summaryrefslogtreecommitdiff
path: root/src/netlink.c
blob: fe979e321c5fb7bd1f0ac02e35763f5d4ceb37bc (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
/*
 * (C) 2006-2011 by Pablo Neira Ayuso <pablo@netfilter.org>
 * (C) 2011 by Vyatta Inc. <http://www.vyatta.com>
 *
 * This program is free software; you can redistribute it and/or modify
 * it under the terms of the GNU General Public License as published by
 * the Free Software Foundation; either version 2 of the License, or
 * (at your option) any later version.
 *
 * This program is distributed in the hope that it will be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 * GNU General Public License for more details.
 *
 * You should have received a copy of the GNU General Public License
 * along with this program; if not, write to the Free Software
 * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
 */

#include "netlink.h"
#include "conntrackd.h"
#include "filter.h"
#include "log.h"

#include <string.h>
#include <errno.h>
#include <sys/types.h>
#include <sys/socket.h>
#include <sys/fcntl.h>
#include <libnetfilter_conntrack/libnetfilter_conntrack_tcp.h>

struct nfct_handle *nl_init_event_handler(void)
{
	struct nfct_handle *h;

	h = nfct_open(CONFIG(netlink).subsys_id, CONFIG(netlink).groups);
	if (h == NULL)
		return NULL;

	if (CONFIG(netlink).events_reliable) {
		int on = 1;

		setsockopt(nfct_fd(h), SOL_NETLINK,
			   NETLINK_BROADCAST_SEND_ERROR, &on, sizeof(int));

		setsockopt(nfct_fd(h), SOL_NETLINK,
			   NETLINK_NO_ENOBUFS, &on, sizeof(int));

		dlog(LOG_NOTICE, "reliable ctnetlink event delivery "
				 "is ENABLED.");
	}

	if (STATE(filter)) {
		if (CONFIG(filter_from_kernelspace)) {
			if (nfct_filter_attach(nfct_fd(h),
					       STATE(filter)) == -1) {
				dlog(LOG_ERR, "cannot set event filtering: %s",
				     strerror(errno));
			}
			dlog(LOG_NOTICE, "using kernel-space event filtering");
		} else
			dlog(LOG_NOTICE, "using user-space event filtering");

		nfct_filter_destroy(STATE(filter));
	}

	fcntl(nfct_fd(h), F_SETFL, O_NONBLOCK);

	/* set up socket buffer size */
	if (CONFIG(netlink_buffer_size) &&
	    CONFIG(netlink_buffer_size) <=
			CONFIG(netlink_buffer_size_max_grown)) {
		/* we divide netlink_buffer_size by 2 here since value passed
		   to kernel gets doubled in SO_RCVBUF; see net/core/sock.c */
		CONFIG(netlink_buffer_size) =
		  nfnl_rcvbufsiz(nfct_nfnlh(h), CONFIG(netlink_buffer_size)/2);
	} else {
		dlog(LOG_NOTICE, "NetlinkBufferSize is either not set or "
				 "is greater than NetlinkBufferSizeMaxGrowth. "
				 "Using current system buffer size");

		socklen_t socklen = sizeof(unsigned int);
		unsigned int read_size;

		/* get current buffer size */
		getsockopt(nfct_fd(h), SOL_SOCKET,
			   SO_RCVBUF, &read_size, &socklen);

		CONFIG(netlink_buffer_size) = read_size;
	}

	dlog(LOG_NOTICE, "netlink event socket buffer size has been set "
			 "to %u bytes", CONFIG(netlink_buffer_size));

	return h;
}

struct nlif_handle *nl_init_interface_handler(void)
{
	struct nlif_handle *h;
	h = nlif_open();
	if (h == NULL)
		return NULL;

	if (nlif_query(h) == -1) {
		free(h);
		return NULL;
	}
	fcntl(nlif_fd(h), F_SETFL, O_NONBLOCK);

	return h;
}

static int warned = 0;

void nl_resize_socket_buffer(struct nfct_handle *h)
{
	unsigned int s = CONFIG(netlink_buffer_size);

	/* already warned that we have reached the maximum buffer size */
	if (warned)
		return;

