]> git.proxmox.com Git - pve-docs.git/blob - pve-firewall.adoc
pve-firewall.adoc: move info about pve ports to the end
[pve-docs.git] / pve-firewall.adoc
1 ifdef::manvolnum[]
2 PVE({manvolnum})
3 ================
4 include::attributes.txt[]
5
6 NAME
7 ----
8
9 pve-firewall - The PVE Firewall Daemon
10
11
12 SYNOPSYS
13 --------
14
15 include::pve-firewall.8-synopsis.adoc[]
16
17
18 DESCRIPTION
19 -----------
20 endif::manvolnum[]
21
22 ifndef::manvolnum[]
23 {pve} Firewall
24 ==============
25 include::attributes.txt[]
26 endif::manvolnum[]
27
28 // Copied from pve wiki: Revision as of 08:45, 9 November 2015
29
30 Proxmox VE Firewall provides an easy way to protect your IT
31 infrastructure. You can easily setup firewall rules for all hosts
32 inside a cluster, or define rules for virtual machines and
33 containers. Features like firewall macros, security groups, IP sets
34 and aliases help making that task easier.
35
36 While all configuration is stored on the cluster file system, the
37 iptables based firewall runs on each cluster node, and thus provides
38 full isolation between virtual machines. The distributed nature of
39 this system also provides much higher bandwidth than a central
40 firewall solution.
41
42 NOTE: If you enable the firewall, all traffic is blocked by default,
43 except WebGUI(8006) and ssh(22) from your local network.
44
45 The firewall has full support for IPv4 and IPv6. IPv6 support is fully
46 transparent, and we filter traffic for both protocols by default. So
47 there is no need to maintain a different set of rules for IPv6.
48
49
50 Zones
51 -----
52
53 The Proxmox VE firewall groups the network into the following logical zones:
54
55 Host::
56
57 Traffic from/to a cluster node
58
59 VM::
60
61 Traffic from/to a specific VM
62
63 For each zone, you can define firewall rules for incoming and/or
64 outgoing traffic.
65
66
67 Configuration
68 -------------
69
70 All firewall related configuration is stored on the proxmox cluster
71 file system. So those files are automatically distributed to all
72 cluster nodes, and the 'pve-firewall' service updates the underlying
73 iptables rules automatically on any change. Any configuration can be
74 done using the GUI (i.e. Datacenter -> Firewall -> Options tab (tabs
75 at the bottom of the page), or on a Node -> Firewall), so the
76 following configuration file snippets are just for completeness.
77
78 Cluster wide configuration is stored at:
79
80 /etc/pve/firewall/cluster.fw
81
82 The firewall is completely disabled by default, so you need to set the
83 enable option here:
84
85 ----
86 [OPTIONS]
87 # enable firewall (cluster wide setting, default is disabled)
88 enable: 1
89 ----
90
91 The cluster wide configuration can contain the following data:
92
93 * IP set definitions
94 * Alias definitions
95 * Security group definitions
96 * Cluster wide firewall rules for all nodes
97
98 VM firewall configuration is read from:
99
100 /etc/pve/firewall/<VMID>.fw
101
102 and contains the following data:
103
104 * IP set definitions
105 * Alias definitions
106 * Firewall rules for this VM
107 * VM specific options
108
109 And finally, any host related configuration is read from:
110
111 /etc/pve/nodes/<nodename>/host.fw
112
113 This is useful if you want to overwrite rules from 'cluster.fw'
114 config. You can also increase log verbosity, and set netfilter related
115 options.
116
117 Enabling the Firewall for VMs and Containers
118 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
119
120 You need to enable the firewall on the virtual network interface configuration
121 in addition to the general 'Enable Firewall' option in the 'Options' tab.
122
123 Firewall Rules
124 ~~~~~~~~~~~~~~
125
126 Firewall rules consists of a direction (`IN` or `OUT`) and an
127 action (`ACCEPT`, `DENY`, `REJECT`). You can also specify a macro
128 name. Macros contain predifined sets of rules and options. Rules can be disabled by prefixing them with '|'.
129
130 .Firewall rules syntax
131 ----
132 [RULES]
133
134 DIRECTION ACTION [OPTIONS]
135 |DIRECTION ACTION [OPTIONS] # disabled rule
136
137 DIRECTION MACRO(ACTION) [OPTIONS] # use predefined macro
138 ----
139
140 The following options can be used to refine rule matches.
141
142 include::pve-firewall-rules-opts.adoc[]
143
144 Here are some examples:
145
146 ----
147 [RULES]
148 IN SSH(ACCEPT) -i net0
149 IN SSH(ACCEPT) -i net0 # a comment
150 IN SSH(ACCEPT) -i net0 -source 192.168.2.192 # only allow SSH from 192.168.2.192
151 IN SSH(ACCEPT) -i net0 -source 10.0.0.1-10.0.0.10 # accept SSH for ip range
152 IN SSH(ACCEPT) -i net0 -source 10.0.0.1,10.0.0.2,10.0.0.3 #accept ssh for ip list
153 IN SSH(ACCEPT) -i net0 -source +mynetgroup # accept ssh for ipset mynetgroup
154 IN SSH(ACCEPT) -i net0 -source myserveralias #accept ssh for alias myserveralias
155
156 |IN SSH(ACCEPT) -i net0 # disabled rule
157
158 IN DROP # drop all incoming packages
159 OUT ACCEPT # accept all outgoing packages
160 ----
161
162 Security Groups
163 ~~~~~~~~~~~~~~~
164
165 A security group is a collection of rules, defined at cluster level, which
166 can be used in all VMs' rules. For example you can define a group named
167 `webserver` with rules to open the http and https ports.
168
169 ----
170 # /etc/pve/firewall/cluster.fw
171
172 [group webserver]
173 IN ACCEPT -p tcp -dport 80
174 IN ACCEPT -p tcp -dport 443
175 ----
176
177 Then, you can add this group to a VM's firewall
178
179 ----
180 # /etc/pve/firewall/<VMID>.fw
181
182 [RULES]
183 GROUP webserver
184 ----
185
186
187 IP Aliases
188 ~~~~~~~~~~
189
190 IP Aliases allow you to associate IP addresses of networks with a
191 name. You can then refer to those names:
192
193 * inside IP set definitions
194 * in `source` and `dest` properties of firewall rules
195
196 Standard IP alias `local_network`
197 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
198
199 This alias is automatically defined. Please use the following command
200 to see assigned values:
201
202 ----
203 # pve-firewall localnet
204 local hostname: example
205 local IP address: 192.168.2.100
206 network auto detect: 192.168.0.0/20
207 using detected local_network: 192.168.0.0/20
208 ----
209
210 The firewall automatically sets up rules to allow everything needed
211 for cluster communication (corosync, API, SSH) using this alias.
212
213 The user can overwrite these values in the cluster.fw alias
214 section. If you use a single host on a public network, it is better to
215 explicitly assign the local IP address
216
217 ----
218 # /etc/pve/firewall/cluster.fw
219 [ALIASES]
220 local_network 1.2.3.4 # use the single ip address
221 ----
222
223 IP Sets
224 ~~~~~~~
225
226 IP sets can be used to define groups of networks and hosts. You can
227 refer to them with `+name` in the firewall rules' `source` and `dest`
228 properties.
229
230 The following example allows HTTP traffic from the `management` IP
231 set.
232
233 IN HTTP(ACCEPT) -source +management
234
235 Standard IP set `management`
236 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
237
238 This IP set applies only to host firewalls (not VM firewalls). Those
239 ips are allowed to do normal management tasks (PVE GUI, VNC, SPICE,
240 SSH).
241
242 The local cluster network is automatically added to this IP set (alias
243 `cluster_network`), to enable inter-host cluster
244 communication. (multicast,ssh,...)
245
246 ----
247 # /etc/pve/firewall/cluster.fw
248
249 [IPSET management]
250 192.168.2.10
251 192.168.2.10/24
252 ----
253
254 Standard IP set 'blacklist'
255 ^^^^^^^^^^^^^^^^^^^^^^^^^^^
256
257 Traffic from these ips is dropped by every host's and VM's firewall.
258
259 ----
260 # /etc/pve/firewall/cluster.fw
261
262 [IPSET blacklist]
263 77.240.159.182
264 213.87.123.0/24
265 ----
266
267 [[ipfilter-section]]
268 Standard IP set 'ipfilter-net*'
269 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
270
271 These filters belong to a VM's network interface and are mainly used to prevent
272 IP spoofing. If such a set exists for an interface then any outgoing traffic
273 with a source IP not matching its interface's corresponding ipfilter set will
274 be dropped.
275
276 For containers with configured IP addresses these sets, if they exist (or are
277 activated via the general `IP Filter` option in the VM's firewall's 'options'
278 tab), implicitly contain the associated IP addresses.
279
280 For both virtual machines and containers they also implicitly contain the
281 standard MAC-derived IPv6 link-local address in order to allow the neighbor
282 discovery protocol to work.
283
284 ----
285 /etc/pve/firewall/<VMID>.fw
286
287 [IPSET ipfilter-net0] # only allow specified IPs on net0
288 192.168.2.10
289 ----
290
291 Services and Commands
292 ~~~~~~~~~~~~~~~~~~~~~
293
294 The firewall runs two service daemons on each node:
295
296 * pvefw-logger: NFLOG daemon (ulogd replacement).
297 * pve-firewall: updates iptables rules
298
299 There is also a CLI command named 'pve-firewall', which can be used to
300 start and stop the firewall service:
301
302 # pve-firewall start
303 # pve-firewall stop
304
305 To get the status use:
306
307 # pve-firewall status
308
309 The above command reads and compiles all firewall rules, so you will
310 see warnings if your firewall configuration contains any errors.
311
312 If you want to see the generated iptables rules you can use:
313
314 # iptables-save
315
316 Tips and Tricks
317 ~~~~~~~~~~~~~~~
318
319 How to allow FTP
320 ^^^^^^^^^^^^^^^^
321
322 FTP is an old style protocol which uses port 21 and several other dynamic ports. So you
323 need a rule to accept port 21. In addition, you need to load the 'ip_conntrack_ftp' module.
324 So please run:
325
326 modprobe ip_conntrack_ftp
327
328 and add `ip_conntrack_ftp` to '/etc/modules' (so that it works after a reboot) .
329
330 Suricata IPS integration
331 ^^^^^^^^^^^^^^^^^^^^^^^^
332
333 If you want to use the http://suricata-ids.org/[Suricata IPS]
334 (Intrusion Prevention System), it's possible.
335
336 Packets will be forwarded to the IPS only after the firewall ACCEPTed
337 them.
338
339 Rejected/Dropped firewall packets don't go to the IPS.
340
341 Install suricata on proxmox host:
342
343 ----
344 # apt-get install suricata
345 # modprobe nfnetlink_queue
346 ----
347
348 Don't forget to add `nfnetlink_queue` to '/etc/modules' for next reboot.
349
350 Then, enable IPS for a specific VM with:
351
352 ----
353 # /etc/pve/firewall/<VMID>.fw
354
355 [OPTIONS]
356 ips: 1
357 ips_queues: 0
358 ----
359
360 `ips_queues` will bind a specific cpu queue for this VM.
361
362 Available queues are defined in
363
364 ----
365 # /etc/default/suricata
366 NFQUEUE=0
367 ----
368
369 Notes on IPv6
370 ^^^^^^^^^^^^^
371
372 The firewall contains a few IPv6 specific options. One thing to note is that
373 IPv6 does not use the ARP protocol anymore, and instead uses NDP (Neighbor
374 Discovery Protocol) which works on IP level and thus needs IP addresses to
375 succeed. For this purpose link-local addresses derived from the interface's MAC
376 address are used. By default the 'NDP' option is enabled on both host and VM
377 level to allow neighbor discovery (NDP) packets to be sent and received.
378
379 Beside neighbor discovery NDP is also used for a couple of other things, like
380 autoconfiguration and advertising routers.
381
382 By default VMs are allowed to send out router solicitation messages (to query
383 for a router), and to receive router advetisement packets. This allows them to
384 use stateless auto configuration. On the other hand VMs cannot advertise
385 themselves as routers unless the 'Allow Router Advertisement' (`radv: 1`) option
386 is set.
387
388 As for the link local addresses required for NDP, there's also an 'IP Filter'
389 (`ipfilter: 1`) option which can be enabled which has the same effect as adding
390 an `ipfilter-net*` ipset for each of the VM's network interfaces containing the
391 corresponding link local addresses. (See the
392 <<ipfilter-section,Standard IP set 'ipfilter-net*'>> section for details.)
393
394 Avoiding link-local addresses on tap and veth devices
395 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
396
397 With IPv6 enabled by default every interface gets a MAC-derived link local
398 address. However, most devices on a typical {pve} setup are connected to a
399 bridge and so the bridge is the only interface which really needs one.
400
401 To disable a link local address on an interface you can set the interface's
402 `disable_ipv6` sysconf variable. Despite the name, this does not prevent IPv6
403 traffic from passing through the interface when routing or bridging, so the
404 only noticeable effect will be the removal of the link local address.
405
406 The easiest method of achieving this setting for all newly started VMs is to
407 set it for the `default` interface configuration and enabling it explicitly on
408 the interfaces which need it. This is also the case for other settings such as
409 `forwarding`, `accept_ra` or `autoconf`.
410
411 Here's a possible setup:
412 ----
413 # /etc/sysconf.d/90-ipv6.conf
414
415 net.ipv6.conf.default.forwarding = 0
416 net.ipv6.conf.default.proxy_ndp = 0
417 net.ipv6.conf.default.autoconf = 0
418 net.ipv6.conf.default.disable_ipv6 = 1
419 net.ipv6.conf.default.accept_ra = 0
420
421 net.ipv6.conf.lo.disable_ipv6 = 0
422 ----
423
424 ----
425 # /etc/network/interfaces
426 (...)
427 iface vmbr0 inet6 static
428 address fc00::31
429 netmask 16
430 gateway fc00::1
431 accept_ra 0
432 pre-up echo 0 > /proc/sys/net/ipv6/conf/$IFACE/disable_ipv6
433 (...)
434 ----
435
436 Ports used by Proxmox VE
437 ------------------------
438
439 * Web interface: 8006
440 * VNC Web console: 5900-5999
441 * SPICE proxy: 3128
442 * sshd (used for cluster actions): 22
443 * rpcbind: 111
444 * corosync multicast (if you run a cluster): 5404, 5405 UDP
445
446
447 ifdef::manvolnum[]
448
449 Macro Definitions
450 -----------------
451
452 include::pve-firewall-macros.adoc[]
453
454
455 include::pve-copyright.adoc[]
456
457 endif::manvolnum[]