#
#
-# Copyright (C) 2006, 2007, 2008 Google Inc.
+# Copyright (C) 2006, 2007, 2008, 2010 Google Inc.
#
# 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
import os.path
import re
import logging
-import tempfile
import time
from ganeti import rpc
from ganeti import ssconf
from ganeti import serializer
from ganeti import hypervisor
+from ganeti import bdev
+from ganeti import netutils
def _InitSSHSetup():
utils.AddAuthorizedKey(auth_keys, utils.ReadFile(pub_key))
-def GenerateSelfSignedSslCert(file_name, validity=(365 * 5)):
- """Generates a self-signed SSL certificate.
+def GenerateHmacKey(file_name):
+ """Writes a new HMAC key.
@type file_name: str
@param file_name: Path to output file
- @type validity: int
- @param validity: Validity for certificate in days
"""
- (fd, tmp_file_name) = tempfile.mkstemp(dir=os.path.dirname(file_name))
- try:
- try:
- # Set permissions before writing key
- os.chmod(tmp_file_name, 0600)
+ utils.WriteFile(file_name, data="%s\n" % utils.GenerateSecret(), mode=0400,
+ backup=True)
+
+
+def GenerateClusterCrypto(new_cluster_cert, new_rapi_cert, new_confd_hmac_key,
+ new_cds, rapi_cert_pem=None, cds=None,
+ nodecert_file=constants.NODED_CERT_FILE,
+ rapicert_file=constants.RAPI_CERT_FILE,
+ hmackey_file=constants.CONFD_HMAC_KEY,
+ cds_file=constants.CLUSTER_DOMAIN_SECRET_FILE):
+ """Updates the cluster certificates, keys and secrets.
+
+ @type new_cluster_cert: bool
+ @param new_cluster_cert: Whether to generate a new cluster certificate
+ @type new_rapi_cert: bool
+ @param new_rapi_cert: Whether to generate a new RAPI certificate
+ @type new_confd_hmac_key: bool
+ @param new_confd_hmac_key: Whether to generate a new HMAC key
+ @type new_cds: bool
+ @param new_cds: Whether to generate a new cluster domain secret
+ @type rapi_cert_pem: string
+ @param rapi_cert_pem: New RAPI certificate in PEM format
+ @type cds: string
+ @param cds: New cluster domain secret
+ @type nodecert_file: string
+ @param nodecert_file: optional override of the node cert file path
+ @type rapicert_file: string
+ @param rapicert_file: optional override of the rapi cert file path
+ @type hmackey_file: string
+ @param hmackey_file: optional override of the hmac key file path
- result = utils.RunCmd(["openssl", "req", "-new", "-newkey", "rsa:1024",
- "-days", str(validity), "-nodes", "-x509",
- "-keyout", tmp_file_name, "-out", tmp_file_name,
- "-batch"])
- if result.failed:
- raise errors.OpExecError("Could not generate SSL certificate, command"
- " %s had exitcode %s and error message %s" %
- (result.cmd, result.exit_code, result.output))
+ """
+ # noded SSL certificate
+ cluster_cert_exists = os.path.exists(nodecert_file)
+ if new_cluster_cert or not cluster_cert_exists:
+ if cluster_cert_exists:
+ utils.CreateBackup(nodecert_file)
- # Make read-only
- os.chmod(tmp_file_name, 0400)
+ logging.debug("Generating new cluster certificate at %s", nodecert_file)
+ utils.GenerateSelfSignedSslCert(nodecert_file)
- os.rename(tmp_file_name, file_name)
- finally:
- utils.RemoveFile(tmp_file_name)
- finally:
- os.close(fd)
+ # confd HMAC key
+ if new_confd_hmac_key or not os.path.exists(hmackey_file):
+ logging.debug("Writing new confd HMAC key to %s", hmackey_file)
+ GenerateHmacKey(hmackey_file)
+ # RAPI
+ rapi_cert_exists = os.path.exists(rapicert_file)
-def GenerateHmacKey(file_name):
- """Writes a new HMAC key.
