/usr/share/doc/ganeti-doc/html/design-node-security.html is in ganeti-doc 2.15.2-3.
This file is owned by root:root, with mode 0o644.
The actual contents of the file can be viewed below.
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 | <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
"http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
<html xmlns="http://www.w3.org/1999/xhtml">
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8" />
<title>Improvements of Node Security — Ganeti 2.15.2 documentation</title>
<link rel="stylesheet" href="_static/style.css" type="text/css" />
<link rel="stylesheet" href="_static/pygments.css" type="text/css" />
<script type="text/javascript">
var DOCUMENTATION_OPTIONS = {
URL_ROOT: './',
VERSION: '2.15.2',
COLLAPSE_INDEX: false,
FILE_SUFFIX: '.html',
HAS_SOURCE: true
};
</script>
<script type="text/javascript" src="_static/jquery.js"></script>
<script type="text/javascript" src="_static/underscore.js"></script>
<script type="text/javascript" src="_static/doctools.js"></script>
<link rel="top" title="Ganeti 2.15.2 documentation" href="index.html" />
<link rel="up" title="Design document drafts" href="design-draft.html" />
<link rel="next" title="Design for adding ifdown script to KVM" href="design-ifdown.html" />
<link rel="prev" title="Instance move improvements" href="design-move-instance-improvements.html" />
</head>
<body role="document">
<div class="related" role="navigation" aria-label="related navigation">
<h3>Navigation</h3>
<ul>
<li class="right" style="margin-right: 10px">
<a href="design-ifdown.html" title="Design for adding ifdown script to KVM"
accesskey="N">next</a></li>
<li class="right" >
<a href="design-move-instance-improvements.html" title="Instance move improvements"
accesskey="P">previous</a> |</li>
<li class="nav-item nav-item-0"><a href="index.html">Ganeti 2.15.2 documentation</a> »</li>
<li class="nav-item nav-item-1"><a href="design-draft.html" accesskey="U">Design document drafts</a> »</li>
</ul>
</div>
<div class="document">
<div class="documentwrapper">
<div class="bodywrapper">
<div class="body" role="main">
<div class="section" id="improvements-of-node-security">
<h1><a class="toc-backref" href="#id3">Improvements of Node Security</a><a class="headerlink" href="#improvements-of-node-security" title="Permalink to this headline">¶</a></h1>
<p>This document describes an enhancement of Ganeti’s security by restricting
the distribution of security-sensitive data to the master and master
candidates only.</p>
<p>Note: In this document, we will use the term ‘normal node’ for a node that
is neither master nor master-candidate.</p>
<div class="contents topic" id="contents">
<p class="topic-title first">Contents</p>
<ul class="simple">
<li><a class="reference internal" href="#improvements-of-node-security" id="id3">Improvements of Node Security</a><ul>
<li><a class="reference internal" href="#objective" id="id4">Objective</a></li>
<li><a class="reference internal" href="#detailed-design" id="id5">Detailed design</a><ul>
<li><a class="reference internal" href="#current-state-and-shortcomings" id="id6">Current state and shortcomings</a></li>
<li><a class="reference internal" href="#proposal-concerning-ssh-host-key-distribution" id="id7">Proposal concerning SSH host key distribution</a><ul>
<li><a class="reference internal" href="#rapi" id="id8">RAPI</a></li>
<li><a class="reference internal" href="#cluster-initialization" id="id9">Cluster initialization</a></li>
<li><a class="reference internal" href="#re-adding-nodes-to-a-cluster" id="id10">(Re-)Adding nodes to a cluster</a></li>
<li><a class="reference internal" href="#pro-and-demoting-a-node-to-from-master-candidate" id="id11">Pro- and demoting a node to/from master candidate</a></li>
<li><a class="reference internal" href="#offlining-and-onlining-a-node" id="id12">Offlining and onlining a node</a></li>
<li><a class="reference internal" href="#cluster-verify" id="id13">Cluster verify</a></li>
<li><a class="reference internal" href="#upgrades" id="id14">Upgrades</a></li>
<li><a class="reference internal" href="#downgrades" id="id15">Downgrades</a></li>
<li><a class="reference internal" href="#renew-crypto" id="id16">Renew-Crypto</a></li>
<li><a class="reference internal" href="#compliance-to-no-ssh-init-and-no-node-setup" id="id17">Compliance to –no-ssh-init and –no-node-setup</a></li>
</ul>
</li>
<li><a class="reference internal" href="#proposal-regarding-node-daemon-certificates" id="id18">Proposal regarding node daemon certificates</a><ul>
<li><a class="reference internal" href="#id1" id="id19">Cluster initialization</a></li>
<li><a class="reference internal" href="#re-adding-nodes" id="id20">(Re-)Adding nodes</a></li>
<li><a class="reference internal" href="#promotion-and-demotion-of-master-candidates" id="id21">Promotion and demotion of master candidates</a></li>
<li><a class="reference internal" href="#id2" id="id22">Cluster verify</a></li>
<li><a class="reference internal" href="#crypto-renewal" id="id23">Crypto renewal</a></li>
</ul>
</li>
<li><a class="reference internal" href="#further-considerations" id="id24">Further considerations</a><ul>
<li><a class="reference internal" href="#watcher" id="id25">Watcher</a></li>
<li><a class="reference internal" href="#other-keys-and-daemons" id="id26">Other Keys and Daemons</a></li>
<li><a class="reference internal" href="#related-and-future-work" id="id27">Related and Future Work</a></li>
</ul>
</li>
</ul>
</li>
</ul>
</li>
</ul>
</div>
<div class="section" id="objective">
<h2><a class="toc-backref" href="#id4">Objective</a><a class="headerlink" href="#objective" title="Permalink to this headline">¶</a></h2>
<p>Up till 2.10, Ganeti distributes security-relevant keys to all nodes,
including nodes that are neither master nor master-candidates. Those
keys are the private and public SSH keys for node communication and the
