def run_zfs_remote_replication(remote_replication_id):
try:
rr, err = remote_replication.get_remote_replications(
remote_replication_id)
if err:
raise Exception('Could not fetch replication details: %s' % err)
replication = rr[0]
mode = replication['mode']
if mode == 'zfs':
now_local_epoch, err = datetime_utils.get_epoch(when='now')
if err:
raise Exception(err)
now_local_str, err = datetime_utils.convert_from_epoch(
now_local_epoch, return_format='str', str_format='%Y%m%d%H%M', to='local')
if err:
raise Exception(err)
source_dataset = replication['zfs'][0]['source_dataset']
ret, err = zfs.create_snapshot(
source_dataset, 'zrr_%s_%s' % (remote_replication_id, now_local_str))
if err:
raise Exception(err)
ret, err = remote_replication.run_zfs_remote_replication(
remote_replication_id)
if err:
raise Exception(err)
else:
raise Exception('Invalid remote replication mode')
except Exception, e:
return False, 'Error adding ZFS remote replication task : %s' % e
def _get_repeat_entry(alert_dict, past_x_seconds=900):
"""Check the db to see if the passed alert is a repeat of an existing alert entry in the past x seconds
"""
result = None
try:
db_path, err = config.get_db_path()
if err:
raise Exception(err)
now, err = datetime_utils.get_epoch(when='now', num_previous_days=0)
if alert_dict['component']:
query = "select * from alerts where severity_type_id = '%d' and subsystem_type_id = '%d' and component = '%s' and alert_str=\"%s\" and first_alert_time >= %d;" % (
alert_dict['severity_type_id'],
alert_dict['subsystem_type_id'], alert_dict['component'],
alert_dict['alert_str'], now - past_x_seconds)
else:
query = "select * from alerts where severity_type_id = '%d' and subsystem_type_id = '%d' and alert_str=\"%s\" and first_alert_time >= %d);" % (
alert_dict['severity_type_id'],
alert_dict['subsystem_type_id'], alert_dict['alert_str'],
now - past_x_seconds)
# print query
result, err = db.get_single_row(db_path, query)
if err:
raise Exception(err)
# print result, err
except Exception, e:
return None, 'Error checking for repeats : %s' % str(e)
def run_zfs_remote_replication(remote_replication_id):
try:
rr, err = remote_replication.get_remote_replications(
remote_replication_id)
if err:
raise Exception('Could not fetch replication details: %s' % err)
replication = rr[0]
mode = replication['mode']
if mode == 'zfs':
now_local_epoch, err = datetime_utils.get_epoch(when='now')
if err:
raise Exception(err)
now_local_str, err = datetime_utils.convert_from_epoch(
now_local_epoch,
return_format='str',
str_format='%Y%m%d%H%M',
to='local')
if err:
raise Exception(err)
source_dataset = replication['zfs'][0]['source_dataset']
ret, err = zfs.create_snapshot(
source_dataset,
'zrr_%s_%s' % (remote_replication_id, now_local_str))
if err:
raise Exception(err)
ret, err = remote_replication.run_zfs_remote_replication(
remote_replication_id)
if err:
raise Exception(err)
else:
raise Exception('Invalid remote replication mode')
except Exception, e:
return False, 'Error adding ZFS remote replication task : %s' % e
def remove_old_files(lg=None, older_than_days=7):
try:
lck, err = lock.get_lock('remove_old_files')
if err:
raise Exception(err)
if not lck:
raise Exception('Could not acquire lock.')