+ if rapi_cert_pem:
+ # Assume rapi_pem contains a valid PEM-formatted certificate and key
+ logging.debug("Writing RAPI certificate at %s", rapicert_file)
+ utils.WriteFile(rapicert_file, data=rapi_cert_pem, backup=True)
- @type file_name: str
- @param file_name: Path to output file
+ elif new_rapi_cert or not rapi_cert_exists:
+ if rapi_cert_exists:
+ utils.CreateBackup(rapicert_file)
- """
- utils.WriteFile(file_name, data="%s\n" % utils.GenerateSecret(), mode=0400)
+ logging.debug("Generating new RAPI certificate at %s", rapicert_file)
+ utils.GenerateSelfSignedSslCert(rapicert_file)
+
+ # Cluster domain secret
+ if cds:
+ logging.debug("Writing cluster domain secret to %s", cds_file)
+ utils.WriteFile(cds_file, data=cds, backup=True)
+
+ elif new_cds or not os.path.exists(cds_file):
+ logging.debug("Generating new cluster domain secret at %s", cds_file)
+ GenerateHmacKey(cds_file)
def _InitGanetiServerSetup(master_name):
"""Setup the necessary configuration for the initial node daemon.
This creates the nodepass file containing the shared password for
- the cluster and also generates the SSL certificate.
+ the cluster, generates the SSL certificate and starts the node daemon.
- """
- GenerateSelfSignedSslCert(constants.SSL_CERT_FILE)
-
- # Don't overwrite existing file
- if not os.path.exists(constants.RAPI_CERT_FILE):
- GenerateSelfSignedSslCert(constants.RAPI_CERT_FILE)
+ @type master_name: str
+ @param master_name: Name of the master node
- if not os.path.exists(constants.HMAC_CLUSTER_KEY):
- GenerateHmacKey(constants.HMAC_CLUSTER_KEY)
+ """
+ # Generate cluster secrets
+ GenerateClusterCrypto(True, False, False, False)
result = utils.RunCmd([constants.DAEMON_UTIL, "start", constants.NODED])
if result.failed:
" 10 seconds" % node_name)
+def _InitFileStorage(file_storage_dir):
+ """Initialize if needed the file storage.
+
+ @param file_storage_dir: the user-supplied value
+ @return: either empty string (if file storage was disabled at build
+ time) or the normalized path to the storage directory
+
+ """
+ if not constants.ENABLE_FILE_STORAGE:
+ return ""
+
+ file_storage_dir = os.path.normpath(file_storage_dir)
+
+ if not os.path.isabs(file_storage_dir):
+ raise errors.OpPrereqError("The file storage directory you passed is"
+ " not an absolute path.", errors.ECODE_INVAL)
+
+ if not os.path.exists(file_storage_dir):
+ try:
+ os.makedirs(file_storage_dir, 0750)
+ except OSError, err:
+ raise errors.OpPrereqError("Cannot create file storage directory"
+ " '%s': %s" % (file_storage_dir, err),
+ errors.ECODE_ENVIRON)
+
+ if not os.path.isdir(file_storage_dir):
+ raise errors.OpPrereqError("The file storage directory '%s' is not"
+ " a directory." % file_storage_dir,
+ errors.ECODE_ENVIRON)
+ return file_storage_dir
+
+
+#pylint: disable-msg=R0913
def InitCluster(cluster_name, mac_prefix,
master_netdev, file_storage_dir, candidate_pool_size,
secondary_ip=None, vg_name=None, beparams=None,
nicparams=None, hvparams=None, enabled_hypervisors=None,
- modify_etc_hosts=True, modify_ssh_setup=True):
+ modify_etc_hosts=True, modify_ssh_setup=True,
+ maintain_node_health=False, drbd_helper=None,
+ uid_pool=None, default_iallocator=None):
"""Initialise the cluster.