SSL certficate and private key for RPC communication. Objective of this
design is to limit the set of nodes that can establish ssh and RPC
connections to the master and master candidates.</p>
<p>As pointed out in
<a class="reference external" href="https://code.google.com/p/ganeti/issues/detail?id=377">issue 377</a>, this
is a security risk. Since all nodes have these keys, compromising
any of those nodes would possibly give an attacker access to all other
machines in the cluster. Reducing the set of nodes that are able to
make ssh and RPC connections to the master and master candidates would
significantly reduce the risk simply because fewer machines would be a
valuable target for attackers.</p>
<p>Note: For bigger installations of Ganeti, it is advisable to run master
candidate nodes as non-vm-capable nodes. This would reduce the attack
surface for the hypervisor exploitation.</p>
</div>
<div class="section" id="detailed-design">
<h2><a class="toc-backref" href="#id5">Detailed design</a><a class="headerlink" href="#detailed-design" title="Permalink to this headline">¶</a></h2>
<div class="section" id="current-state-and-shortcomings">
<h3><a class="toc-backref" href="#id6">Current state and shortcomings</a><a class="headerlink" href="#current-state-and-shortcomings" title="Permalink to this headline">¶</a></h3>
<p>Currently (as of 2.10), all nodes hold the following information:</p>
<ul class="simple">
<li>the ssh host keys (public and private)</li>
<li>the ssh root keys (public and private)</li>
<li>node daemon certificate (the SSL client certificate and its
corresponding private key)</li>
</ul>
<p>Concerning ssh, this setup contains the following security issue. Since
all nodes of a cluster can ssh as root into any other cluster node, one
compromised node can harm all other nodes of a cluster.</p>
<p>Regarding the SSL encryption of the RPC communication with the node
daemon, we currently have the following setup. There is only one
certificate which is used as both, client and server certificate. Besides
the SSL client verification, we check if the used client certificate is
the same as the certificate stored on the server.</p>
<p>This means that any node running a node daemon can also act as an RPC
client and use it to issue RPC calls to other cluster nodes. This in
turn means that any compromised node could be used to make RPC calls to
any node (including itself) to gain full control over VMs. This could
be used by an attacker to for example bring down the VMs or exploit bugs
in the virtualization stacks to gain access to the host machines as well.</p>
</div>
<div class="section" id="proposal-concerning-ssh-host-key-distribution">
<h3><a class="toc-backref" href="#id7">Proposal concerning SSH host key distribution</a><a class="headerlink" href="#proposal-concerning-ssh-host-key-distribution" title="Permalink to this headline">¶</a></h3>
<p>We propose the following design regarding the SSH host key handling. The
root keys are untouched by this design.</p>
<p>Each node gets its own ssh private/public key pair, but only the public
keys of the master candidates get added to the <code class="docutils literal"><span class="pre">authorized_keys</span></code> file
of all nodes. This has the advantages, that:</p>
<ul class="simple">
<li>Only master candidates can ssh into other nodes, thus compromised
nodes cannot compromise the cluster further.</li>
<li>One can remove a compromised master candidate from a cluster
(including removing it’s public key from all nodes’ <code class="docutils literal"><span class="pre">authorized_keys</span></code>
file) without having to regenerate and distribute new ssh keys for all
master candidates. (Even though it is be good practice to do that anyway,
since the compromising of the other master candidates might have taken
place already.)</li>
<li>If a (uncompromised) master candidate is offlined to be sent for
repair due to a hardware failure before Ganeti can remove any keys
from it (for example when the network adapter of the machine is broken),
we don’t have to worry about the keys being on a machine that is
physically accessible.</li>
</ul>
<p>To ensure security while transferring public key information and
updating the <code class="docutils literal"><span class="pre">authorized_keys</span></code>, there are several other changes
necessary:</p>
<ul class="simple">
<li>Any distribution of keys (in this case only public keys) is done via
SSH and not via RPC. An attacker who has RPC control should not be
able to get SSH access where he did not have SSH access before
already.</li>
<li>The only RPC calls that are made in this context are from the master
daemon to the node daemon on its own host and noded ensures as much
as possible that the change to be made does not harm the cluster’s
security boundary.</li>
<li>The nodes that are potential master candidates keep a list of public
keys of potential master candidates of the cluster in a separate
file called <code class="docutils literal"><span class="pre">ganeti_pub_keys</span></code> to keep track of which keys could
possibly be added <code class="docutils literal"><span class="pre">authorized_keys</span></code> files of the nodes. We come
to what “potential” means in this case in the next section. The key
list is only transferred via SSH or written directly by noded. It
is not stored in the cluster config, because the config is
distributed via RPC.</li>
</ul>
<p>The following sections describe in detail which Ganeti commands are
affected by the proposed changes.</p>
<div class="section" id="rapi">
<h4><a class="toc-backref" href="#id8">RAPI</a><a class="headerlink" href="#rapi" title="Permalink to this headline">¶</a></h4>
<p>The design goal to limit SSH powers to master candidates conflicts with
the current powers a user of the RAPI interface would have. The
<code class="docutils literal"><span class="pre">master_capable</span></code> flag of nodes can be modified via RAPI.