exported_logs_dir, err = config.get_exported_logs_dir_path()
if err:
raise Exception(err)
status_report_dir, err = config.get_status_reports_dir_path()
if err:
raise Exception(err)
patterns = [exported_logs_dir + '/alerts_*', status_report_dir + '/*']
now, err = datetime_utils.get_epoch()
# print now
for pattern in patterns:
list = glob.glob(pattern)
for f in list:
# print f
ctime = os.path.getctime(f)
# print os.path.getctime(f)
if (now - ctime) > (60 * 60 * 24 * older_than_days):
# print 'removing %s'%f
os.remove(f)
else:
# print 'not removing %s'%f
pass
except Exception, e:
logger.log_or_print('Error removing old files: %s' %
e, lg, level='critical')
lock.release_lock('remove_old_files')
return -1, str(e)
def remove_old_files(lg=None, older_than_days=7):
try:
lck, err = lock.get_lock('remove_old_files')
if err:
raise Exception(err)
if not lck:
raise Exception('Could not acquire lock.')
exported_logs_dir, err = config.get_exported_logs_dir_path()
if err:
raise Exception(err)
status_report_dir, err = config.get_status_reports_dir_path()
if err:
raise Exception(err)
patterns = [exported_logs_dir + '/alerts_*', status_report_dir + '/*']
now, err = datetime_utils.get_epoch()
# print now
for pattern in patterns:
list = glob.glob(pattern)
for f in list:
# print f
ctime = os.path.getctime(f)
# print os.path.getctime(f)
if (now - ctime) > (60 * 60 * 24 * older_than_days):
# print 'removing %s'%f
os.remove(f)
else:
# print 'not removing %s'%f
pass
except Exception, e:
logger.log_or_print('Error removing old files: %s' % e,
lg,
level='critical')
lock.release_lock('remove_old_files')
return -1, str(e)
def record_alerts(alerts_list):
"""Record a set of passed list containing alert dictionaries into the DB."""
try:
db_path, err = config.get_db_path()
if err:
raise Exception(err)
if alerts_list:
now, err = datetime_utils.get_epoch(
when='now', num_previous_days=0)
if err:
raise Exception(er)
for alert in alerts_list:
# print alert
command_list = []
result, err = _get_repeat_entry(alert)
# print 'repeat', result, err
if err:
raise Exception(err)
update_alert_id = None
if result:
cmd = ['update alerts set repeat_count=%d, last_update_time="%d" where alert_id=%d' % (
int(result['repeat_count']) + 1, now, result['alert_id'])]
update_alert_id = result['alert_id']
# print 'updating!', cmd
else:
if 'component' not in alert or (not alert['component']):
cmd = [
'insert into alerts(first_alert_time, last_update_time, severity_type_id, subsystem_type_id, alert_str) values (?,?,?,?,?)', (now, now, alert['severity_type_id'], alert['subsystem_type_id'], alert['alert_str'],)]
else:
cmd = [
'insert into alerts(first_alert_time, last_update_time, severity_type_id, subsystem_type_id, component, alert_str) values (?,?,?,?,?,?)', (now, now, alert['severity_type_id'], alert['subsystem_type_id'], alert['component'], alert['alert_str'],)]
command_list.append(cmd)
# print command_list
rowid, err = db.execute_iud(
db_path, command_list, get_rowid=True)
if err:
raise Exception(err)
if update_alert_id:
alert_id = update_alert_id
else:
alert_id = rowid
ret, err = event_notifications.record_event_notification_holding(
event_id=alert_id, event_type_id=1, subsystem_type_id=alert['subsystem_type_id'], severity_type_id=alert['severity_type_id'])
# print ret, err
if err:
raise Exception(err)
except Exception, e:
# print str(e)
return False, 'Error recording alerts : %s' % str(e)
def main():
lg = None
try:
scripts_log, err = config.get_scripts_log_path()
if err:
raise Exception(err)
lg, err = logger.get_script_logger(
'Current logs archive generation', scripts_log, level=logging.DEBUG)
logs_archives_dir, err = config.get_logs_archives_dir_path()
if err:
raise Exception(err)
lck, err = lock.get_lock('generate_current_logs_archive')
if err:
raise Exception(err)
if not lck:
raise Exception('Could not acquire lock.')
