/usr/share/heartbeat/cts/CM_hb.py is in heartbeat 1:3.0.5-3ubuntu2.
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 | #!/usr/bin/python
'''CTS: Cluster Testing System: heartbeat dependent modules...
Classes related to testing high-availability clusters...
Lots of things are implemented.
Lots of things are not implemented.
We have many more ideas of what to do than we've implemented.
'''
__copyright__='''
Copyright (C) 2000,2001,2005 Alan Robertson <alanr@unix.sh>
Licensed under the GNU GPL.
'''
#
# 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., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
from CTS import *
class HeartbeatCM(ClusterManager):
'''
The heartbeat cluster manager class.
It implements the things we need to talk to and manipulate
heartbeat clusters
'''
def __init__(self, Environment, randseed=None):
self.ResourceDirs = ["/etc/ha.d/resource.d", "/etc/rc.d/init.d", "/etc/rc.d/"]
self.ResourceFile = Environment["HAdir"] + "/haresources"
self.ConfigFile = Environment["HAdir"]+ "/ha.cf"
ClusterManager.__init__(self, Environment, randseed=randseed)
self.update({
"Name" : "heartbeat",
"DeadTime" : 30,
"StableTime" : 30,
"StartCmd" : "/usr/lib/heartbeat/ha_logd -d >/dev/null 2>&1; MALLOC_CHECK_=2 /usr/lib/heartbeat/heartbeat >/dev/null 2>&1",
"StopCmd" : "/usr/lib/heartbeat/heartbeat -k",
"StatusCmd" : "/usr/lib/heartbeat/heartbeat -s",
"RereadCmd" : "/usr/lib/heartbeat/heartbeat -r",
"StartDRBDCmd" : "/etc/init.d/drbd start >/dev/null 2>&1",
"StopDRBDCmd" : "/etc/init.d/drbd stop",
"StatusDRBDCmd" : "/etc/init.d/drbd status",
"DRBDCheckconf" : "/etc/init.d/drbd checkconfig >/var/run/drbdconf 2>&1",
"BreakCommCmd" : "/usr/share/heartbeat/TestHeartbeatComm break-communication >/dev/null 2>&1",
"FixCommCmd" : "/usr/share/heartbeat/TestHeartbeatComm fix-communication >/dev/null 2>&1",
"DelFileCommCmd" : "/usr/share/heartbeat/TestHeartbeatComm delete-testingfile >/dev/null 2>&1",
"SaveFileCmd" : "/usr/share/heartbeat/TestHeartbeatComm save-testingfile /tmp/OnlyForTesting >/dev/null 2>&1",
"ReduceCommCmd" : "/usr/share/heartbeat/TestHeartbeatComm reduce-communication %s %s>/dev/null 2>&1",
"RestoreCommCmd" : "/usr/share/heartbeat/TestHeartbeatComm restore-communication /tmp/OnlyForTesting >/dev/null 2>&1",
"IPaddrCmd" : "/etc/ha.d/resource.d/IPaddr %s status",
"Standby" : "/usr/share/heartbeat/hb_standby >/dev/null 2>&1",
"TestConfigDir" : "/etc/ha.d/testconfigs",
"LogFileName" : Environment["LogFileName"],
# Patterns to look for in the log files for various occasions...
"Pat:We_started" : " (%s) .* Initial resource acquisition complete",
"Pat:They_started" : " (%s) .* Initial resource acquisition complete",
"Pat:We_stopped" : "%s heartbeat.*Heartbeat shutdown complete",
"Pat:Logd_stopped" : "%s logd:.*Exiting write process",
"Pat:They_stopped" : "%s heartbeat.*node (%s).*: is dead",
"Pat:They_dead" : "node (%s).*: is dead",
"Pat:All_stopped" : " (%s).*heartbeat.*Heartbeat shutdown complete",
"Pat:StandbyOK" : "Standby resource acquisition done",
"Pat:StandbyNONE" : "No reply to standby request",
"Pat:StandbyTRANSIENT" : "standby message.*ignored.*in flux",
"Pat:Return_partition" : "Cluster node %s returning after partition",
# Bad news Regexes. Should never occur.
"BadRegexes" : (
r"Shutting down\.",
r"Forcing shutdown\.",
r"Both machines own .* resources!",
r"No one owns .* resources!",
r", exiting\.",
r"ERROR:",
r"CRIT.*:",
),
})
self.cf=HBConfig(Environment["HAdir"])
self._finalConditions()
def SetClusterConfig(self, configpath="default", nodelist=None):
'''Activate the named test configuration throughout the cluster.