That means, an attacker that has access to the RAPI interface, can make
all non-master-capable nodes master-capable, and then increase the master
candidate pool size till all machines are master candidates (or at least
a particular machine that he is aming for). This means that with RAPI
access and a compromised normal node, one can make this node a master
candidate and then still have the power to compromise the whole cluster.</p>
<p>To mitigate this issue, we propose the following changes:</p>
<ul class="simple">
<li>Add a flag <code class="docutils literal"><span class="pre">master_capability_rapi_modifiable</span></code> to the cluster
configuration which indicates whether or not it should be possible
to modify the <code class="docutils literal"><span class="pre">master_capable</span></code> flag of nodes via RAPI. The flag is
set to <code class="docutils literal"><span class="pre">False</span></code> by default and can itself only be changed on the
commandline. In this design doc, we refer to the flag as the
“rapi flag” from here on.</li>
<li>Only if the <code class="docutils literal"><span class="pre">master_capabability_rapi_modifiable</span></code> switch is set to
<code class="docutils literal"><span class="pre">True</span></code>, it is possible to modify the master-capability flag of
nodes.</li>
</ul>
<p>With this setup, there are the following definitions of “potential
master candidates” depending on the rapi flag:</p>
<ul class="simple">
<li>If the rapi flag is set to <code class="docutils literal"><span class="pre">True</span></code>, all cluster nodes are potential
master candidates, because as described above, all of them can
eventually be made master candidates via RAPI and thus security-wise,
we haven’t won anything above the current SSH handling.</li>
<li>If the rapi flag is set to <code class="docutils literal"><span class="pre">False</span></code>, only the master capable nodes
are considered potential master candidates, as it is not possible to
make them master candidates via RAPI at all.</li>
</ul>
<p>Note that when the rapi flag is changed, the state of the
<code class="docutils literal"><span class="pre">ganeti_pub_keys</span></code> file on all nodes has to be updated accordingly.
This should be done in the client script <code class="docutils literal"><span class="pre">gnt_cluster</span></code> before the
RPC call to update the configuration is made, because this way, if
someone would try to perform that RPC call on master to trick it into
thinking that the flag is enabled, this would not help as the content of
the <code class="docutils literal"><span class="pre">ganeti_pub_keys</span></code> file is a crucial part in the design of the
distribution of the SSH keys.</p>
<p>Note: One could think of always allowing to disable the master-capability
via RAPI and just restrict the enabling of it, thus making it possible
to RAPI-“freeze” the nodes’ master-capability state once it disabled.
However, we think these are rather confusing semantics of the involved
flags and thus we go with proposed design.</p>
<p>Note that this change will break RAPI compatibility, at least if the
rapi flag is not explicitely set to <code class="docutils literal"><span class="pre">True</span></code>. We made this choice to
have the more secure option as default, because otherwise it is
unlikely to be widely used.</p>
</div>
<div class="section" id="cluster-initialization">
<h4><a class="toc-backref" href="#id9">Cluster initialization</a><a class="headerlink" href="#cluster-initialization" title="Permalink to this headline">¶</a></h4>
<p>On cluster initialization, the following steps are taken in
bootstrap.py:</p>
<ul class="simple">
<li>A public/private key pair is generated (as before), but only used
by the first (and thus master) node. In particular, the private key
never leaves the node.</li>
<li>A mapping of node UUIDs to public SSH keys is created and stored
as text file in <code class="docutils literal"><span class="pre">/var/lib/ganeti/ganeti_pub_keys</span></code> only accessible
by root (permissions 0600). The master node’s uuid and its public
key is added as first entry. The format of the file is one
line per node, each line composed as <code class="docutils literal"><span class="pre">node_uuid</span> <span class="pre">ssh_key</span></code>.</li>
<li>The node’s public key is added to it’s own <code class="docutils literal"><span class="pre">authorized_keys</span></code> file.</li>
</ul>
</div>
<div class="section" id="re-adding-nodes-to-a-cluster">
<h4><a class="toc-backref" href="#id10">(Re-)Adding nodes to a cluster</a><a class="headerlink" href="#re-adding-nodes-to-a-cluster" title="Permalink to this headline">¶</a></h4>
<p>According to <a class="reference internal" href="design-node-add.html"><em>Design for adding a node to a cluster</em></a>, Ganeti transfers the ssh keys to
every node that gets added to the cluster.</p>
<p>Adding a new node will require the following steps.</p>
<p>In gnt_node.py:</p>
<ul class="simple">
<li>On the new node, a new public/private SSH key pair is generated.</li>
<li>The public key of the new node is fetched (via SSH) to the master
node and if it is a potential master candidate (see definition above),
it is added to the <code class="docutils literal"><span class="pre">ganeti_pub_keys</span></code> list on the master node.</li>
<li>The public keys of all current master candidates are added to the
new node’s <code class="docutils literal"><span class="pre">authorized_keys</span></code> file (also via SSH).</li>
</ul>
<p>In LUNodeAdd in cmdlib/node.py:</p>
<ul class="simple">
<li>The LUNodeAdd determines whether or not the new node is a master
candidate and in any case updates the cluster’s configuration with the
new nodes information. (This is not changed by the proposed design.)</li>
<li>If the new node is a master candidate, we make an RPC call to the node
daemon of the master node to add the new node’s public key to all
nodes’ <code class="docutils literal"><span class="pre">authorized_keys</span></code> files. The implementation of this RPC call
has to be extra careful as described in the next steps, because
compromised RPC security should not compromise SSH security.</li>
</ul>
<p>RPC call execution in noded (on master node):</p>
<ul class="simple">
<li>Check that the public key of the new node is in the
<code class="docutils literal"><span class="pre">ganeti_pub_keys</span></code> file of the master node to make sure that no keys
of nodes outside the Ganeti cluster and no keys that are not potential
master candidates gain SSH access in the cluster.</li>
<li>Via SSH, transfer the new node’s public key to all nodes (including
the new node) and add it to their <code class="docutils literal"><span class="pre">authorized_keys</span></code> file.</li>
<li>The <code class="docutils literal"><span class="pre">ganeti_pub_keys</span></code> file is transferred via SSH to all
potential master candidates nodes except the master node
(including the new one).</li>
</ul>
<p>In case of readding a node that used to be in the cluster before,
handling of the SSH keys would basically be the same, in particular also
a new SSH key pair is generated for the node, because we cannot be sure
that the old key pair has not been compromised while the node was
offlined. Note that for reasons of data hygiene, a node’s
<code class="docutils literal"><span class="pre">ganeti_pub_keys</span></code> file is cleared before the node is readded.