logger.log_or_print(
'Current logs archive generation initiated.', lg, level='info')
now_local_epoch, err = datetime_utils.get_epoch(when='now')
if err:
raise Exception(err)
now_local_str, err = datetime_utils.convert_from_epoch(
now_local_epoch, return_format='str', str_format='%Y_%m_%d_%H_%M', to='local')
if err:
raise Exception(err)
zf_name = 'IntegralSTOR_system_logs_%s.zip' % now_local_str
try:
os.makedirs(logs_archives_dir)
except:
pass
zf = zipfile.ZipFile('%s/%s' % (logs_archives_dir, zf_name), 'w')
for root, dirs, files in os.walk('/var/log/integralstor'):
if root.startswith('/var/log/integralstor/archives'):
continue
for file in files:
# print '%s/%s'%(root[len('/var/log/integralstor/'):], file)
zf.write(os.path.join(root, file), '%s/%s' %
(root[len('/var/log/integralstor/'):], file))
zf.close()
except Exception, e:
# print str(e)
lock.release_lock('generate_current_logs_archive')
logger.log_or_print('Error generating current logs archive : %s' %
e, lg, level='critical')
return -1, 'Error generating current logs archive: %s' % e
def purge_email_queue(older_than_days):
try:
cutoff_seconds, err = datetime_utils.get_epoch(
when='now', num_previous_days=older_than_days)
if err:
raise Exception(err)
db_path, err = config.get_db_path()
if err:
raise Exception(err)
delete_command_list = [
["delete from email_queue where create_time < %d;" % cutoff_seconds]]
ret, err = db.execute_iud(db_path, delete_command_list)
if err:
raise Exception(err)
except Exception, e:
return False, 'Error purging email queue database: %s' % str(e)
def process_email_queue():
"""Send out all the mails that are pending in the email_queue table and update the status accordingly.
"""
try:
d, err = load_email_settings()
if err:
raise Exception(err)
if d:
eq_list, err = get_queued_emails()
# print eq_list, err
if err:
raise Exception(err)
if eq_list:
db_path, err = config.get_db_path()
if err:
raise Exception(err)
now, err = datetime_utils.get_epoch(
when='now', num_previous_days=0)
if err:
raise Exception(er)
for eq_entry in eq_list:
if 'attachment_file_location' in eq_entry and eq_entry['attachment_file_location']:
attachment_file_list = [
eq_entry['attachment_file_location']]
else:
attachment_file_list = None
result, err = send_mail(d['server'], d['port'], d['username'], d['pswd'], d['tls'],
eq_entry['recipients'], eq_entry['subject'], eq_entry['body'], attachment_file_list)
if err:
if eq_entry['attempts'] == 2:
# Failed
cmd = ['update email_queue set attempts=2, status=3, last_attempt_time="%s", error_message="%s" where eq_id=%d' % (
now, err, eq_entry['eq_id'])]
else:
# Retry once more
cmd = ['update email_queue set attempts=%d, status=2, last_attempt_time="%s", error_message="%s" where eq_id=%d' % (
((int)(eq_entry['attempts']) + 1), now, err, eq_entry['eq_id'])]
else:
# Success
cmd = ['update email_queue set status=0, last_attempt_time="%s" where eq_id=%d' % (
now, eq_entry['eq_id'])]
ret, err = db.execute_iud(db_path, [cmd])
if err:
raise Exception(err)
except Exception, e:
return False, 'Error processing email queue: %s' % str(e)
def update_last_view_time(request):
"""Called to update the last time a particular user opened their alerts..
"""
try:
db_path, err = config.get_db_path()
if err:
raise Exception(err)
now, err = datetime_utils.get_epoch(when='now', num_previous_days=0)
if err:
raise Exception(er)
cmd_list = []
cmd = ["INSERT OR REPLACE INTO admin_alerts (user, last_refresh_time) values (?,?);", (
request.user.username, now)]
cmd_list.append(cmd)
test, err = db.execute_iud(db_path, cmd_list)
if err:
raise Exception(err)
except Exception, e:
return False, 'Error updating last view time for alerts : %s' % str(e)
def update_last_view_time(request):
"""Called to update the last time a particular user opened their alerts..