@type candidate_pool_size: int
" entries: %s" % invalid_hvs,
errors.ECODE_INVAL)
- hostname = utils.GetHostInfo()
+ hostname = netutils.GetHostInfo()
if hostname.ip.startswith("127."):
raise errors.OpPrereqError("This host's IP resolves to the private"
(hostname.ip, constants.ETC_HOSTS),
errors.ECODE_ENVIRON)
- if not utils.OwnIpAddress(hostname.ip):
+ if not netutils.OwnIpAddress(hostname.ip):
raise errors.OpPrereqError("Inconsistency: this host's name resolves"
" to %s,\nbut this ip address does not"
" belong to this host. Aborting." %
hostname.ip, errors.ECODE_ENVIRON)
- clustername = utils.GetHostInfo(cluster_name)
+ clustername = \
+ netutils.GetHostInfo(netutils.HostInfo.NormalizeName(cluster_name))
- if utils.TcpPing(clustername.ip, constants.DEFAULT_NODED_PORT,
+ if netutils.TcpPing(clustername.ip, constants.DEFAULT_NODED_PORT,
timeout=5):
raise errors.OpPrereqError("Cluster IP already active. Aborting.",
errors.ECODE_NOTUNIQUE)
if secondary_ip:
- if not utils.IsValidIP(secondary_ip):
+ if not netutils.IsValidIP4(secondary_ip):
raise errors.OpPrereqError("Invalid secondary ip given",
errors.ECODE_INVAL)
if (secondary_ip != hostname.ip and
- not utils.OwnIpAddress(secondary_ip)):
+ not netutils.OwnIpAddress(secondary_ip)):
raise errors.OpPrereqError("You gave %s as secondary IP,"
" but it does not belong to this host." %
secondary_ip, errors.ECODE_ENVIRON)
" you are not using lvm" % vgstatus,
errors.ECODE_INVAL)
- file_storage_dir = os.path.normpath(file_storage_dir)
-
- if not os.path.isabs(file_storage_dir):
- raise errors.OpPrereqError("The file storage directory you passed is"
- " not an absolute path.", errors.ECODE_INVAL)
-
- if not os.path.exists(file_storage_dir):
+ if drbd_helper is not None:
try:
- os.makedirs(file_storage_dir, 0750)
- except OSError, err:
- raise errors.OpPrereqError("Cannot create file storage directory"
- " '%s': %s" % (file_storage_dir, err),
+ curr_helper = bdev.BaseDRBD.GetUsermodeHelper()
+ except errors.BlockDeviceError, err:
+ raise errors.OpPrereqError("Error while checking drbd helper"
+ " (specify --no-drbd-storage if you are not"
+ " using drbd): %s" % str(err),
errors.ECODE_ENVIRON)
+ if drbd_helper != curr_helper:
+ raise errors.OpPrereqError("Error: requiring %s as drbd helper but %s"
+ " is the current helper" % (drbd_helper,
+ curr_helper),
+ errors.ECODE_INVAL)
- if not os.path.isdir(file_storage_dir):
- raise errors.OpPrereqError("The file storage directory '%s' is not"
- " a directory." % file_storage_dir,
- errors.ECODE_ENVIRON)
+ file_storage_dir = _InitFileStorage(file_storage_dir)
if not re.match("^[0-9a-z]{2}:[0-9a-z]{2}:[0-9a-z]{2}$", mac_prefix):
raise errors.OpPrereqError("Invalid mac prefix given '%s'" % mac_prefix,
hv_class = hypervisor.GetHypervisor(hv_name)
hv_class.CheckParameterSyntax(hv_params)
- # set up the inter-node password and certificate
+ # set up the inter-node password and certificate, start noded
_InitGanetiServerSetup(hostname.name)
# set up ssh config and /etc/hosts
if modify_ssh_setup:
_InitSSHSetup()
+ if default_iallocator is not None:
+ alloc_script = utils.FindFile(default_iallocator,
+ constants.IALLOCATOR_SEARCH_PATH,
+ os.path.isfile)
+ if alloc_script is None:
+ raise errors.OpPrereqError("Invalid default iallocator script '%s'"