This code is specialized to heartbeat.
'''
rc=1
Command='''
cd %s%s%s; : cd to test configuration directory
for j in *
do
if
[ -f "/etc/ha.d/$j" ];
then
if
cmp $j /etc/ha.d/$j >/dev/null 2>&1;
then
: Config file $j is already up to correct.
else
echo "Touching $j"
cp $j /etc/ha.d/$j
fi
fi
done
''' % (self["TestConfigDir"], os.sep, configpath)
if nodelist == None:
nodelist=self.Env["nodes"]
for node in nodelist:
if not self.rsh(node, Command): rc=None
self.rereadall()
return rc
def ResourceGroups(self):
'''
Return the list of resources groups defined in this configuration.
This code is specialized to heartbeat.
We make the assumption that the resource file on the local machine
is the same as that of a cluster member.
We aren't necessarily a member of the cluster
(In fact, we usually aren't).
'''
RscGroups=[]
file = open(self.ResourceFile, "r")
while (1):
line = file.readline()
if line == "": break
idx=string.find(line, '#')
if idx >= 0:
line=line[:idx]
if line == "": continue
line = string.strip(line)
# Is this wrong?
tokens = re.split("[ \t]+", line)
# Ignore the default server for this resource group
del tokens[0]
Group=[]
for token in tokens:
if token != "":
idx=string.find(token, "::")
if idx > 0:
tuple=string.split(token, "::")
else:
#
# Is this an IPaddr default resource type?
#
if re.match("^[0-9]+\.[0-9]+\.[0-9]+\.[0-9]+$"
, token):
tuple=["IPaddr", token]
else:
tuple = [token, None]
Resource = self.hbResource(tuple[0], tuple[1])
Group.append(Resource)
RscGroups.append(Group)
file.close()
return RscGroups
def InternalCommConfig(self):
'''
Return a list of communication paths. Each path consists of a tuple like this:
mediatype serial | ip
interface/dev name eth0 | /dev/ttyS0...
protocol tcp?? | udp | None
port Number | None
'''
Path = {"mediatype" : None, "interface": None, "protocol" : None, "port": None}
cf = self.cf
for cfp in cf.Parameters:
if cfp == "serial":
if Path["mediatype"] == None:
Path["mediatype"] = ["serial"]
else: Path["mediatype"].append("serial")
if Path["interface"] == None:
Path["interface"] = cf.Parameters["serial"]
else:
for serial in cf.Parameters["serial"]:
Path["interface"].append(serial)
if cfp == "bcast" or cfp == "mcast" or cfp == "ucast" :
if Path["mediatype"] == None:
Path["mediatype"] = ["ip"]
else: Path["mediatype"].append("ip")
if cfp == "bcast":
interfaces = cf.Parameters[cfp]
if cfp == "ucast":
interfaces = [cf.Parameters[cfp][0]]
if cfp == "mcast":
Path["port"] = [cf.Parameters[cfp][0][2]]
Path["protocol"] = "udp"
interfaces = [cf.Parameters[cfp][0][0]]
if Path["interface"] == None:
Path["interface"] = interfaces
else:
for interface in interfaces:
if interface not in Path["interface"]:
Path["interface"].append(interface)
if cfp == "udpport":
Path["port"] = cf.Parameters["udpport"]
Path["protocol"] = ["udp"]
if Path["port"] == None:
Path["port"] = [694]
return Path
def HasQuorum(self, node_list):
(
'''Return TRUE if the cluster currently has quorum. According to
current heartbeat code this means one node is up.
''')
return self.upcount() >= 1
def hbResource(self, type, instance):
'''
Our job is to create the right kind of resource. For most
resources, we just create an HBResource object, but for
IP addresses, we create an HBipResource instead.
Some other types of resources may also be added as special
cases.
'''
if type == "IPaddr":
return HBipResource(self, type, instance)
return HBResource(self, type, instance)
class HBResource(Resource):
def IsRunningOn(self, nodename):
'''
This member function returns true if our resource is running
on the given node in the cluster.
We call the status operation for the resource script.
'''
return self._ResourceOperation("status", "OK|running", nodename)
def _ResourceOperation(self, operation, pattern, nodename):
'''
We call the requested operation for the resource script.
We don't care what kind of operation we were called to do
particularly.
When we were created, we were bound to a cluster manager, which
has its own remote execution method (which we use here).
