def kinit(user, pswd, realm):
try:
cmd_to_run = 'echo "%s" | kinit %s@%s' % (pswd, user, realm.upper())
# print cmd_to_run
use_salt, err = config.use_salt()
if err:
raise Exception(err)
if use_salt:
import salt.client
errors = []
client = salt.client.LocalClient()
# print 'Running %s'%cmd_to_run
#assert False
r1 = client.cmd('*', 'cmd.run_all', [cmd_to_run])
if r1:
for node, ret in r1.items():
# print ret
if ret["retcode"] != 0:
e = "Error initiating kerberos on GRIDCell %s" % node
if "stderr" in ret:
e += " : %s" % ret["stderr"]
errors.append(e)
print errors
# print r1
if errors:
raise Exception(' '.join(errors))
else:
lines, err = command.get_command_output(cmd_to_run, shell=True)
# print lines, err
if err:
raise Exception(err)
except Exception, e:
return False, 'Error initializing kerberos : %s' % str(e)
def reload_configuration():
try:
use_salt, err = config.use_salt()
if err:
raise Exception(err)
if use_salt:
import salt.client
errors = ''
client = salt.client.LocalClient()
r1 = client.cmd('*', 'cmd.run_all',
['smbcontrol all reload-config'])
if r1:
for node, ret in r1.items():
# print ret
if ret["retcode"] != 0:
errors += "Error reloading samba on node %s " % node
if errors:
raise Exception(errors)
else:
cmd_to_run = 'smbcontrol all reload-config'
lines, err = command.get_command_output(cmd_to_run)
if err:
raise Exception(err)
ret, err = services_management.update_service_status(
'winbind', 'restart')
if err:
raise Exception(err)
except Exception, e:
return False, 'Error reloading CIFS configuration : %s' % str(e)
def delete_local_user(username):
try:
if not username:
raise Exception('No username specified')
d, err = get_local_user(username)
if not d:
if err:
raise Exception('Error locating user : %s' % err)
else:
raise Exception('Error locating user')
use_salt, err = config.use_salt()
if err:
raise Exception(err)
if use_salt:
import salt.client
client = salt.client.LocalClient()
rc = client.cmd('*', 'user.delete', [username])
# print rc
if rc:
for hostname, status in rc.items():
if not status:
raise Exception("Error deleting the system user")
else:
raise Exception("Error deleting the system user")
else:
cmd_to_run = 'userdel %s' % (username)
lines, err = command.get_command_output(cmd_to_run)
if err:
raise Exception(err)
if d['smb_user']:
lines, err = command.get_command_output(r'pdbedit -d 1 -x %s' %
username)
if err:
raise Exception(err)
except Exception, e:
return False, 'Error deleting local user : %s' % str(e)
def create_local_group(grpname, gid=None):
try:
# First check if groups exists. if so kick out
gl, err = get_local_groups()
if gl:
for gd in gl:
if gd["grpname"] == grpname:
raise Exception(
"Error creating group. The group \"%s\" already exists. "
% grpname)
elif err:
raise Exception("Error retrieving group list : %s" % err)
use_salt, err = config.use_salt()
if err:
raise Exception(err)
if use_salt:
import salt.client
client = salt.client.LocalClient()
if not gid:
rc = client.cmd('*', 'group.add', [grpname])
else:
rc = client.cmd('*', 'group.add', [grpname, gid])
print rc
if not rc:
raise Exception('Group creation failed')
for hostname, status in rc.items():
if not status:
raise Exception('Group creation failed')
else:
if gid:
cmd_to_run = 'groupadd -g %s %s' % (gid, grpname)
else:
cmd_to_run = 'groupadd %s' % (grpname)
lines, err = command.get_command_output(cmd_to_run)
if err:
raise Exception(err)
except Exception, e:
return False, 'Error creating a local group : %s' % str(e)
def display_status():
try:
hostname = socket.gethostname()
use_salt, err = config.use_salt()
if err:
raise Exception(err)
if use_salt:
print "Salt master service status :",
(r,
rc), err = command.execute_with_rc('service salt-master status')
if err:
raise Exception(err)
l, err = command.get_output_list(r)
if err:
raise Exception(err)
if l:
print '\n'.join(l)
else:
l, err = command.get_error_list(r)
if err:
raise Exception(err)
if l:
print '\n'.join(l)
print "Salt minion service status :",
(r,
rc), err = command.execute_with_rc('service salt-minion status')
if err:
raise Exception(err)
l, err = command.get_output_list(r)
if err:
raise Exception(err)
if l:
print '\n'.join(l)
else:
l, err = command.get_error_list(r)
if err:
raise Exception(err)
print l
if l:
print '\n'.join(l)
print "Samba service status :",
(r, rc), err = command.execute_with_rc('service smb status')
if err:
raise Exception(err)
l, err = command.get_output_list(r)
if err:
raise Exception(err)
if l:
print '\n'.join(l)
else:
l, err = command.get_error_list(r)
if err:
raise Exception(err)
if l:
print '\n'.join(l)
print "Winbind service status :",
(r, rc), err = command.execute_with_rc('service winbind status')
if err:
raise Exception(err)
l, err = command.get_output_list(r)