+ " specified" % default_iallocator,
+ errors.ECODE_INVAL)
+
now = time.time()
# init of cluster config file
candidate_pool_size=candidate_pool_size,
modify_etc_hosts=modify_etc_hosts,
modify_ssh_setup=modify_ssh_setup,
+ uid_pool=uid_pool,
ctime=now,
mtime=now,
uuid=utils.NewUUID(),
+ maintain_node_health=maintain_node_health,
+ drbd_usermode_helper=drbd_helper,
+ default_iallocator=default_iallocator,
)
master_node_config = objects.Node(name=hostname.name,
primary_ip=hostname.ip,
"""
sshrunner = ssh.SshRunner(cluster_name)
- noded_cert = utils.ReadFile(constants.SSL_CERT_FILE)
+ noded_cert = utils.ReadFile(constants.NODED_CERT_FILE)
rapi_cert = utils.ReadFile(constants.RAPI_CERT_FILE)
- hmac_key = utils.ReadFile(constants.HMAC_CLUSTER_KEY)
+ confd_hmac_key = utils.ReadFile(constants.CONFD_HMAC_KEY)
# in the base64 pem encoding, neither '!' nor '.' are valid chars,
# so we use this to detect an invalid certificate; as long as the
# cert doesn't contain this, the here-document will be correctly
# parsed by the shell sequence below. HMAC keys are hexadecimal strings,
# so the same restrictions apply.
- for content in (noded_cert, rapi_cert, hmac_key):
+ for content in (noded_cert, rapi_cert, confd_hmac_key):
if re.search('^!EOF\.', content, re.MULTILINE):
raise errors.OpExecError("invalid SSL certificate or HMAC key")
noded_cert += "\n"
if not rapi_cert.endswith("\n"):
rapi_cert += "\n"
- if not hmac_key.endswith("\n"):
- hmac_key += "\n"
+ if not confd_hmac_key.endswith("\n"):
+ confd_hmac_key += "\n"
# set up inter-node password and certificate and restarts the node daemon
# and then connect with ssh to set password and start ganeti-noded
# note that all the below variables are sanitized at this point,
# either by being constants or by the checks above
+ # TODO: Could this command exceed a shell's maximum command length?
mycommand = ("umask 077 && "
"cat > '%s' << '!EOF.' && \n"
"%s!EOF.\n"
"%s!EOF.\n"
"chmod 0400 %s %s %s && "
"%s start %s" %
- (constants.SSL_CERT_FILE, noded_cert,
+ (constants.NODED_CERT_FILE, noded_cert,
constants.RAPI_CERT_FILE, rapi_cert,
- constants.HMAC_CLUSTER_KEY, hmac_key,
- constants.SSL_CERT_FILE, constants.RAPI_CERT_FILE,
- constants.HMAC_CLUSTER_KEY,
+ constants.CONFD_HMAC_KEY, confd_hmac_key,
+ constants.NODED_CERT_FILE, constants.RAPI_CERT_FILE,
+ constants.CONFD_HMAC_KEY,
constants.DAEMON_UTIL, constants.NODED))
result = sshrunner.Run(node, 'root', mycommand, batch=False,
logging.error("Could not disable the master role on the old master"
" %s, please disable manually: %s", old_master, msg)
+ master_ip = sstore.GetMasterIP()
+ total_timeout = 30
# Here we have a phase where no master should be running
+ def _check_ip():
+ if netutils.TcpPing(master_ip, constants.DEFAULT_NODED_PORT):
+ raise utils.RetryAgain()
+
+ try:
+ utils.Retry(_check_ip, (1, 1.5, 5), total_timeout)
+ except utils.RetryTimeout:
+ logging.warning("The master IP is still reachable after %s seconds,"
+ " continuing but activating the master on the current"
+ " node will probably fail", total_timeout)
# instantiate a real config writer, as we now know we have the
# configuration data
@return: list of (node, votes)
"""
- myself = utils.HostInfo().name
+ myself = netutils.HostInfo().name
try:
node_list.remove(myself)
except ValueError: