From: Zhang Chen <zhangchen.fnst@cn.fujitsu.com>
To: Jason Wang <jasowang@redhat.com>, qemu devel <qemu-devel@nongnu.org>
Cc: Li Zhijian <lizhijian@cn.fujitsu.com>,
Gui jianfeng <guijianfeng@cn.fujitsu.com>,
"eddie.dong" <eddie.dong@intel.com>,
"Dr. David Alan Gilbert" <dgilbert@redhat.com>,
Yang Hongyang <hongyang.yang@easystack.cn>,
zhanghailiang <zhang.zhanghailiang@huawei.com>
Subject: Re: [Qemu-devel] [PATCH V2] net/traffic-mirror:Add traffic-mirror
Date: Thu, 28 Jan 2016 15:44:36 +0800 [thread overview]
Message-ID: <56A9C6E4.2050309@cn.fujitsu.com> (raw)
In-Reply-To: <56A9AAC8.4010803@redhat.com>
On 01/28/2016 01:44 PM, Jason Wang wrote:
>
> On 01/27/2016 10:40 AM, Zhang Chen wrote:
>> From: ZhangChen <zhangchen.fnst@cn.fujitsu.com>
>>
>> Traffic-mirror is a netfilter plugin.
>> It gives qemu the ability to copy and mirror guest's
>> net packet. we output packet to chardev.
>>
>> usage:
>>
>> -netdev tap,id=hn0
>> -chardev socket,id=mirror0,host=ip_primary,port=X,server,nowait
>> -traffic-mirror,id=m0,netdev=hn0,queue=tx/rx/all,outdev=mirror0
>>
>> Signed-off-by: ZhangChen <zhangchen.fnst@cn.fujitsu.com>
>> Signed-off-by: Wen Congyang <wency@cn.fujitsu.com>
>> Reviewed-by: Yang Hongyang <hongyang.yang@easystack.cn>
> Thanks for the patch. Several questions:
>
> - I'm curious about how the patch was tested? Simple setup e.g:
>
> -netdev tap,id=hn0 -device virtio-net-pci,netdev=hn0 -chardev
> socket,id=c0,host=localhost,port=4444,server,nowait -object
> traffic-mirror,netdev=hn0,outdev=c0,id=f0 -netdev
> socket,id=s0,connect=127.0.0.1:4444 -device e1000,netdev=s0
>
> does not works for me.
I test it in this way.
primary:
-netdev tap,id=hn0 -device e1000,netdev=hn0 -chardev
socket,id=mirror0,host=3.3.3.3,port=9003,server,nowait
-object traffic-mirror,id=f0,netdev=hn0,queue=tx,outdev=mirror0
secondary:
-netdev tap,id=hn0 -device e1000,netdev=hn0 -chardev
socket,id=mirror0,host=3.3.3.3,port=9003 -object
traffic-reader,id=f1,netdev=hn0,queue=rx,indev=mirror0
I write a traffic-reader demo to read chardev socket and print it in
monitor.
>
> - Is a reliable mirroring (e.g no packet drops during mirroring) is
> needed for COLO? If yes, this patch seems could not guarantee this.
I will fix it in V3
> - Please consider to write a unit test for this patch.
write a unit test like tests/test-netfilter.c ?
> And see comments below.
>
> Thanks
>
>
>> ---
>> net/Makefile.objs | 1 +
>> net/traffic-mirror.c | 173 +++++++++++++++++++++++++++++++++++++++++++++++++++
>> qemu-options.hx | 5 ++
>> vl.c | 3 +-
>> 4 files changed, 181 insertions(+), 1 deletion(-)
>> create mode 100644 net/traffic-mirror.c
>>
>> diff --git a/net/Makefile.objs b/net/Makefile.objs
>> index 5fa2f97..de06ebe 100644
>> --- a/net/Makefile.objs
>> +++ b/net/Makefile.objs
>> @@ -15,3 +15,4 @@ common-obj-$(CONFIG_VDE) += vde.o
>> common-obj-$(CONFIG_NETMAP) += netmap.o
>> common-obj-y += filter.o
>> common-obj-y += filter-buffer.o
>> +common-obj-y += traffic-mirror.o
> Let's s/traffic-mirror/filter-mirror/g to be consistent with other filters.
>
OK~ I will fix it in V3
>> diff --git a/net/traffic-mirror.c b/net/traffic-mirror.c
>> new file mode 100644
>> index 0000000..bed915c
>> --- /dev/null
>> +++ b/net/traffic-mirror.c
>> @@ -0,0 +1,173 @@
>> +/*
>> + * Copyright (c) 2016 HUAWEI TECHNOLOGIES CO., LTD.