if err:
raise Exception(err)
if l:
print '\n'.join(l)
else:
l, err = command.get_error_list(r)
if err:
raise Exception(err)
if l:
print '\n'.join(l)
except Exception, e:
print "Error displaying system status : %s" % e
return -1
def configure_interface():
try:
os.system('clear')
interfaces, err = networking.get_interfaces()
if err:
raise Exception(
'Error retrieving interface information : %s' % err)
if not interfaces:
raise Exception('No interfaces detected')
print
print
print 'IntegralSTOR interface configuration'
print '--------------------------------------------'
print
print
print 'Current network interfaces : '
print
for if_name, iface in interfaces.items():
if if_name.startswith('lo'):
continue
print '- %s' % if_name
print
valid_input = False
while not valid_input:
ifname = raw_input(
'Enter the name of the interface that you wish to configure : ')
if ifname not in interfaces or ifname.startswith('lo'):
print 'Invalid interface name'
else:
valid_input = True
print
ip_info, err = networking.get_ip_info(ifname)
'''
if err:
raise Exception('Error retrieving interface information : %s'%err)
'''
if ip_info:
ip = ip_info["ipaddr"]
netmask = ip_info["netmask"]
if "default_gateway" in ip_info:
gateway = ip_info["default_gateway"]
else:
gateway = None
else:
ip = None
netmask = None
gateway = None
old_boot_proto, err = networking.get_interface_bootproto(ifname)
if err:
raise Exception(
'Error retrieving interface information : %s' % err)
time.sleep(5)
config_changed = False
str_to_print = "Configure for DHCP or static addressing (dhcp/static)? : "
valid_input = False
while not valid_input:
input = raw_input(str_to_print)
if input:
if input.lower() in ['static', 'dhcp']:
valid_input = True
boot_proto = input.lower()
if boot_proto != old_boot_proto:
config_changed = True
if not valid_input:
print "Invalid value. Please try again."
print
if boot_proto == 'static':
if ip:
str_to_print = "Enter IP address (currently %s, press enter to retain current value) : " % ip
else:
str_to_print = "Enter IP address (currently not set) : "
valid_input = False
while not valid_input:
input = raw_input(str_to_print)
if input:
ok, err = networking.validate_ip(input)
if err:
raise Exception('Error validating IP : %s' % err)
if ok:
valid_input = True
ip = input
config_changed = True
elif ip:
valid_input = True
if not valid_input:
print "Invalid value. Please try again."
print
if netmask:
str_to_print = "Enter netmask (currently %s, press enter to retain current value) : " % netmask
else:
str_to_print = "Enter netmask (currently not set) : "
valid_input = False
while not valid_input:
input = raw_input(str_to_print)
if input:
ok, err = networking.validate_netmask(input)
if err:
raise Exception('Error validating netmask : %s' % err)
if ok:
valid_input = True
netmask = input
config_changed = True
elif netmask:
valid_input = True
if not valid_input:
print "Invalid value. Please try again."
print
if gateway:
str_to_print = "Enter gateway (currently %s, press enter to retain current value) : " % gateway
else:
str_to_print = "Enter gateway (currently not set) : "
valid_input = False
while not valid_input:
input = raw_input(str_to_print)
if input:
ok, err = networking.validate_ip(input)
if err:
raise Exception('Error validating gateway : %s' % err)
if ok:
valid_input = True
gateway = input
config_changed = True
elif gateway:
valid_input = True
if not valid_input:
print "Invalid value. Please try again."
print
if config_changed:
d = {}
d['addr_type'] = boot_proto
if boot_proto == 'static':
d['ip'] = ip
d['netmask'] = netmask
d['default_gateway'] = gateway
ret, err = networking.update_interface_ip(ifname, d)
if not ret:
if err:
raise Exception(
'Error changing interface address : %s' % err)
else:
raise Exception('Error changing interface address')
restart = False
print
print
valid_input = False
while not valid_input:
str_to_print = 'Restart network services now (y/n) :'
print
input = raw_input(str_to_print)
if input:
if input.lower() in ['y', 'n']:
valid_input = True
if input.lower() == 'y':
restart = True
if not valid_input:
print "Invalid value. Please try again."
print
if restart:
ret, err = networking.restart_networking()
if not ret:
if err:
raise Exception(err)
else:
raise Exception("Couldn't restart.")
use_salt, err = config.use_salt()
if err:
raise Exception(err)
if use_salt:
(r, rc), err = command.execute_with_rc(
'service salt-minion restart')
if err:
raise Exception(err)
if rc == 0:
print "Salt minion service restarted succesfully."
else:
print "Error restarting salt minion services."
raw_input('Press enter to return to the main menu')
return -1
else:
print
print
raw_input(
'No changes have been made to the configurations. Press enter to return to the main menu.')
return 0
except Exception, e:
print "Error configuring network settings : %s" % e
return -1
def replace_disk(request):
return_dict = {}
try:
form = None
si, err = system_info.load_system_config()
if err:
raise Exception(err)
if not si:
raise Exception('Error loading system config')
return_dict['system_config_list'] = si
template = 'logged_in_error.html'
use_salt, err = config.use_salt()
if err:
raise Exception(err)
if request.method == "GET":
raise Exception("Incorrect access method. Please use the menus")
else:
if 'node' in request.POST:
node = request.POST["node"]
else:
node = si.keys()[0]
serial_number = request.POST["serial_number"]
if "conf" in request.POST:
if "node" not in request.POST or "serial_number" not in request.POST:
raise Exception(
"Incorrect access method. Please use the menus")
elif request.POST["node"] not in si:
raise Exception("Unknown node. Please use the menus")
elif "step" not in request.POST:
raise Exception("Incomplete request. Please use the menus")
elif request.POST["step"] not in [
"replace_method", "select_replacement_disk",
"offline_disk", "scan_for_new_disk", "online_new_disk"
]:
raise Exception("Incomplete request. Please use the menus")
else:
step = request.POST["step"]
# Which step of the replace disk are we in?
if step == "offline_disk":
# get the pool corresponding to the disk
# zpool offline pool disk
# send a screen asking them to replace the disk
if 'replacement_method' not in request.POST or request.POST[
'replacement_method'] not in [
'use_existing_disk', 'swap_out_disk'
]:
raise Exception('Invalid request')
return_dict['replacement_method'] = request.POST[
'replacement_method']
if request.POST[
'replacement_method'] == 'use_existing_disk':
# Then we should have landed here after already
# selecting the new disk so get and record the new
# disk details
if 'new_serial_number' not in request.POST:
raise Exception(
'Incomplete request. Please try again')
new_serial_number = request.POST[
'new_serial_number']
all_disks, err = disks.get_disk_info_all()
if new_serial_number not in all_disks:
raise Exception('Invalid disk selection')
# print new_serial_number
# print all_disks[new_serial_number]['id']
return_dict[
'new_serial_number'] = new_serial_number
return_dict['new_id'] = all_disks[
new_serial_number]['id']
pool = None
if serial_number in si[node]["disks"]:
disk = si[node]["disks"][serial_number]
if "pool" in disk:
pool = disk["pool"]
disk_id = disk["id"]
if not pool:
raise Exception(
"Could not find the storage pool on that disk. Please use the menus"
)
else:
cmd_to_run = 'zpool offline %s %s' % (pool,
disk_id)
# print 'Running %s'%cmd_to_run
#assert False
ret, err = command.get_command_output(cmd_to_run)
if err:
raise Exception(err)
audit_str = "Replace disk - Disk with serial number %s brought offline" % serial_number
audit.audit("replace_disk_offline_disk", audit_str,
request)
return_dict["serial_number"] = serial_number
return_dict["node"] = node
return_dict["pool"] = pool
return_dict["old_id"] = disk_id
template = "replace_disk_offlined_conf.html"
elif step == "replace_method":
return_dict["node"] = node
return_dict["serial_number"] = serial_number
template = "replace_disk_method.html"
elif step == "select_replacement_disk":
if 'replacement_method' not in request.POST or request.POST[
'replacement_method'] not in [
'use_existing_disk', 'swap_out_disk'
]:
raise Exception('Invalid request')
return_dict['replacement_method'] = request.POST[
'replacement_method']
return_dict["node"] = node
return_dict["serial_number"] = serial_number
free_disks, err = zfs.get_free_disks()
if err:
raise Exception(err)
if not free_disks:
raise Exception(
'There are no unused disks presently')
return_dict['free_disks'] = free_disks
template = "replace_disk_choose_disk.html"
elif step == "scan_for_new_disk":
# they have replaced the disk so scan for the new disk
# and prompt for a confirmation of the new disk serial
# number
pool = request.POST["pool"]
old_id = request.POST["old_id"]
return_dict["node"] = node
return_dict["serial_number"] = serial_number
return_dict["pool"] = pool
return_dict["old_id"] = old_id
old_disks = si[node]["disks"].keys()
result = False
rc, err = manifest_status.get_disk_info_and_status()
if err:
raise Exception(err)
if rc:
result = True
new_disks = rc
if result:
# print '1'
if new_disks:
# print new_disks.keys()
# print old_disks
for disk in new_disks.keys():
# print disk
if disk not in old_disks:
# print 'new disk : ', disk
return_dict[
"inserted_disk_serial_number"] = disk
return_dict["new_id"] = new_disks[
disk]["id"]
break
if "inserted_disk_serial_number" not in return_dict:
raise Exception(
"Could not detect any new disk. Please check the new disk is inserted and give the system a few seconds to detect the drive and refresh the page to try again."
)
else:
template = "replace_disk_confirm_new_disk.html"
elif step == "online_new_disk":
pool = request.POST["pool"]
old_id = request.POST["old_id"]
new_id = request.POST["new_id"]
new_serial_number = request.POST["new_serial_number"]
common_python_scripts_path, err = config.get_common_python_scripts_path(
)
if err:
raise Exception(err)
cmd_list = []
cmd_list.append({
'Replace old disk':
'zpool replace -f %s %s %s' %
(pool, old_id, new_id)
})
cmd_list.append({
'Online the new disk':
'zpool online -e %s %s' % (pool, new_id)
})
cmd_list.append({
'Regenerate the system configuration':
'%s/generate_manifest.py' %
common_python_scripts_path
})
ret, err = scheduler_utils.create_task(
'Disk replacement',
cmd_list,
task_type_id=1,
attempts=1)
if err:
raise Exception(err)
if not ret:
raise Exception(
'Error scheduling disk replacement tasks')
audit_str = "Replace disk - Scheduled a task for replacing the old disk with serial number %s with the new disk with serial number %s" % (
serial_number, new_serial_number)
audit.audit("replace_disk_replaced_disk", audit_str,
request)
return_dict["node"] = node
return_dict["old_serial_number"] = serial_number
return_dict["new_serial_number"] = new_serial_number
template = "replace_disk_success.html"
return django.shortcuts.render_to_response(
template,
return_dict,
context_instance=django.template.context.
RequestContext(request))
else:
if "serial_number" not in request.POST:
raise Exception(
"Incorrect access method. Please use the menus")
else:
if 'node' in request.POST:
return_dict["node"] = request.POST["node"]
else:
node = si.keys()[0]
return_dict["serial_number"] = request.POST[
"serial_number"]
template = "replace_disk_conf.html"
return django.shortcuts.render_to_response(
template,
return_dict,
context_instance=django.template.context.RequestContext(request))
except Exception, e:
return_dict['base_template'] = "storage_base.html"
return_dict["page_title"] = 'Replace a disk in a ZFS pool'
return_dict['tab'] = 'view_zfs_pools_tab'
return_dict["error"] = 'Error replacing a disk in a ZFS pool'
return_dict["error_details"] = str(e)
return django.shortcuts.render_to_response(
"logged_in_error.html",
return_dict,
context_instance=django.template.context.RequestContext(request))
def create_local_user(username, name, pswd, gid=None, smb_user=True):
try:
# First check if user exists. if so kick out
ul, err = get_local_users()
if ul:
for ud in ul:
if ud["username"] == username:
raise Exception(
"Error creating user. The user \"%s\" already exists. "
% username)
elif err:
raise Exception("Error retrieving user list : %s" % err)
enc_pswd = crypt.crypt(pswd, "28")
use_salt, err = config.use_salt()
if err:
raise Exception(err)
if use_salt:
import salt.client
client = salt.client.LocalClient()
if gid:
rc = client.cmd('*', 'user.add', [username, None, gid])
else:
rc = client.cmd('*', 'user.add', [username])
# print rc
if not rc:
error_list.append("Error creating the username")
for hostname, status in rc.items():
if not status:
error_list.append("Error creating the username on node" %
hostname)
rc = client.cmd('*', 'shadow.set_password', [username, enc_pswd])
for hostname, status in rc.items():
if not status:
error_list.append(
"Error setting the password for username on GRIDCell %s"
% hostname)
rc = client.cmd('*', 'user.chfullname',
[username, "integralstor_user_%s" % name])
for hostname, status in rc.items():
if not status:
error_list.append(
"Error setting the name for username on node %s" %
hostname)
else:
# print '1'
if gid:
cmd_to_run = 'useradd -g %s -p %s -c integralstor_user_%s %s' % (
gid, enc_pswd, name, username)
else:
cmd_to_run = 'useradd -p %s -c integralstor_user_%s %s' % (
enc_pswd, name, username)
lines, err = command.get_command_output(cmd_to_run)
# print '2'
# print lines, err
if err:
raise Exception(err)
if smb_user:
# print '/usr/bin/pdbedit -d 1 -t -a -u %s -f %s'%(username, name), "%s\n%s"%(pswd, pswd)
# Now all set to create samba user
# print '3'
#lines, err = command.get_command_output(r'/usr/bin/pdbedit -d 1 -t -a -u %s -f %s'%(username, name), "%s\n%s"%(pswd, pswd))
(ret, rc), err = command.execute_with_conf_and_rc(
r'/usr/bin/pdbedit -d 1 -t -a -u %s -f %s' %
(username, name), "%s\n%s" % (pswd, pswd))
if rc == 0:
lines, er = command.get_output_list(ret)
if er:
raise Exception(er)
else:
err = ''
tl, er = command.get_output_list(ret)
if er:
raise Exception(er)
if tl:
err = ','.join(tl)
tl, er = command.get_error_list(ret)
if er:
raise Exception(er)
if tl:
err = err + ','.join(tl)
raise Exception(err)
# print '4'
# print lines, err
except Exception, e:
return False, 'Error creating local user : %s' % str(e)
def generate_status_info(path):
status_dict = None
try:
pp = pprint.PrettyPrinter(indent=4)
# First load the status
fqdn = socket.getfqdn()
use_salt, err = config.use_salt()
if err:
raise Exception(err)
if use_salt:
import salt.client
local = salt.client.LocalClient()
sd = local.cmd('*', 'integralstor.status')
else:
tmpsd, err = get_status()
if err:
raise Exception(err)
sd = {}
sd[fqdn] = tmpsd
# pp.pprint(sd)
if not sd:
raise Exception('Did not get a response from salt')
# pp.pprint(sd[fqdn]['interfaces'])
# Load the manifest to check for discrepencies
try:
with open(path, 'r') as f:
md = json.load(f)
except Exception, e:
raise Exception('Error reading the manifest file : %s' % str(e))
# pp.pprint(md)
status_dict = {}
# Match the status against the manifest entries for discrepencies
for hostname, manifest in md.items():
# print hostname, len(hostname)
temp_d = {}
temp_d["errors"] = []
node_status = 0
# print sd.keys()
if hostname not in sd.keys():
node_status = -1
# print "Not found in sd"
else:
if sd[hostname] and 'hardware_specific_dict' in sd[hostname]:
temp_d['hardware_specific_dict'] = sd[hostname][
'hardware_specific_dict']
# Process disk information
disks = {}
for disk_sn, disk in manifest["disks"].items():
# print disk_sn
# print sd[hostname]['disks'].keys()
dd = {}
if disk_sn in sd[hostname]["disks"]:
dd["status"] = sd[hostname]["disks"][disk_sn]["status"]
if dd['status']:
if 'hw_raid' in disk and (not disk['hw_raid']):
if (dd["status"] not in ['PASSED', 'OK']):
node_status = 1
temp_d["errors"].append(
"Disk with serial number %s is reporting SMART errors."