'''
if self.Instance == None:
instance = ""
else:
instance = self.Instance
Rlist = 'LIST="'
for dir in self.CM.ResourceDirs:
Rlist = Rlist + " " + dir
Rlist = Rlist + '"; '
Script= Rlist + '''
T="''' + self.ResourceType + '''";
I="''' + instance + '''";
for dir in $LIST;
do
if
[ -f "$dir/$T" -a -x "$dir/$T" ]
then
"$dir/$T" $I ''' + operation + '''
exit $?
fi
done 2>&1;
exit 1;'''
#print "Running " + Script + "\n"
line = self.CM.rsh.readaline(nodename, Script)
if operation == "status":
if re.search(pattern, line):
return 1
return self.CM.rsh.lastrc == 0
def Start(self, nodename):
'''
This member function starts or activates the resource.
'''
return self._ResourceOperation("start", None, nodename)
def Stop(self, nodename):
'''
This member function stops or deactivates the resource.
'''
return self._ResourceOperation("stop", None, nodename)
# def IsWorkingCorrectly(self, nodename):
# "We default to returning TRUE for this one..."
# if self.Instance == None:
# self.CM.log("Faking out: " + self.ResourceType)
# else:
# self.CM.log("Faking out: " + self.ResourceType + self.Instance)
# return 1
def IsWorkingCorrectly(self, nodename):
return self._ResourceOperation("monitor", "OK", nodename)
class HBipResource(HBResource):
'''
We are a specialized IP address resource which knows how to test
to see if our resource type is actually being served.
We are cheat and run the IPaddr resource script on
the current machine, because it's a more interesting case.
'''
def IsWorkingCorrectly(self, nodename):
return self._ResourceOperation("monitor", "OK", self.CM.OurNode)
#
# A heartbeat configuration class...
# It reads and parses the heartbeat config
# files
#
class HBConfig:
# Which options have multiple words on the line?
MultiTokenKeywords = {"mcast" : None , "stonith_host": None}
def __init__(self, configdir="/etc/ha.d"):
self.Parameters = {}
self.ResourceGroups = {}
self._ReadConfig(os.path.join(configdir, "ha.cf"))
FirstUp_NodeSelection()
LastUp_NodeSelection()
no_failback = NoAutoFailbackPolicy()
auto_failback = AutoFailbackPolicy()
#
# We allow each resource group to have its own failover/back
# policies
#
if self.Parameters.has_key("nice_failback") \
and self.Parameters["nice_failback"] == "on":
HBConfig.DefaultFailbackPolicy = no_failback
elif self.Parameters.has_key("auto_failback") \
and self.Parameters["auto_failback"] == "off":
HBConfig.DefaultFailbackPolicy = no_failback
else:
HBConfig.DefaultFailbackPolicy = auto_failback
HBConfig.DefaultNodeSelectionPolicy = NodeSelectionPolicies["FirstUp"]
self._ReadResourceGroups(os.path.join(configdir, "haresources"))
# Read ha.cf config file
def _ReadConfig(self, ConfigFile):
self.ConfigPath = ConfigFile;
fp = open(ConfigFile)
while 1:
line=fp.readline()
if not line:
return
line = re.sub("#.*", "", line)
line = string.rstrip(line)
if len(line) < 1:
continue
tokens = line.split()
key = tokens[0]
values = tokens[1:]
if HBConfig.MultiTokenKeywords.has_key(key):
# group items from this line together, and separate
# from the items on other lines
values = [values]
if self.Parameters.has_key(key):
if key == "node":
self.Parameters[key].extend(values)
else:
self.Parameters[key].append(values[0])
else:
self.Parameters[key] = values
# Read a line from the haresources file...
# - allow for \ continuations...
def _GetRscLine(self, fp):
linesofar = None
continuation=1
while continuation:
continuation = 0
line=fp.readline()
if not line:
break
line = re.sub("#.*", "", line)
if line[len(line)-2] == "\\":
line = line[0:len(line)-2] + "\n"
continuation=1
if linesofar == None:
linesofar = line
else:
linesofar = linesofar + line
return linesofar
def _ReadResourceGroups(self, RscFile):
self.RscPath = RscFile;
fp = open(RscFile)
thisline = ""
while 1:
line=self._GetRscLine(fp)
if not line:
return
line = line.strip()
if len(line) < 1:
continue
tokens = line.split()
node = tokens[0]
resources = tokens[1:]
rscargs=[]
for resource in resources:
name=resource.split("::", 1)
if len(name) > 1:
args=name[1].split("::")
else:
args=None
name = name[0]
rscargs.append(Resource(name, args))
name = tokens[0] + "__" + tokens[1]
assert not self.ResourceGroups.has_key(name)
#
# Create the resource group
#
self.ResourceGroups[name] = ResourceGroup(name \
, rscargs
, node.split(",")
# Provide default value
, HBConfig.DefaultNodeSelectionPolicy
, HBConfig.DefaultFailbackPolicy)