>> + * Copyright (c) 2016 FUJITSU LIMITED
>> + * Copyright (c) 2016 Intel Corporation
>> + *
>> + * Author: Zhang Chen <zhangchen.fnst@cn.fujitsu.com>
>> + *
>> + * This work is licensed under the terms of the GNU GPL, version 2 or
>> + * later. See the COPYING file in the top-level directory.
>> + */
>> +
>> +#include "net/filter.h"
>> +#include "net/net.h"
>> +#include "qemu-common.h"
>> +#include "qapi/qmp/qerror.h"
>> +#include "qapi-visit.h"
>> +#include "qom/object.h"
>> +#include "qemu/main-loop.h"
>> +#include "qemu/error-report.h"
>> +#include "trace.h"
>> +#include "sysemu/char.h"
>> +#include "qemu/iov.h"
>> +
>> +#define FILTER_TRAFFIC_MIRROR(obj) \
>> + OBJECT_CHECK(MirrorState, (obj), TYPE_FILTER_TRAFFIC_MIRROR)
>> +
>> +#define TYPE_FILTER_TRAFFIC_MIRROR "traffic-mirror"
>> +
>> +typedef struct MirrorState {
>> + NetFilterState parent_obj;
>> + char *outdev;
>> + CharDriverState *chr_out;
>> +
>> +} MirrorState;
>> +
>> +static ssize_t traffic_mirror_send(NetFilterState *nf,
>> + const struct iovec *iov,
>> + int iovcnt)
>> +{
>> + MirrorState *s = FILTER_TRAFFIC_MIRROR(nf);
>> + ssize_t ret = 0;
>> + ssize_t size = 0;
>> + char *buf;
>> +
>> + size = iov_size(iov, iovcnt);
>> + if (!size) {
>> + return 0;
>> + }
>> +
>> + buf = g_malloc0(size);
>> + iov_to_buf(iov, iovcnt, 0, buf, size);
>> + ret = qemu_chr_fe_write(s->chr_out, (uint8_t *)&size, sizeof(size));
> htonl(size)?
We do not need this.
>> + if (ret < 0) {
> This check is not sufficient, for some reason, only part of the packets
> maybe sent by the socket. Need to handle this properly, otherwise it may
> confuse receiver.
I will fix it in next version.
>> + g_free(buf);
>> + return ret;
>> + }
>> +
>> + ret = qemu_chr_fe_write(s->chr_out, (uint8_t *)buf, size);
>> + g_free(buf);
>> + return ret;
> Ditto.
I will fix it in next version.
>> +}
>> +
>> +static ssize_t traffic_mirror_receive_iov(NetFilterState *nf,
>> + NetClientState *sender,
>> + unsigned flags,
>> + const struct iovec *iov,
>> + int iovcnt,
>> + NetPacketSent *sent_cb)
>> +{
>> + /*
>> + * We copy and mirror packet to outdev,
>> + * then put back the packet.
>> + */
> The code could explain itself, so the comment is unnecessary.
OK,I will remove it.
Thanks
zhangchen
>> + ssize_t ret = 0;
>> +
>> + ret = traffic_mirror_send(nf, iov, iovcnt);
>> + if (ret < 0) {
>> + error_report("traffic_mirror_send failed");
> Monitor could be flooded by this.
>
>> + }
>> +
>> + return 0;
>> +}
>> +
> Other looks good.
>
>
>
> .
>
--
Thanks
zhangchen
next prev parent reply other threads:[~2016-01-28 7:45 UTC|newest]
Thread overview: 12+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-01-27 2:40 [Qemu-devel] [PATCH V2] net/traffic-mirror:Add traffic-mirror Zhang Chen
2016-01-27 9:23 ` Hailiang Zhang
2016-01-28 5:44 ` Jason Wang
2016-01-28 7:44 ` Zhang Chen [this message]
2016-01-28 8:37 ` Jason Wang
2016-01-28 9:52 ` Zhang Chen
2016-01-28 10:06 ` Jason Wang
2016-01-29 1:38 ` Li Zhijian
2016-02-01 2:57 ` Jason Wang
2016-02-01 7:50 ` Li Zhijian
2016-02-01 9:11 ` Dr. David Alan Gilbert
2016-02-01 9:42 ` Li Zhijian
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=56A9C6E4.2050309@cn.fujitsu.com \
--to=zhangchen.fnst@cn.fujitsu.com \
--cc=dgilbert@redhat.com \
--cc=eddie.dong@intel.com \
--cc=guijianfeng@cn.fujitsu.com \
--cc=hongyang.yang@easystack.cn \
--cc=jasowang@redhat.com \
--cc=lizhijian@cn.fujitsu.com \
--cc=qemu-devel@nongnu.org \
--cc=zhang.zhanghailiang@huawei.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.