	/* since sock_setsockopt in net/core/sock.c doubles the size of socket
	   buffer passed to it using nfnl_rcvbufsiz, only call nfnl_rcvbufsiz
	   if new value is not greater than netlink_buffer_size_max_grown */
	if (s*2 > CONFIG(netlink_buffer_size_max_grown)) {
		dlog(LOG_WARNING,
		     "netlink event socket buffer size cannot "
		     "be doubled further since it will exceed "
		     "NetlinkBufferSizeMaxGrowth. We are likely to "
		     "be losing events, this may lead to "
		     "unsynchronized replicas. Please, consider "
		     "increasing netlink socket buffer size via "
		     "NetlinkBufferSize and "
		     "NetlinkBufferSizeMaxGrowth clauses in "
		     "conntrackd.conf");
		warned = 1;
		return;
	}

	CONFIG(netlink_buffer_size) = nfnl_rcvbufsiz(nfct_nfnlh(h), s);

	/* notify the sysadmin */
	dlog(LOG_NOTICE, "netlink event socket buffer size has been doubled "
			 "to %u bytes", CONFIG(netlink_buffer_size));
}

int nl_dump_conntrack_table(struct nfct_handle *h)
{
	return nfct_query(h, NFCT_Q_DUMP, &CONFIG(family));
}

int nl_flush_conntrack_table(struct nfct_handle *h)
{
	return nfct_query(h, NFCT_Q_FLUSH, &CONFIG(family));
}

int nl_send_resync(struct nfct_handle *h)
{
	int family = CONFIG(family);
	return nfct_send(h, NFCT_Q_DUMP, &family);
}

/* if the handle has no callback, check for existence, otherwise, update */
int nl_get_conntrack(struct nfct_handle *h, const struct nf_conntrack *ct)
{
	int ret = 1;
	struct nf_conntrack *tmp;

	tmp = nfct_new();
	if (tmp == NULL)
		return -1;

	/* use the original tuple to check if it is there */
	nfct_copy(tmp, ct, NFCT_CP_ORIG);

	if (nfct_query(h, NFCT_Q_GET, tmp) == -1)
		ret = (errno == ENOENT) ? 0 : -1;

	nfct_destroy(tmp);
	return ret;
}

int nl_create_conntrack(struct nfct_handle *h, 
			const struct nf_conntrack *orig,
			int timeout)
{
	int ret;
	struct nf_conntrack *ct;

	ct = nfct_clone(orig);
	if (ct == NULL)
		return -1;

	if (timeout > 0)
		nfct_set_attr_u32(ct, ATTR_TIMEOUT, timeout);

	/* we hit error if we try to change the expected bit */
	if (nfct_attr_is_set(ct, ATTR_STATUS)) {
		uint32_t status = nfct_get_attr_u32(ct, ATTR_STATUS);
		status &= ~IPS_EXPECTED;
		nfct_set_attr_u32(ct, ATTR_STATUS, status);
	}

	nfct_setobjopt(ct, NFCT_SOPT_SETUP_REPLY);

	/* disable TCP window tracking for recovered connections if required */
	if (nfct_attr_is_set(ct, ATTR_TCP_STATE)) {
		uint8_t flags = IP_CT_TCP_FLAG_SACK_PERM;

		if (!CONFIG(sync).tcp_window_tracking)
			flags |= IP_CT_TCP_FLAG_BE_LIBERAL;
		else
			flags |= IP_CT_TCP_FLAG_WINDOW_SCALE;

		/* FIXME: workaround, we should send TCP flags in updates */
		if (nfct_get_attr_u8(ct, ATTR_TCP_STATE) >=
						TCP_CONNTRACK_TIME_WAIT) {
			flags |= IP_CT_TCP_FLAG_CLOSE_INIT;
		}
		nfct_set_attr_u8(ct, ATTR_TCP_FLAGS_ORIG, flags);
		nfct_set_attr_u8(ct, ATTR_TCP_MASK_ORIG, flags);
		nfct_set_attr_u8(ct, ATTR_TCP_FLAGS_REPL, flags);
		nfct_set_attr_u8(ct, ATTR_TCP_MASK_REPL, flags);
	}

	ret = nfct_query(h, NFCT_Q_CREATE, ct);
	nfct_destroy(ct);

	return ret;
}

int nl_update_conntrack(struct nfct_handle *h,
			const struct nf_conntrack *orig,
			int timeout)
{
	int ret;
	struct nf_conntrack *ct;

	ct = nfct_clone(orig);
	if (ct == NULL)
		return -1;

	if (timeout > 0)
		nfct_set_attr_u32(ct, ATTR_TIMEOUT, timeout);

	/* unset NAT info, otherwise we hit error */
	nfct_attr_unset(ct, ATTR_SNAT_IPV4);
	nfct_attr_unset(ct, ATTR_DNAT_IPV4);
	nfct_attr_unset(ct, ATTR_SNAT_PORT);
	nfct_attr_unset(ct, ATTR_DNAT_PORT);

	if (nfct_attr_is_set(ct, ATTR_STATUS)) {
		uint32_t status = nfct_get_attr_u32(ct, ATTR_STATUS);
		status &= ~IPS_NAT_MASK;
		nfct_set_attr_u32(ct, ATTR_STATUS, status);
	}
	/* we have to unset the helper to avoid EBUSY in reset timers */
	if (nfct_attr_is_set(ct, ATTR_HELPER_NAME))
		nfct_attr_unset(ct, ATTR_HELPER_NAME);

	/* we hit error if we try to update the master conntrack */
	if (ct_is_related(ct)) {
		nfct_attr_unset(ct, ATTR_MASTER_L3PROTO);
		nfct_attr_unset(ct, ATTR_MASTER_L4PROTO);
		nfct_attr_unset(ct, ATTR_MASTER_IPV4_SRC);
		nfct_attr_unset(ct, ATTR_MASTER_IPV4_DST);
		nfct_attr_unset(ct, ATTR_MASTER_IPV6_SRC);
		nfct_attr_unset(ct, ATTR_MASTER_IPV6_DST);
		nfct_attr_unset(ct, ATTR_MASTER_PORT_SRC);
		nfct_attr_unset(ct, ATTR_MASTER_PORT_DST);
	}

	/* disable TCP window tracking for recovered connections if required */
	if (nfct_attr_is_set(ct, ATTR_TCP_STATE)) {
		uint8_t flags = IP_CT_TCP_FLAG_SACK_PERM;

		if (!CONFIG(sync).tcp_window_tracking)
			flags |= IP_CT_TCP_FLAG_BE_LIBERAL;
		else
			flags |= IP_CT_TCP_FLAG_WINDOW_SCALE;

		/* FIXME: workaround, we should send TCP flags in updates */
		if (nfct_get_attr_u8(ct, ATTR_TCP_STATE) >=
						TCP_CONNTRACK_TIME_WAIT) {
			flags |= IP_CT_TCP_FLAG_CLOSE_INIT;
		}
		nfct_set_attr_u8(ct, ATTR_TCP_FLAGS_ORIG, flags);
		nfct_set_attr_u8(ct, ATTR_TCP_MASK_ORIG, flags);
		nfct_set_attr_u8(ct, ATTR_TCP_FLAGS_REPL, flags);
		nfct_set_attr_u8(ct, ATTR_TCP_MASK_REPL, flags);
	}

	ret = nfct_query(h, NFCT_Q_UPDATE, ct);
	nfct_destroy(ct);

	return ret;
}

int nl_destroy_conntrack(struct nfct_handle *h, const struct nf_conntrack *ct)
{
	return nfct_query(h, NFCT_Q_DESTROY, ct);
}

int nl_create_expect(struct nfct_handle *h, const struct nf_expect *orig,
		     int timeout)
{
	int ret;
	struct nf_expect *exp;

	exp = nfexp_clone(orig);
	if (exp == NULL)
		return -1;

	if (timeout > 0)
		nfexp_set_attr_u32(exp, ATTR_EXP_TIMEOUT, timeout);

	ret = nfexp_query(h, NFCT_Q_CREATE, exp);
	nfexp_destroy(exp);

	return ret;
}

int nl_destroy_expect(struct nfct_handle *h, const struct nf_expect *exp)
{
	return nfexp_query(h, NFCT_Q_DESTROY, exp);
}

/* if the handle has no callback, check for existence, otherwise, update */
int nl_get_expect(struct nfct_handle *h, const struct nf_expect *exp)
{
	int ret = 1;
	struct nf_expect *tmp;

	/* XXX: we only need the expectation, not the mask and the master. */
	tmp = nfexp_clone(exp);
	if (tmp == NULL)
		return -1;

	if (nfexp_query(h, NFCT_Q_GET, tmp) == -1)
		ret = (errno == ENOENT) ? 0 : -1;

	nfexp_destroy(tmp);
	return ret;
}

int nl_dump_expect_table(struct nfct_handle *h)
{
	return nfexp_query(h, NFCT_Q_DUMP, &CONFIG(family));
}

int nl_flush_expect_table(struct nfct_handle *h)
{
	return nfexp_query(h, NFCT_Q_FLUSH, &CONFIG(family));
}

int nl_send_expect_resync(struct nfct_handle *h)
{
	int family = CONFIG(family);
	return nfexp_send(h, NFCT_Q_DUMP, &family);
}