#
# Return the list of nodes in the cluster...
#
def nodes(self):
result = self.Parameters["node"]
result.sort()
return result
class ClusterState:
pass
class ResourceGroup:
def __init__(self, name, resourcelist, possiblenodes
, nodeselection_policy, failback_policy):
self.name = name
self.resourcelist = resourcelist
self.possiblenodes = possiblenodes
self.nodeselection_policy = nodeselection_policy
self.failback_policy = failback_policy
self.state = None
self.attributes = {}
self.history = []
def __str__(self):
result = string.join(self.possiblenodes, ",")
for rsc in self.resourcelist:
result = result + " " + str(rsc)
return result
class Resource:
def __init__(self, name, arguments=None):
self.name = name
self.arguments = arguments
def __str__(self):
result = self.name
try:
for arg in self.arguments:
result = result + "::" + arg
except TypeError:
pass
return result
#######################################################################
#
# Base class defining policies for where we put resources
# when we're starting, or when a failure has occurred...
#
#######################################################################
NodeSelectionPolicies = {}
class NodeSelectionPolicy:
def __init__(self, name):
self.name = name
NodeSelectionPolicies[name] = self
def name(self):
return self.name
#
# nodenames: the list of nodes eligible to run this resource
# ResourceGroup: the group to be started...
# ClusterState: Cluster state information
#
def SelectNode(self, nodenames, ResourceGroup, ClusterState):
return None
#
# Choose the first node in the list...
#
class FirstUp_NodeSelection(NodeSelectionPolicy):
def __init__(self):
NodeSelectionPolicy.__init__(self, "FirstUp")
def SelectNode(self, nodenames, ResourceGroup, ClusterState):
return nodenames[0]
#
# Choose the last node in the list...
# (kind of a dumb example)
#
class LastUp_NodeSelection(NodeSelectionPolicy):
def __init__(self):
NodeSelectionPolicy.__init__(self, "LastUp")
def SelectNode(self, nodenames, ResourceGroup, ClusterState):
return nodenames[len(nodenames)-1]
#######################################################################
#
# Failback policies...
#
# Where to locate a resource group when an eligible node rejoins
# the cluster...
#
#######################################################################
FailbackPolicies = {}
class FailbackPolicy:
def __init__(self, name):
self.name = name
FailbackPolicies[name] = self
def name(self):
return self.name
#
# currentnode: The node the service is currently on
# returningnode: The node which just rejoined
# eligiblenodes: Permitted nodes which are up
# SelectionPolicy: the normal NodeSelectionPolicy
# Cluster state information...
#
def SelectNewNode(self, currentnode, returningnode, eligiblenodes
, SelectionPolicy, ResourceGroup, ClusterState):
return None
#
# This FailbackPolicy is like "normal failback" in heartbeat
#
class AutoFailbackPolicy(FailbackPolicy):
def __init__(self):
FailbackPolicy.__init__(self, "failback")
def SelectNewNode(self, currentnode, returningnode, eligiblenodes
, SelectionPolicy, ResourceGroup, ClusterState):
# Select where it should run based on current normal policy
# just as though we were starting it for the first time.
return SelectionPolicy(eligiblenodes, ResourceGroup, ClusterState)
#
# This FailbackPolicy is like "nice failback" in heartbeat
#
class NoAutoFailbackPolicy(FailbackPolicy):
def __init__(self):
FailbackPolicy.__init__(self, "failuresonly")
def SelectNewNode(self, currentnode, returningnode, eligiblenodes
, SelectionPolicy, ResourceGroup):
# Always leave the resource where it is...
return currentnode
#######################################################################
#
# A little test code...
#
# Which you are advised to completely ignore...
#
#######################################################################
if __name__ == '__main__':
FirstUp_NodeSelection()
LastUp_NodeSelection()
no_failback = NoAutoFailbackPolicy()
auto_failback = AutoFailbackPolicy()
cf=HBConfig("/etc/ha.d")
print "Cluster configuration:\n"
print "Nodes:", cf.nodes(), "\n"
print "Config Parameters:", cf.Parameters, "\n"
for groupname in cf.ResourceGroups.keys():
print "Resource Group %s:\n\t%s\n" % (groupname, cf.ResourceGroups[groupname])
|