src/openvpn/multi.c
> index 697994b8..27c6e935 100644
> --- a/src/openvpn/multi.c
> +++ b/src/openvpn/multi.c
> @@ -1641,9 +1641,13 @@ multi_client_connect_post(struct multi_context *m,
> const char *dc_file,
> unsigned int *option_types_found)
> {
> - /* Did script generate a dynamic config file? */
> - if (platform_test_file(dc_file))
> - {
> + /* Did script write config to dynamic config file? */
> + platform_stat_t dc_stat = {0};
> + if (platform_stat(dc_file, &dc_stat) < 0) {
> + return;
> + }
> +
> + if (dc_stat.st_size > 0) {
> options_server_import(&mi->context.options,
> dc_file,
> D_IMPORT_ERRORS|M_OPTERR,
> @@ -2474,13 +2478,13 @@ multi_connection_established(struct
> multi_context *m, struct multi_instance *mi)
> /* Initially we have no handler that has returned a result */
> mi->client_connect_status = CC_STATUS_DEFERRED_NO_RESULT;
> from_deferred = false;
> + multi_client_connect_early_setup(m, mi);
> }
> else
> {
> from_deferred = true;
> }
>
> - multi_client_connect_early_setup(m, mi);
>
> bool cc_succeeded = true;
>
> Gert Doering <g...@greenie.muc.de> 于2019年4月20日周六 下午11:07写道：
> >
> > Hi,
> >
> > On Sat, Apr 20, 2019 at 04:37:03PM +0800, wei wang wrote:
> > > I have found that the patch set
> > > "https://patchwork.openvpn.net/project/openvpn2/list/?series=413"; has
> > > implement it.
> > > But it has not been updated for several months. Is there someone still
> > > work on it? We test with it for client-connect script, and find some
> > > bug.
> > > I am very happy to provide if needed.
> >
> > We're working on reviewing, fixing and merging it.
> >
> > So if you have feedback or even found a bug, please send it in (preferrably
> > in reply to the original mail, so patchwork can pick it up, but just here
> > in the thread is good enough).
> >
> > gert
> > --
> > "If was one thing all people took for granted, was conviction that if you
> > feed honest figures into a computer, honest figures come out. Never doubted
> > it myself till I met a computer with a sense of humor."
> > Robert A. Heinlein, The Moon is a Harsh
> > Mistress
> >
> > Gert Doering - Munich, Germany
> > g...@greenie.muc.de
>
>
>
> --
> Regards,
> Wang Wei
--
Regards,
Wang Wei
_______________________________________________
Openvpn-devel mailing list
Openvpn-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/openvpn-devel