% disk_sn)
elif dd["status"].lower() != 'ok':
node_status = 1
temp_d["errors"].append(
"Disk with serial number %s has errors." %
disk_sn)
dd["name"] = sd[hostname]["disks"][disk_sn]["name"]
else:
sd[hostname]['disks'][disk_sn] = disk
sd[hostname]['disks'][disk_sn][
'status'] = 'Disk missing'
dd["status"] = "Disk Missing"
node_status = 1
temp_d["errors"].append(
"Disk with serial number %s seems to be missing." %
disk_sn)
disks[disk_sn] = dd
new_disk = False
for td in sd[hostname]["disks"].keys():
if td not in manifest["disks"]:
new_disk = True
temp_d["errors"].append(
"New disk detected. Disk with serial number %s seems to be new."
% td)
node_status = 2
#temp_d["disks"] = disks
temp_d["disks"] = sd[hostname]['disks']
# pp.pprint(temp_d['disks'])
if sd[hostname]['services']:
for service_name, service_info in sd[hostname][
'services'].items():
if service_info[0] != 0:
temp_d['errors'].append(
'Service %s seems to have failed.' %
service_name)
node_status = 1
temp_d["services"] = sd[hostname]['services']
# Process interface information
interfaces = {}
for ifname, ifdict in manifest["interfaces"].items():
# Check for all initially present interfaces
id = {}
if ifname in sd[hostname]["interfaces"]:
id, err = _convert_to_status_interface_dict(
sd[hostname]["interfaces"][ifname])
if err:
raise Exception(err)
else:
id["status"] = "Interface Missing"
node_status = 1
temp_d["errors"].append(
"Interface with number %s seems to be missing." %
ifname)
interfaces[ifname] = id
for ifname, ifinfo in sd[hostname]["interfaces"].items():
# Check for all newly created interfaces - bonds, vlans, etc
# print 'ifname is ', ifname
if ifname not in manifest["interfaces"]:
id, err = _convert_to_status_interface_dict(
sd[hostname]["interfaces"][ifname])
if err:
raise Exception(err)
interfaces[ifname] = id
temp_d["interfaces"] = interfaces
for ifname, id in temp_d['interfaces'].items():
if ('ip_configured' in id
or 'slave_to' in id) and id["status"] != 'up':
node_status = 1
temp_d["errors"].append("Interface %s is not up." %
ifname)
# print 'interfaces are ', temp_d['interfaces']
if "memory" in sd[hostname]:
if sd[hostname]["memory"]["mem_total"]["unit"] == "kB":
sd[hostname]["memory"]["mem_total"]["value"] = str(
int(sd[hostname]["memory"]["mem_total"]["value"]) /
1024)
sd[hostname]["memory"]["mem_total"]["unit"] = "MB"
if sd[hostname]["memory"]["mem_free"]["unit"] == "kB":
sd[hostname]["memory"]["mem_free"]["value"] = str(
int(sd[hostname]["memory"]["mem_free"]["value"]) /
1024)
sd[hostname]["memory"]["mem_free"]["unit"] = "MB"
temp_d["memory"] = sd[hostname]["memory"]
if "disk_usage" in sd[hostname]:
temp_d["disk_usage"] = sd[hostname]["disk_usage"]
if "pools" in sd[hostname]:
temp_d["pools"] = sd[hostname]["pools"]
if "load_avg" in sd[hostname]:
# To get around a django quirk of not recognising hyphens
# in dicts
sd[hostname]["load_avg"]["15_min"] = sd[hostname][
"load_avg"]["15-min"]
sd[hostname]["load_avg"]["5_min"] = sd[hostname][
"load_avg"]["5-min"]
sd[hostname]["load_avg"]["1_min"] = sd[hostname][
"load_avg"]["1-min"]
sd[hostname]["load_avg"].pop("15-min", None)
sd[hostname]["load_avg"].pop("5-min", None)
sd[hostname]["load_avg"].pop("1-min", None)
temp_d["load_avg"] = sd[hostname]["load_avg"]
if "cpu_model" in manifest:
temp_d["cpu_model"] = manifest["cpu_model"]
if "fqdn" in manifest:
temp_d["fqdn"] = manifest["fqdn"]
if 'ipmi_status' in sd[hostname]:
temp_d["ipmi_status"] = sd[hostname]["ipmi_status"]
if temp_d["load_avg"]['15_min'] >= temp_d["load_avg"][
'cpu_cores']:
temp_d["errors"].append(
"The 15-minute load average (%.2f) has been high." %
temp_d["load_avg"]['15_min'])
node_status = "Degraded"
if temp_d["load_avg"]['5_min'] >= temp_d["load_avg"][
'cpu_cores']:
temp_d["errors"].append(
"The 5-minute load average (%.2f) has been high." %
temp_d["load_avg"]['5_min'])
if 'ipmi_status' in temp_d:
for status_item in temp_d['ipmi_status']:
if status_item["status"] not in ['ok', 'nr']:
temp_d["errors"].append(
'The %s of the %s is reporting errors' %
(status_item["parameter_name"],
status_item["component_name"]))
pools = temp_d["pools"]
component_status_dict, err = zfs.get_all_components_status(
pools)
if err:
raise Exception(err)
if component_status_dict:
for pool_name, component_status_list in component_status_dict.items(
):
msg = None
for component in component_status_list:
if 'status' in component and 'state' in component[
'status'] and component['status'][
'state'] != 'ONLINE':
if not msg:
msg = "The ZFS pool '%s' has the following issue(s) : " % pool_name
msg += "The component %s of type '%s' has a state of '%s'. " % (
component['name'], component['type'],
component['status']['state'])
if msg:
temp_d['errors'].append(msg)
temp_d['zfs_version'] = sd[hostname]['zfs_version']
temp_d['os_version'] = sd[hostname]['os_version']
temp_d["node_status"] = node_status
if node_status == 0:
temp_d["node_status_str"] = "Healthy"
elif node_status == 1:
temp_d["node_status_str"] = "Degraded"
elif node_status == 2:
temp_d["node_status_str"] = "New on-node hardware detected"
elif node_status == -1:
temp_d["node_status_str"] = "No response. Down?"
temp_d['errors'].append('Node %s seems to be down' % hostname)
status_dict[hostname] = temp_d
def generate_manifest_info():
"""Generate a dictionary containing all manifest information. Will be dumped into the master.manifest file in a json format."""
manifest_dict = {}
try:
#pp = pprint.PrettyPrinter(indent=4)
use_salt, err = config.use_salt()
if err:
raise Exception(err)
fqdn = socket.getfqdn()
if use_salt:
import salt.modules.network
import salt.modules.ps
import salt.modules.status
import salt.client
import salt.wheel
import salt.config
local = salt.client.LocalClient()
cfg, err = config.get_salt_master_config()
if err:
raise Exception(err)
opts = salt.config.master_config(cfg)
wheel = salt.wheel.Wheel(opts)
keys = wheel.call_func('key.list_all')
if not keys:
raise Exception('No GRIDCells found!')
nodes = keys['minions']
# print nodes
for node in nodes:
manifest_dict[node] = {}
roles = local.cmd('*', 'grains.item', ['roles'])
for node, info in roles.items():
if node not in manifest_dict:
manifest_dict[node] = {}
manifest_dict[node]['roles'] = info['roles']
ret = local.cmd('*', 'integralstor.status')
for node, info in ret.items():
if node not in manifest_dict:
manifest_dict[node] = {}
manifest_dict[node]['cpu_model'] = info['cpu_model']
manifest_dict[node]['disks'] = info['disks']
for dn, dv in manifest_dict[node]['disks'].items():
if 'pool' in dv:
dv.pop('pool')
manifest_dict[node]['interfaces'] = info['interfaces']
manifest_dict[node]['memory'] = info['memory']
manifest_dict[node]['fqdn'] = info['fqdn']
if 'hardware_specific_dict' in info:
manifest_dict[node]['hardware_specific_dict'] = info[
'hardware_specific_dict']
else:
# Single node so get the info using a direct call and just bung it
# into the fqdn key!
manifest_dict[fqdn] = {}
status_dict, err = get_status()
if err:
raise Exception(err)
manifest_dict[fqdn]['cpu_model'] = status_dict['cpu_model']
manifest_dict[fqdn]['disks'] = status_dict['disks']
for dn, dv in manifest_dict[fqdn]['disks'].items():
if 'pool' in dv:
dv.pop('pool')
manifest_dict[fqdn]['interfaces'] = status_dict['interfaces']
manifest_dict[fqdn]['memory'] = status_dict['memory']
manifest_dict[fqdn]['fqdn'] = fqdn
# Remove transitory info and only keep the actual hardware info
for node in manifest_dict.keys():
if 'interfaces' in manifest_dict[node]:
for int_name, interface in manifest_dict[node][
'interfaces'].items():
if 'up' in interface:
interface.pop('up')
if 'inet' in interface:
interface.pop('inet')
if 'disks' in manifest_dict[node]:
for disk_name, diskinfo in manifest_dict[node]['disks'].items(
):
if 'status' in diskinfo:
diskinfo.pop('status')
if 'memory' in manifest_dict[node]:
if 'mem_free' in manifest_dict[node]['memory']:
manifest_dict[node]['memory'].pop('mem_free')
if not manifest_dict:
raise Exception('Error getting manifest information')
except Exception, e:
return None, 'Error generating the manifest dictionary : %s' % str(e)
def generate_status_info(path):
status_dict = None
try:
pp = pprint.PrettyPrinter(indent=4)
# First load the status
fqdn = socket.getfqdn()
use_salt, err = config.use_salt()
if err:
raise Exception(err)
if use_salt:
import salt.client
local = salt.client.LocalClient()
sd = local.cmd('*', 'integralstor.status')
else:
tmpsd, err = get_status()
if err:
raise Exception(err)
sd = {}
sd[fqdn] = tmpsd
# pp.pprint(sd)
if not sd:
raise Exception('Did not get a response from salt')
# pp.pprint(sd[fqdn]['interfaces'])
# Load the manifest to check for discrepencies
try:
with open(path, 'r') as f:
md = json.load(f)
except Exception, e:
raise Exception('Error reading the manifest file : %s' % str(e))
# pp.pprint(md)
status_dict = {}
# Match the status against the manifest entries for discrepencies
for hostname, manifest in md.items():
# print hostname, len(hostname)
temp_d = {}
temp_d["errors"] = []
node_status = 0
# print sd.keys()
if hostname not in sd.keys():
node_status = -1
# print "Not found in sd"
else:
if sd[hostname] and 'hardware_specific_dict' in sd[hostname]:
temp_d['hardware_specific_dict'] = sd[hostname]['hardware_specific_dict']
# Process disk information
disks = {}
for disk_sn, disk in manifest["disks"].items():
# print disk_sn
# print sd[hostname]['disks'].keys()
dd = {}
if disk_sn in sd[hostname]["disks"]:
dd["status"] = sd[hostname]["disks"][disk_sn]["status"]
if dd['status']:
if 'hw_raid' in disk and (not disk['hw_raid']):
if (dd["status"] not in ['PASSED', 'OK']):
node_status = 1
temp_d["errors"].append(
"Disk with serial number %s is reporting SMART errors." % disk_sn)
elif dd["status"].lower() != 'ok':
node_status = 1
temp_d["errors"].append(
"Disk with serial number %s has errors." % disk_sn)
dd["name"] = sd[hostname]["disks"][disk_sn]["name"]
else:
sd[hostname]['disks'][disk_sn] = disk
sd[hostname]['disks'][disk_sn]['status'] = 'Disk missing'
dd["status"] = "Disk Missing"
node_status = 1
temp_d["errors"].append(
"Disk with serial number %s seems to be missing." % disk_sn)
disks[disk_sn] = dd
new_disk = False
for td in sd[hostname]["disks"].keys():
if td not in manifest["disks"]:
new_disk = True
temp_d["errors"].append(
"New disk detected. Disk with serial number %s seems to be new." % td)
node_status = 2
#temp_d["disks"] = disks
temp_d["disks"] = sd[hostname]['disks']
# pp.pprint(temp_d['disks'])
if sd[hostname]['services']:
for service_name, service_info in sd[hostname]['services'].items():
if service_info[0] != 0:
temp_d['errors'].append(
'Service %s seems to have failed.' % service_name)
node_status = 1
temp_d["services"] = sd[hostname]['services']
# Process interface information
interfaces = {}
for ifname, ifdict in manifest["interfaces"].items():
# Check for all initially present interfaces
id = {}
if ifname in sd[hostname]["interfaces"]:
id, err = _convert_to_status_interface_dict(
sd[hostname]["interfaces"][ifname])
if err:
raise Exception(err)
else:
id["status"] = "Interface Missing"
node_status = 1
temp_d["errors"].append(
"Interface with number %s seems to be missing." % ifname)
interfaces[ifname] = id
for ifname, ifinfo in sd[hostname]["interfaces"].items():
# Check for all newly created interfaces - bonds, vlans, etc
# print 'ifname is ', ifname
if ifname not in manifest["interfaces"]:
id, err = _convert_to_status_interface_dict(
sd[hostname]["interfaces"][ifname])
if err:
raise Exception(err)
interfaces[ifname] = id
temp_d["interfaces"] = interfaces
for ifname, id in temp_d['interfaces'].items():
if ('ip_configured' in id or 'slave_to' in id)and id["status"] != 'up':
node_status = 1
temp_d["errors"].append(
"Interface %s is not up." % ifname)
# print 'interfaces are ', temp_d['interfaces']
if "memory" in sd[hostname]:
if sd[hostname]["memory"]["mem_total"]["unit"] == "kB":
sd[hostname]["memory"]["mem_total"]["value"] = str(
int(sd[hostname]["memory"]["mem_total"]["value"]) / 1024)
sd[hostname]["memory"]["mem_total"]["unit"] = "MB"
if sd[hostname]["memory"]["mem_free"]["unit"] == "kB":
sd[hostname]["memory"]["mem_free"]["value"] = str(
int(sd[hostname]["memory"]["mem_free"]["value"]) / 1024)
sd[hostname]["memory"]["mem_free"]["unit"] = "MB"
temp_d["memory"] = sd[hostname]["memory"]
if "disk_usage" in sd[hostname]:
temp_d["disk_usage"] = sd[hostname]["disk_usage"]
if "pools" in sd[hostname]:
temp_d["pools"] = sd[hostname]["pools"]
if "load_avg" in sd[hostname]:
# To get around a django quirk of not recognising hyphens
# in dicts
sd[hostname]["load_avg"]["15_min"] = sd[hostname]["load_avg"]["15-min"]
sd[hostname]["load_avg"]["5_min"] = sd[hostname]["load_avg"]["5-min"]
sd[hostname]["load_avg"]["1_min"] = sd[hostname]["load_avg"]["1-min"]
sd[hostname]["load_avg"].pop("15-min", None)
sd[hostname]["load_avg"].pop("5-min", None)
sd[hostname]["load_avg"].pop("1-min", None)
temp_d["load_avg"] = sd[hostname]["load_avg"]
if "cpu_model" in manifest:
temp_d["cpu_model"] = manifest["cpu_model"]
if "fqdn" in manifest:
temp_d["fqdn"] = manifest["fqdn"]
if 'ipmi_status' in sd[hostname]:
temp_d["ipmi_status"] = sd[hostname]["ipmi_status"]
if temp_d["load_avg"]['15_min'] >= temp_d["load_avg"]['cpu_cores']:
temp_d["errors"].append(
"The 15-minute load average (%.2f) has been high." % temp_d["load_avg"]['15_min'])
node_status = "Degraded"
if temp_d["load_avg"]['5_min'] >= temp_d["load_avg"]['cpu_cores']:
temp_d["errors"].append(
"The 5-minute load average (%.2f) has been high." % temp_d["load_avg"]['5_min'])
if 'ipmi_status' in temp_d:
for status_item in temp_d['ipmi_status']:
if status_item["status"] not in ['ok', 'nr']:
temp_d["errors"].append('The %s of the %s is reporting errors' % (
status_item["parameter_name"], status_item["component_name"]))
pools = temp_d["pools"]
component_status_dict, err = zfs.get_all_components_status(
pools)
if err:
raise Exception(err)
if component_status_dict:
for pool_name, component_status_list in component_status_dict.items():
msg = None
for component in component_status_list:
if 'status' in component and 'state' in component['status'] and component['status']['state'] != 'ONLINE':
if not msg:
msg = "The ZFS pool '%s' has the following issue(s) : " % pool_name
msg += "The component %s of type '%s' has a state of '%s'. " % (
component['name'], component['type'], component['status']['state'])
if msg:
temp_d['errors'].append(msg)
temp_d['zfs_version'] = sd[hostname]['zfs_version']
temp_d['os_version'] = sd[hostname]['os_version']
temp_d["node_status"] = node_status
if node_status == 0:
temp_d["node_status_str"] = "Healthy"
elif node_status == 1:
temp_d["node_status_str"] = "Degraded"
elif node_status == 2:
temp_d["node_status_str"] = "New on-node hardware detected"
elif node_status == -1:
temp_d["node_status_str"] = "No response. Down?"
temp_d['errors'].append('Node %s seems to be down' % hostname)
status_dict[hostname] = temp_d
def generate_manifest_info():
"""Generate a dictionary containing all manifest information. Will be dumped into the master.manifest file in a json format."""
manifest_dict = {}
try:
#pp = pprint.PrettyPrinter(indent=4)
use_salt, err = config.use_salt()
if err:
raise Exception(err)
fqdn = socket.getfqdn()
if use_salt:
import salt.modules.network
import salt.modules.ps
import salt.modules.status
import salt.client
import salt.wheel
import salt.config
local = salt.client.LocalClient()
cfg, err = config.get_salt_master_config()
if err:
raise Exception(err)
opts = salt.config.master_config(cfg)
wheel = salt.wheel.Wheel(opts)
keys = wheel.call_func('key.list_all')
if not keys:
raise Exception('No GRIDCells found!')
nodes = keys['minions']
# print nodes
for node in nodes:
manifest_dict[node] = {}
roles = local.cmd('*', 'grains.item', ['roles'])
for node, info in roles.items():
if node not in manifest_dict:
manifest_dict[node] = {}
manifest_dict[node]['roles'] = info['roles']
ret = local.cmd('*', 'integralstor.status')
for node, info in ret.items():
if node not in manifest_dict:
manifest_dict[node] = {}
manifest_dict[node]['cpu_model'] = info['cpu_model']
manifest_dict[node]['disks'] = info['disks']
for dn, dv in manifest_dict[node]['disks'].items():
if 'pool' in dv:
dv.pop('pool')
manifest_dict[node]['interfaces'] = info['interfaces']
manifest_dict[node]['memory'] = info['memory']
manifest_dict[node]['fqdn'] = info['fqdn']
if 'hardware_specific_dict' in info:
manifest_dict[node]['hardware_specific_dict'] = info['hardware_specific_dict']
else:
# Single node so get the info using a direct call and just bung it
# into the fqdn key!
manifest_dict[fqdn] = {}
status_dict, err = get_status()
if err:
raise Exception(err)
manifest_dict[fqdn]['cpu_model'] = status_dict['cpu_model']
manifest_dict[fqdn]['disks'] = status_dict['disks']
for dn, dv in manifest_dict[fqdn]['disks'].items():
if 'pool' in dv:
dv.pop('pool')
manifest_dict[fqdn]['interfaces'] = status_dict['interfaces']
manifest_dict[fqdn]['memory'] = status_dict['memory']
manifest_dict[fqdn]['fqdn'] = fqdn
# Remove transitory info and only keep the actual hardware info
for node in manifest_dict.keys():
if 'interfaces' in manifest_dict[node]:
for int_name, interface in manifest_dict[node]['interfaces'].items():
if 'up' in interface:
interface.pop('up')
if 'inet' in interface:
interface.pop('inet')
if 'disks' in manifest_dict[node]:
for disk_name, diskinfo in manifest_dict[node]['disks'].items():
if 'status' in diskinfo:
diskinfo.pop('status')
if 'memory' in manifest_dict[node]:
if 'mem_free' in manifest_dict[node]['memory']:
manifest_dict[node]['memory'].pop('mem_free')
if not manifest_dict:
raise Exception('Error getting manifest information')
except Exception, e:
return None, 'Error generating the manifest dictionary : %s' % str(e)