Also, Ganeti attempts to remove any Ganeti keys from the <code class="docutils literal"><span class="pre">authorized_keys</span></code>
file before the node is readded. However, since Ganeti does not keep a list
of all keys ever used in the cluster, this applies only to keys which
are currently used in the cluster. Note that Ganeti won’t touch any keys
that were added to the <code class="docutils literal"><span class="pre">authorized_keys</span></code> by other systems than Ganeti.</p>
</div>
<div class="section" id="pro-and-demoting-a-node-to-from-master-candidate">
<h4><a class="toc-backref" href="#id11">Pro- and demoting a node to/from master candidate</a><a class="headerlink" href="#pro-and-demoting-a-node-to-from-master-candidate" title="Permalink to this headline">¶</a></h4>
<p>If the role of a node is changed from ‘normal’ to ‘master_candidate’,
the procedure is the same as for adding nodes from the step “In
LUNodeAdd ...” on.</p>
<p>If a node gets demoted to ‘normal’, the master daemon makes a similar
RPC call to the master node’s node daemon as for adding a node.</p>
<p>In the RPC call, noded will perform the following steps:</p>
<ul class="simple">
<li>Check that the public key of the node to be demoted is indeed in the
<code class="docutils literal"><span class="pre">ganeti_pub_keys</span></code> file to avoid deleting ssh keys of machines that
don’t belong to the cluster (and thus potentially lock out the
administrator).</li>
<li>Via SSH, remove the key from all node’s <code class="docutils literal"><span class="pre">authorized_keys</span></code> files.</li>
</ul>
<p>This affected the behavior of the following commands:</p>
<dl class="docutils">
<dt>::</dt>
<dd>gnt-node modify –master-candidate=yes
gnt-node modify –master-candidate=no [–auto-promote]</dd>
</dl>
<p>If the node has been master candidate already before the command to promote
it was issued, Ganeti does not do anything.</p>
<p>Note that when you demote a node from master candidate to normal node, another
master-capable and normal node will be promoted to master candidate. For this
newly promoted node, the same changes apply as if it was explicitely promoted.</p>
<p>The same behavior should be ensured for the corresponding rapi command.</p>
</div>
<div class="section" id="offlining-and-onlining-a-node">
<h4><a class="toc-backref" href="#id12">Offlining and onlining a node</a><a class="headerlink" href="#offlining-and-onlining-a-node" title="Permalink to this headline">¶</a></h4>
<p>When offlining a node, it immediately loses its role as master or master
candidate as well. When it is onlined again, it will become master
candidate again if it was so before. The handling of the keys should be done
in the same way as when the node is explicitely promoted or demoted to or from
master candidate. See the previous section for details.</p>
<p>This affects the command:</p>
<dl class="docutils">
<dt>::</dt>
<dd>gnt-node modify –offline=yes
gnt-node modify –offline=no [–auto-promote]</dd>
</dl>
<p>For offlining, the removal of the keys is particularly important, as the
detection of a compromised node might be the very reason for the offlining.
Of course we cannot guarantee that removal of the key is always successful,
because the node might not be reachable anymore. Even though it is a
best-effort operation, it is still an improvement over the status quo,
because currently Ganeti does not even try to remove any keys.</p>
<p>The same behavior should be ensured for the corresponding rapi command.</p>
</div>
<div class="section" id="cluster-verify">
<h4><a class="toc-backref" href="#id13">Cluster verify</a><a class="headerlink" href="#cluster-verify" title="Permalink to this headline">¶</a></h4>
<p>So far, <code class="docutils literal"><span class="pre">gnt-cluster</span> <span class="pre">verify</span></code> checks the SSH connectivity of all nodes to
all other nodes. We propose to replace this by the following checks:</p>
<ul class="simple">
<li>For all master candidates, we check if they can connect any other node
in the cluster (other master candidates and normal nodes).</li>
<li>We check if the <code class="docutils literal"><span class="pre">ganeti_pub_keys</span></code> file contains keys of nodes that
are no longer in the cluster or that are not potential master
candidates.</li>
<li>For all normal nodes, we check if their key does not appear in other
node’s <code class="docutils literal"><span class="pre">authorized_keys</span></code>. For now, we will only emit a warning
rather than an error if this check fails, because Ganeti might be
run in a setup where Ganeti is not the only system manipulating the
SSH keys.</li>
</ul>
</div>
<div class="section" id="upgrades">
<h4><a class="toc-backref" href="#id14">Upgrades</a><a class="headerlink" href="#upgrades" title="Permalink to this headline">¶</a></h4>
<p>When upgrading from a version that has the previous SSH setup to the one
proposed in this design, the upgrade procedure has to involve the
following steps in the post-upgrade hook:</p>
<ul class="simple">
<li>For all nodes, new SSH key pairs are generated.</li>
<li>All nodes and their public keys are added to the <code class="docutils literal"><span class="pre">ganeti_pub_keys</span></code>
file and the file is copied to all nodes.</li>
<li>All keys of master candidate nodes are added to the
<code class="docutils literal"><span class="pre">authorized_keys</span></code> files of all other nodes.</li>
</ul>
<p>Since this upgrade significantly changes the configuration of the
clusters’ nodes, we will add a note to the UPGRADE notes to make the
administrator aware of this fact (in case he intends to enable access
from normal nodes to master candidates for other reasons than Ganeti
uses the machines).</p>
<p>Also, in any operation where Ganeti creates new SSH keys, the old keys
will be backed up and not simply overridden.</p>
</div>
<div class="section" id="downgrades">
<h4><a class="toc-backref" href="#id15">Downgrades</a><a class="headerlink" href="#downgrades" title="Permalink to this headline">¶</a></h4>
<p>These downgrading steps will be implemtented from 2.13 to 2.12:</p>
<ul class="simple">
<li>The master node’s private/public key pair will be distributed to all
nodes (via SSH) and the individual SSH keys will be backed up.</li>
<li>The obsolete individual ssh keys will be removed from all nodes’
<code class="docutils literal"><span class="pre">authorized_keys</span></code> file.</li>
</ul>
</div>
<div class="section" id="renew-crypto">
<h4><a class="toc-backref" href="#id16">Renew-Crypto</a><a class="headerlink" href="#renew-crypto" title="Permalink to this headline">¶</a></h4>
<p>The <code class="docutils literal"><span class="pre">gnt-cluster</span> <span class="pre">renew-crypto</span></code> command will be extended by a new
option <code class="docutils literal"><span class="pre">--new-ssh-keys</span></code>, which will renew all SSH keys on all nodes
and rebuild the <code class="docutils literal"><span class="pre">authorized_keys</span></code> files and the <code class="docutils literal"><span class="pre">ganeti_pub_keys</span></code>
files according to the previous sections. This operation will be
performed considering the already stated security considerations, for
example minimizing RPC calls, distribution of keys via SSH only etc.</p>
</div>
<div class="section" id="compliance-to-no-ssh-init-and-no-node-setup">
<h4><a class="toc-backref" href="#id17">Compliance to –no-ssh-init and –no-node-setup</a><a class="headerlink" href="#compliance-to-no-ssh-init-and-no-node-setup" title="Permalink to this headline">¶</a></h4>
<p>With this design, Ganeti will do more manipulations of SSH keys and
<code class="docutils literal"><span class="pre">authorized_keys</span></code> files than before. If this is not feasible in
a Ganeti environment, the administrator has the option to prevent
Ganeti from performing any manipulations on the SSH setup of the nodes.
The options for doing so, are <code class="docutils literal"><span class="pre">--no-ssh-init</span></code> for <code class="docutils literal"><span class="pre">gnt-cluster</span>
<span class="pre">init</span></code>, and <code class="docutils literal"><span class="pre">--no-node-setup</span></code> for <code class="docutils literal"><span class="pre">gnt-node</span> <span class="pre">add</span></code>. Note that
these options already existed before the implementation of this
design, we just confirm that they will be complied to with the
new design as well.</p>
</div>
</div>
<div class="section" id="proposal-regarding-node-daemon-certificates">
<h3><a class="toc-backref" href="#id18">Proposal regarding node daemon certificates</a><a class="headerlink" href="#proposal-regarding-node-daemon-certificates" title="Permalink to this headline">¶</a></h3>
<p>Regarding the node daemon certificates, we propose the following changes
in the design.</p>
<ul class="simple">
<li>Instead of using the same certificate for all nodes as both, server
and client certificate, we generate a common server certificate (and
the corresponding private key) for all nodes and a different client
certificate (and the corresponding private key) for each node. The
server certificate will be self-signed. The client certficate will
be signed by the server certificate. The client certificates will
use the node UUID as serial number to ensure uniqueness within the
cluster. They will use the host’s hostname as the certificate
common name (CN).</li>
<li>In addition, we store a mapping of
(node UUID, client certificate digest) in the cluster’s configuration
and ssconf for hosts that are master or master candidate.
The client certificate digest is a hash of the client certificate.
We suggest a ‘sha1’ hash here. We will call this mapping ‘candidate map’
from here on.</li>
<li>The node daemon will be modified in a way that on an incoming RPC
request, it first performs a client verification (same as before) to
ensure that the requesting host is indeed the holder of the
corresponding private key. Additionally, it compares the digest of
the certificate of the incoming request to the respective entry of
the candidate map. If the digest does not match the entry of the host
in the mapping or is not included in the mapping at all, the SSL
connection is refused.</li>
</ul>
<p>This design has the following advantages:</p>
<ul class="simple">
<li>A compromised normal node cannot issue RPC calls, because it will
not be in the candidate map. (See the <code class="docutils literal"><span class="pre">Drawbacks</span></code> section regarding
an indirect way of achieving this though.)</li>
<li>A compromised master candidate would be able to issue RPC requests,
but on detection of its compromised state, it can be removed from the
cluster (and thus from the candidate map) without the need for
redistribution of any certificates, because the other master candidates
can continue using their own certificates. However, it is best
practice to issue a complete key renewal even in this case, unless one
can ensure no actions compromising other nodes have not already been
carried out.</li>
<li>A compromised node would not be able to use the other (possibly master
candidate) nodes’ information from the candidate map to issue RPCs,
because the config just stores the digests and not the certificate
itself.</li>
<li>A compromised node would be able to obtain another node’s certificate
by waiting for incoming RPCs from this other node. However, the node
cannot use the certificate to issue RPC calls, because the SSL client
verification would require the node to hold the corresponding private
key as well.</li>
</ul>
<p>Drawbacks of this design:</p>
<ul class="simple">
<li>Complexity of node and certificate management will be increased (see
following sections for details).</li>
<li>If the candidate map is not distributed fast enough to all nodes after
an update of the configuration, it might be possible to issue RPC calls
from a compromised master candidate node that has been removed
from the Ganeti cluster already. However, this is still a better
situation than before and an inherent problem when one wants to
distinguish between master candidates and normal nodes.</li>
<li>A compromised master candidate would still be able to issue RPC calls,
if it uses ssh to retrieve another master candidate’s client
certificate and the corresponding private SSL key. This is an issue
even with the first part of the improved handling of ssh keys in this
design (limiting ssh keys to master candidates), but it will be
eliminated with the second part of the design (separate ssh keys for
each master candidate).</li>
<li>Even though this proposal is an improvement towards the previous
situation in Ganeti, it still does not use the full power of SSL. For
further improvements, see Section “Related and future work”.</li>
<li>Signing the client certificates with the server certificate will
increase the complexity of the renew-crypto, as a renewal of the
server certificates requires the renewal (and signing) of all client
certificates as well.</li>
</ul>
<p>Alternative proposals:</p>
<ul class="simple">
<li>The initial version of this document described a setup where the
client certificates were also self-signed. This led to a serious
problem (Issue 1094), which would only have been solvable by
distributing all client certificates to all nodes and load them
as trusted CAs. As this would have resulted in having to restart
noded on all nodes every time a node is added, removed, demoted
or promoted, this was not feasible and we switched to client
certficates which are signed by the server certificate.</li>
<li>Instead of generating a client certificate per node, one could think
of just generating two different client certificates, one for normal
nodes and one for master candidates. Noded could then just check if
the requesting node has the master candidate certificate. Drawback of
this proposal is that once one master candidate gets compromised, all
master candidates would need to get a new certificate even if the
compromised master candidate had not yet fetched the certificates
from the other master candidates via ssh.</li>
<li>In addition to our main proposal, one could think of including a
piece of data (for example the node’s host name or UUID) in the RPC
call which is encrypted with the requesting node’s private key. The
node daemon could check if the datum can be decrypted using the node’s
certificate. However, this would ensure similar functionality as
SSL’s built-in client verification and add significant complexity
to Ganeti’s RPC protocol.</li>
</ul>
<p>In the following sections, we describe how our design affects various
Ganeti operations.</p>
<div class="section" id="id1">
<h4><a class="toc-backref" href="#id19">Cluster initialization</a><a class="headerlink" href="#id1" title="Permalink to this headline">¶</a></h4>
<p>On cluster initialization, so far only the node daemon certificate was
created. With our design, two certificates (and corresponding keys)
need to be created, a server certificate to be distributed to all nodes
and a client certificate only to be used by this particular node. In the
following, we use the term node daemon certificate for the server
certficate only.</p>
<p>In the cluster configuration, the candidate map is created. It is
populated with the respective entry for the master node. It is also
written to ssconf.</p>
</div>
<div class="section" id="re-adding-nodes">
<h4><a class="toc-backref" href="#id20">(Re-)Adding nodes</a><a class="headerlink" href="#re-adding-nodes" title="Permalink to this headline">¶</a></h4>
<p>When a node is added, the server certificate is copied to the node (as
before). Additionally, a new client certificate (and the corresponding
private key) is created on the new node to be used only by the new node
as client certifcate.</p>
<p>If the new node is a master candidate, the candidate map is extended by
the new node’s data. As before, the updated configuration is distributed
to all nodes (as complete configuration on the master candidates and
ssconf on all nodes). Note that distribution of the configuration after
adding a node is already implemented, since all nodes hold the list of
nodes in the cluster in ssconf anyway.</p>
<p>If the configuration for whatever reason already holds an entry for this
node, it will be overriden.</p>
<p>When readding a node, the procedure is the same as for adding a node.</p>
</div>
<div class="section" id="promotion-and-demotion-of-master-candidates">
<h4><a class="toc-backref" href="#id21">Promotion and demotion of master candidates</a><a class="headerlink" href="#promotion-and-demotion-of-master-candidates" title="Permalink to this headline">¶</a></h4>
<p>When a normal node gets promoted to be master candidate, an entry to the
candidate map has to be added and the updated configuration has to be
distributed to all nodes. If there was already an entry for the node,
we override it.</p>
<p>On demotion of a master candidate, the node’s entry in the candidate map
gets removed and the updated configuration gets redistibuted.</p>
<p>The same procedure applied to onlining and offlining master candidates.</p>
</div>
<div class="section" id="id2">
<h4><a class="toc-backref" href="#id22">Cluster verify</a><a class="headerlink" href="#id2" title="Permalink to this headline">¶</a></h4>
<p>Cluster verify will be extended by the following checks:</p>
<ul class="simple">
<li>Whether each entry in the candidate map indeed corresponds to a master
candidate.</li>
<li>Whether the master candidate’s certificate digest match their entry
in the candidate map.</li>
<li>Whether no node tries to use the certificate of another node. In
particular, it is important to check that no normal node tries to
use the certificate of a master candidate.</li>
<li>Whether there are still self-signed client certificates in use (from
a pre 2.12.4 Ganeti version).</li>
</ul>
</div>
<div class="section" id="crypto-renewal">
<h4><a class="toc-backref" href="#id23">Crypto renewal</a><a class="headerlink" href="#crypto-renewal" title="Permalink to this headline">¶</a></h4>
<p>Currently, when the cluster’s cryptographic tokens are renewed using the
<code class="docutils literal"><span class="pre">gnt-cluster</span> <span class="pre">renew-crypto</span></code> command the node daemon certificate is
renewed (among others). Option <code class="docutils literal"><span class="pre">--new-cluster-certificate</span></code> renews the
node daemon certificate only.</p>
<p>By adding an option <code class="docutils literal"><span class="pre">--new-node-certificates</span></code> we offer to renew the
client certificate. Whenever the client certificates are renewed, the
candidate map has to be updated and redistributed.</p>
<p>If for whatever reason, the candidate map becomes inconsistent, for example
due inconsistent updating after a demotion or offlining), the user can use
this option to renew the client certificates and update the candidate
certificate map.</p>
<p>Note that renewing the server certificate requires all client certificates
being renewed and signed by the new server certificate, because
otherwise their signature can not be verified by the server who only has
the new server certificate then.</p>
<p>As there was a different design in place in Ganeti 2.12.4 and previous
versions, we have to ensure that renew-crypto works on pre 2.12 versions and
2.12.1-4. Users that got hit by Issue 1094 will be encouraged to run
renew-crypto at least once after switching to 2.12.5. Those who did not
encounter this bug yet, will still get nagged friendly by gnt-cluster
verify.</p>
</div>
</div>
<div class="section" id="further-considerations">
<h3><a class="toc-backref" href="#id24">Further considerations</a><a class="headerlink" href="#further-considerations" title="Permalink to this headline">¶</a></h3>
<div class="section" id="watcher">
<h4><a class="toc-backref" href="#id25">Watcher</a><a class="headerlink" href="#watcher" title="Permalink to this headline">¶</a></h4>
<p>The watcher is a script that is run on all nodes in regular intervals. The
changes proposed in this design will not affect the watcher’s implementation,
because it behaves differently on the master than on non-master nodes.</p>
<p>Only on the master, it issues query calls which would require a client
certificate of a node in the candidate mapping. This is the case for the
master node. On non-master node, it’s only external communication is done via
the ConfD protocol, which uses the hmac key, which is present on all nodes.
Besides that, the watcher does not make any ssh connections, and thus is
not affected by the changes in ssh key handling either.</p>
</div>
<div class="section" id="other-keys-and-daemons">
<h4><a class="toc-backref" href="#id26">Other Keys and Daemons</a><a class="headerlink" href="#other-keys-and-daemons" title="Permalink to this headline">¶</a></h4>
<p>Ganeti handles a couple of other keys/certificates that have not been mentioned
in this design so far. Also, other daemons than the ones mentioned so far
perform intra-cluster communication. Neither the keys nor the daemons will
be affected by this design for several reasons:</p>
<ul class="simple">
<li>The hmac key used by ConfD (see <a class="reference internal" href="design-2.1.html"><em>Ganeti 2.1 design</em></a>): the hmac key is still
distributed to all nodes, because it was designed to be used for
communicating with ConfD, which should be possible from all nodes.
For example, the monitoring daemon which runs on all nodes uses it to
retrieve information from ConfD. However, since communication with ConfD
is read-only, a compromised node holding the hmac key does not enable an
attacker to change the cluster’s state.</li>
<li>The WConfD daemon writes the configuration to all master candidates
via RPC. Since it only runs on the master node, it’s ability to run
RPC requests is maintained with this design.</li>
<li>The rapi SSL key certificate and rapi user/password file ‘rapi_users’ is
already only copied to the master candidates (see <a class="reference internal" href="design-2.1.html"><em>Ganeti 2.1 design</em></a>,
Section <code class="docutils literal"><span class="pre">Redistribute</span> <span class="pre">Config</span></code>).</li>
<li>The spice certificates are still distributed to all nodes, since it should
be possible to use spice to access VMs on any cluster node.</li>
<li>The cluster domain secret is used for inter-cluster instance moves.
Since instances can be moved from any normal node of the source cluster to
any normal node of the destination cluster, the presence of this
secret on all nodes is necessary.</li>
</ul>
</div>
<div class="section" id="related-and-future-work">
<h4><a class="toc-backref" href="#id27">Related and Future Work</a><a class="headerlink" href="#related-and-future-work" title="Permalink to this headline">¶</a></h4>
<p>There a couple of suggestions on how to improve the SSL setup even more.
As a trade-off wrt to complexity and implementation effort, we did not
implement them yet (as of version 2.11) but describe them here for
future reference.</p>
<ul class="simple">
<li>The server certificate is currently self-signed and the client certificates
are signed by the server certificate. It would increase the security if they
were signed by a common CA. There is already a design doc for a Ganeti CA
which was suggested in a different context (related to import/export).
This would also be a benefit for the RPC calls. See design doc
<a class="reference internal" href="design-impexp2.html"><em>Design for import/export version 2</em></a> for more information. Implementing a CA is rather
complex, because it would mean also to support renewing the CA certificate and
providing and supporting infrastructure to revoke compromised certificates.</li>
<li>An extension of the previous suggestion would be to even enable the
system administrator to use an external CA. Especially in bigger
setups, where already an SSL infrastructure exists, it would be useful
if Ganeti can simply be integrated with it, rather than forcing the
user to use the Ganeti CA.</li>
<li>Ganeti RPC calls are currently done without checking if the hostname
of the node complies with the common name of the certificate. This
might be a desirable feature, but would increase the effort when a
node is renamed.</li>
<li>The typical use case for SSL is to have one certificate per node
rather than one shared certificate (Ganeti’s noded server certificate)
and a client certificate. One could change the design in a way that
only one certificate per node is used, but this would require a common
CA so that the validity of the certificate can be established by every
node in the cluster.</li>
<li>With the proposed design, the serial numbers of the client
certificates are set to the node UUIDs. This is technically also not
complying to how SSL is supposed to be used, as the serial numbers
should reflect the enumeration of certificates created by the CA. Once
a CA is implemented, it might be reasonable to change this
accordingly. The implementation of the proposed design also has the
drawback of the serial number not changing even if the certificate is
replaced by a new one (for example when calling <code class="docutils literal"><span class="pre">gnt-cluster</span> <span class="pre">renew-</span>
<span class="pre">crypt</span></code>), which also does not comply to way SSL was designed to be
used.</li>
</ul>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
<div class="sphinxsidebar" role="navigation" aria-label="main navigation">
<div class="sphinxsidebarwrapper">
<h3><a href="index.html">Table Of Contents</a></h3>
<ul>
<li><a class="reference internal" href="#">Improvements of Node Security</a><ul>
<li><a class="reference internal" href="#objective">Objective</a></li>
<li><a class="reference internal" href="#detailed-design">Detailed design</a><ul>
<li><a class="reference internal" href="#current-state-and-shortcomings">Current state and shortcomings</a></li>
<li><a class="reference internal" href="#proposal-concerning-ssh-host-key-distribution">Proposal concerning SSH host key distribution</a><ul>
<li><a class="reference internal" href="#rapi">RAPI</a></li>
<li><a class="reference internal" href="#cluster-initialization">Cluster initialization</a></li>
<li><a class="reference internal" href="#re-adding-nodes-to-a-cluster">(Re-)Adding nodes to a cluster</a></li>
<li><a class="reference internal" href="#pro-and-demoting-a-node-to-from-master-candidate">Pro- and demoting a node to/from master candidate</a></li>
<li><a class="reference internal" href="#offlining-and-onlining-a-node">Offlining and onlining a node</a></li>
<li><a class="reference internal" href="#cluster-verify">Cluster verify</a></li>
<li><a class="reference internal" href="#upgrades">Upgrades</a></li>
<li><a class="reference internal" href="#downgrades">Downgrades</a></li>
<li><a class="reference internal" href="#renew-crypto">Renew-Crypto</a></li>
<li><a class="reference internal" href="#compliance-to-no-ssh-init-and-no-node-setup">Compliance to –no-ssh-init and –no-node-setup</a></li>
</ul>
</li>
<li><a class="reference internal" href="#proposal-regarding-node-daemon-certificates">Proposal regarding node daemon certificates</a><ul>
<li><a class="reference internal" href="#id1">Cluster initialization</a></li>
<li><a class="reference internal" href="#re-adding-nodes">(Re-)Adding nodes</a></li>
<li><a class="reference internal" href="#promotion-and-demotion-of-master-candidates">Promotion and demotion of master candidates</a></li>
<li><a class="reference internal" href="#id2">Cluster verify</a></li>
<li><a class="reference internal" href="#crypto-renewal">Crypto renewal</a></li>
</ul>
</li>
<li><a class="reference internal" href="#further-considerations">Further considerations</a><ul>
<li><a class="reference internal" href="#watcher">Watcher</a></li>
<li><a class="reference internal" href="#other-keys-and-daemons">Other Keys and Daemons</a></li>
<li><a class="reference internal" href="#related-and-future-work">Related and Future Work</a></li>
</ul>
</li>
</ul>
</li>
</ul>
</li>
</ul>
<h4>Previous topic</h4>
<p class="topless"><a href="design-move-instance-improvements.html"
title="previous chapter">Instance move improvements</a></p>
<h4>Next topic</h4>
<p class="topless"><a href="design-ifdown.html"
title="next chapter">Design for adding ifdown script to KVM</a></p>
<div role="note" aria-label="source link">
<h3>This Page</h3>
<ul class="this-page-menu">
<li><a href="_sources/design-node-security.txt"
rel="nofollow">Show Source</a></li>
</ul>
</div>
<div id="searchbox" style="display: none" role="search">
<h3>Quick search</h3>
<form class="search" action="search.html" method="get">
<input type="text" name="q" />
<input type="submit" value="Go" />
<input type="hidden" name="check_keywords" value="yes" />
<input type="hidden" name="area" value="default" />
</form>
<p class="searchtip" style="font-size: 90%">
Enter search terms or a module, class or function name.
</p>
</div>
<script type="text/javascript">$('#searchbox').show(0);</script>
</div>
</div>
<div class="clearer"></div>
</div>
<div class="related" role="navigation" aria-label="related navigation">
<h3>Navigation</h3>
<ul>
<li class="right" style="margin-right: 10px">
<a href="design-ifdown.html" title="Design for adding ifdown script to KVM"
>next</a></li>
<li class="right" >
<a href="design-move-instance-improvements.html" title="Instance move improvements"
>previous</a> |</li>
<li class="nav-item nav-item-0"><a href="index.html">Ganeti 2.15.2 documentation</a> »</li>
<li class="nav-item nav-item-1"><a href="design-draft.html" >Design document drafts</a> »</li>
</ul>
</div>
<div class="footer" role="contentinfo">
© Copyright 2006, 2007, 2008, 2009, 2010, 2011, 2012, 2013 Google Inc..
Created using <a href="http://sphinx-doc.org/">Sphinx</a> 1.3.6.
</div>
</body>
</html>
|