"""
try:
db_path, err = config.get_db_path()
if err:
raise Exception(err)
now, err = datetime_utils.get_epoch(when='now', num_previous_days=0)
if err:
raise Exception(er)
cmd_list = []
cmd = [
"INSERT OR REPLACE INTO admin_alerts (user, last_refresh_time) values (?,?);",
(request.user.username, now)
]
cmd_list.append(cmd)
test, err = db.execute_iud(db_path, cmd_list)
if err:
raise Exception(err)
except Exception, e:
return False, 'Error updating last view time for alerts : %s' % str(e)
def _get_repeat_entry(alert_dict, past_x_seconds=900):
"""Check the db to see if the passed alert is a repeat of an existing alert entry in the past x seconds
"""
result = None
try:
db_path, err = config.get_db_path()
if err:
raise Exception(err)
now, err = datetime_utils.get_epoch(when='now', num_previous_days=0)
if alert_dict['component']:
query = "select * from alerts where severity_type_id = '%d' and subsystem_type_id = '%d' and component = '%s' and alert_str=\"%s\" and first_alert_time >= %d;" % (
alert_dict['severity_type_id'], alert_dict['subsystem_type_id'], alert_dict['component'], alert_dict['alert_str'], now - past_x_seconds)
else:
query = "select * from alerts where severity_type_id = '%d' and subsystem_type_id = '%d' and alert_str=\"%s\" and first_alert_time >= %d);" % (
alert_dict['severity_type_id'], alert_dict['subsystem_type_id'], alert_dict['alert_str'], now - past_x_seconds)
# print query
result, err = db.get_single_row(db_path, query)
if err:
raise Exception(err)
# print result, err
except Exception, e:
return None, 'Error checking for repeats : %s' % str(e)
def audit(audit_code, audit_str, request, system_initiated=False):
"""Record an audit entry in the db
"""
try:
db_path, err = config.get_db_path()
if err:
raise Exception(err)
if system_initiated is False:
ip, err = networking.get_client_ip(request.META)
if err:
raise Exception(err)
now, err = datetime_utils.get_epoch(when='now', num_previous_days=0)
if err:
raise Exception(er)
if system_initiated:
username = '******'
source_ip = 'System'
else:
username = request.user.username
source_ip, err = networking.get_client_ip(request.META)
if err:
raise Exception(err)
command_list = []
cmd = [
'insert into audit(audit_time, username, source_ip, audit_code, audit_str) values (?,?,?,?,?)', (now, username, source_ip, audit_code, audit_str,)]
command_list.append(cmd)
audit_id, err = db.execute_iud(db_path, command_list, get_rowid=True)
if err:
raise Exception(err)
ret, err = event_notifications.record_event_notification_holding(
event_id=audit_id, event_type_id=2)
if err:
raise Exception(err)
except Exception, e:
return False, 'Error performing an audit operation : %s' % str(e)
def process_tasks(node=socket.getfqdn()):
"""When called, processes/runs subtasks of each entry from tasks table if they satisfy/pass the required checks like status, last_run_time, retries, etc."""
'''
TODO
- Needs a better docstring comment beriefly explaning what the function does
'''
try:
error_list = []
db_path, err = config.get_db_path()
if err:
raise Exception(err)
# now = int(time.time())
now, err = datetime_utils.get_epoch(when='now')
if err:
raise Exception(err)
tasks_query = "select * from tasks where node == '" + node + \
"' and (status == 'error-retrying' or status == 'queued') and (initiate_time <= '%d');" % (now)
tasks_to_process, err = db.get_multiple_rows(db_path, tasks_query)
if err:
raise Exception(err)
if tasks_to_process is not None:
for task in tasks_to_process:
ret, err = run_task(task['task_id'])
if err:
error_list.append(str(err))
if error_list:
raise Exception(str(error_list))
except Exception as e:
return False, 'Error processing tasks: %s' % e
else:
return True, None
def process_tasks(node=socket.getfqdn()):
"""When called, processes/runs subtasks of each entry from tasks table if they satisfy/pass the required checks like status, last_run_time, retries, etc."""
'''
TODO
- Needs a better docstring comment beriefly explaning what the function does
'''
try:
error_list = []
db_path, err = config.get_db_path()
if err:
raise Exception(err)
# now = int(time.time())
now, err = datetime_utils.get_epoch(when='now')
if err:
raise Exception(err)
tasks_query = "select * from tasks where node == '" + node + \
"' and (status == 'error-retrying' or status == 'queued') and (initiate_time <= '%d');" % (now)
tasks_to_process, err = db.get_multiple_rows(db_path, tasks_query)
if err:
raise Exception(err)
if tasks_to_process is not None:
for task in tasks_to_process:
ret, err = run_task(task['task_id'])
if err:
error_list.append(str(err))
if error_list:
raise Exception(str(error_list))
except Exception as e:
return False, 'Error processing tasks: %s' % e
else:
return True, None
def audit(audit_code, audit_str, request, system_initiated=False):
"""Record an audit entry in the db
"""
try:
db_path, err = config.get_db_path()
if err:
raise Exception(err)
if system_initiated is False:
ip, err = networking.get_client_ip(request.META)
if err:
raise Exception(err)
now, err = datetime_utils.get_epoch(when='now', num_previous_days=0)
if err:
raise Exception(er)
if system_initiated:
username = '******'
source_ip = 'System'
else:
username = request.user.username
source_ip, err = networking.get_client_ip(request.META)
if err:
raise Exception(err)
command_list = []
cmd = [
'insert into audit(audit_time, username, source_ip, audit_code, audit_str) values (?,?,?,?,?)', (now, username, source_ip, audit_code, audit_str,)]
command_list.append(cmd)
audit_id, err = db.execute_iud(db_path, command_list, get_rowid=True)
if err:
raise Exception(err)
ret, err = event_notifications.record_event_notification_holding(
event_id=audit_id, event_type_id=2)
if err:
raise Exception(err)
except Exception, e:
return False, 'Error performing an audit operation : %s' % str(e)
def enqueue(recipients, subject, body, attachment_file_location=None, delete_attachment_file=True):
"""Enter an entry into the email queue table
"""
try:
db_path, err = config.get_db_path()
if err:
raise Exception(err)
now, err = datetime_utils.get_epoch(when='now', num_previous_days=0)
if err:
raise Exception(er)
command_list = []
if attachment_file_location:
cmd = [
'insert into email_queue(recipients, subject, body, attachment_file_location, delete_attachment_file, create_time, status) values (?,?,?,?,?,?,?)', (recipients, subject, body, attachment_file_location, delete_attachment_file, now, 1,)]
else:
cmd = [
'insert into email_queue(recipients, subject, body, create_time, status) values (?,?,?,?,?)', (recipients, subject, body, now, 1,)]
command_list.append(cmd)
ret, err = db.execute_iud(db_path, command_list)
if err:
raise Exception(err)
except Exception, e:
return False, 'Error enqueuing email: %s' % str(e)
def _record_pool_usage_stats():
try:
pool_list, err = zfs.get_pools()
if err:
raise Exception(err)
if pool_list:
midnight, err = datetime_utils.get_epoch(when='midnight')
if err:
raise Exception(err)
db_path, err = config.get_db_path()
if err:
raise Exception(err)
for pool_info in pool_list:
cmd_list = []
cmd_list.append(["insert into pool_usage_stats(pool_name, date, used_bytes, available_bytes) values (?,?,?,?)", (
pool_info['pool_name'], midnight, pool_info['usage']['total_space_used_bytes'], pool_info['usage']['total_space_avail_bytes'],)])
# Run multiple times as a duplicate entry will cause other
# inserts to fail otherwise..
ret, err = db.execute_iud(db_path, cmd_list)
# Best effort.. continue if duplicate dates cause a problem when rerunning
# print ret, err
except Exception, e:
# print str(e)
return False, "Error recording ZFS pool usage statistics : %s" % str(e)
def main():
lg = None
try:
scripts_log, err = config.get_scripts_log_path()
if err:
raise Exception(err)
lg, err = logger.get_script_logger(
'Current configuration archive generation',
scripts_log,
level=logging.DEBUG)
config_archives_dir, err = config.get_config_archives_dir_path()
if err:
raise Exception(err)
lck, err = lock.get_lock('generate_current_config_archive')
if err:
raise Exception(err)
if not lck:
raise Exception('Could not acquire lock.')
logger.log_or_print('Current config archive generation initiated.',
lg,
level='info')
db_path, err = config.get_db_path()
if err:
raise Exception(err)
pki_dir, err = config.get_pki_dir()
if err:
raise Exception(err)
config_file_list = [('/etc/samba/smb.conf', 'smb.conf'),
('/etc/krb5.conf', 'krb5.conf'),
(db_path, 'integral_view_config.db'),
('/etc/exports', 'exports'),
('/etc/vsftpd/vsftpd.conf', 'vsftpd.conf'),
('/etc/tgt/targets.conf', 'targets.conf'),
('/etc/resolv.conf', 'resolv.conf'),
('/etc/hosts', 'hosts'), ('/etc/passwd', 'passwd'),
('/etc/group', 'group')]
config_dir_list = [(pki_dir, 'pki')]
now_local_epoch, err = datetime_utils.get_epoch(when='now')
if err:
raise Exception(err)
now_local_str, err = datetime_utils.convert_from_epoch(
now_local_epoch,
return_format='str',
str_format='%Y_%m_%d_%H_%M',
to='local')
if err:
raise Exception(err)
zf_name = 'IntegralSTOR_system_configuration_%s.zip' % now_local_str
try:
os.makedirs(config_archives_dir)
except:
pass
try:
zf = zipfile.ZipFile('%s/%s' % (config_archives_dir, zf_name), 'w')
for entry in config_file_list:
if os.path.exists(entry[0]):
zf.write(entry[0], arcname=entry[1])
for entry in config_dir_list:
if os.path.exists(entry[0]):
if entry[0][-1] == '/':
path = entry[0][:-1]
else:
path = entry[0]
for root, dirs, files in os.walk(path):
base = root[len(path) + 1:]
for file in files:
if base:
zf.write(os.path.join(root, file),
'%s/%s/%s' % (entry[1], base, file))
else:
zf.write(os.path.join(root, file),
'%s/%s' % (entry[1], file))
zf.close()
except Exception as e:
raise Exception("Error compressing log file : %s" % str(e))
except Exception, e:
# print str(e)
lock.release_lock('generate_current_config_archive')
logger.log_or_print('Error generating current config archive : %s' % e,
lg,
level='critical')
return -1, 'Error generating current config archive: %s' % e
def generate_global_header(f):
try:
ep, err = datetime_utils.get_epoch(when='now')
if err:
raise Exception(err)
date_str, err = datetime_utils.convert_from_epoch(
ep, return_format='str', str_format='%Y/%m/%d %H:%M', to='local')
if err:
raise Exception(err)
ver, err = config.get_version()
if err:
raise Exception(err)
uuid, err = system_info.get_integralstor_uuid()
if err:
raise Exception(err)
org_info, err = system_info.get_org_info()
if err:
raise Exception(err)
f.write('\n\n')
f.write(
'################### IntegralSTOR system status report ####################\n\n'
)
f.write(
' IntegralSTOR version : %s \n\n'
% ver)
f.write(
' Hostname : %s \n\n'
% socket.getfqdn())
f.write(
' Report generated at : %s \n\n'
% date_str)
if org_info:
if 'org_name' in org_info:
f.write(
' Organization name : %s \n\n'
% org_info['org_name'])
if 'unit_name' in org_info:
f.write(
' Unit name : %s \n\n'
% org_info['unit_name'])
if 'unit_id' in org_info:
f.write(
' Unit ID : %s \n\n'
% org_info['unit_id'])
if 'subunit_name' in org_info:
f.write(
' Subunit name : %s \n\n'
% org_info['subunit_name'])
if 'subunit_id' in org_info:
f.write(
' Subunit ID : %s \n\n'
% org_info['subunit_id'])
if uuid:
f.write(
' Installation ID : %s \n\n'
% uuid['uuid_str'])
f.write(
'##########################################################################\n\n'
)
f.write('\n\n')
except Exception, e:
return False, 'Error generating global header : %s' % str(e)
def main():
lg = None
try:
scripts_log, err = config.get_scripts_log_path()
if err:
raise Exception(err)
lg, err = logger.get_script_logger('System status report generation',
scripts_log,
level=logging.DEBUG)
status_reports_dir, err = config.get_staus_reports_dir_path()
if err:
raise Exception(err)
lck, err = lock.get_lock('generate_system_status_report')
if err:
raise Exception(err)
if not lck:
raise Exception('Could not acquire lock.')
logger.log_or_print('System status report generation initiated.',
lg,
level='info')
if len(sys.argv) != 2:
raise Exception(
'Usage : python generate_system_status_report.py <past_x_days>'
)
past_x_days = int(sys.argv[1])
start_time, err = datetime_utils.get_epoch(
when='midnight', num_previous_days=past_x_days)
if err:
raise Exception(err)
now, err = datetime_utils.get_epoch(when='now')
if err:
raise Exception(err)
now_local_str, err = datetime_utils.convert_from_epoch(
now, return_format='str', str_format='%Y_%m_%d_%H_%M', to='local')
if err:
raise Exception(err)
tmp_file_name = 'integralstor_status_%s' % now_local_str
tmp_file_name_with_path = '/tmp/%s' % tmp_file_name
with open(tmp_file_name_with_path, 'w') as f:
ret, err = generate_global_header(f)
# print ret, err
f.write('\n')
ret, err = generate_dmidecode_section(f)
# print ret, err
f.write('\n\n')
ret, err = generate_cpu_section(f)
# print ret, err
f.write('\n\n')
ret, err = generate_memory_section(f)
# print ret, err
f.write('\n\n')
ret, err = generate_command_based_section(
f, 'nmcli con', 'Networking connections')
# print ret, err
f.write('\n\n')
ret, err = generate_command_based_section(f, 'ip addr',
'IP addresses')
# print ret, err
f.write('\n\n')
hw_platform, err = config.get_hardware_platform()
# print ret, err
if hw_platform:
if hw_platform == 'dell':
ret, err = generate_dell_hw_status(f)
# print ret, err
f.write('\n\n')
ret, err = generate_disks_status_section(f)
# print ret, err
f.write('\n\n')
ret, err = generate_command_based_section(
f,
'df -HT --exclude-type=devtmpfs --exclude-type=tmpfs --exclude-type=zfs',
'OS disk space usage')
# print ret, err
f.write('\n\n')
ret, err = generate_zfs_info_section(f)
# print ret, err
f.write('\n\n')
ret, err = generate_command_based_section(f, 'zpool list',
'ZFS pool space usage')
# print ret, err
f.write('\n\n')
ret, err = generate_command_based_section(
f,
'zfs list -t filesystem -o name,used,avail,refer,mountpoint,dedup,compression,quota,xattr,recordsize,acltype',
'ZFS datasets')
# print ret, err
f.write('\n\n')
ret, err = generate_command_based_section(
f,
'zfs list -t volume -o name,used,avail,refer,mountpoint,dedup,compression,volsize,volblocksize',
'ZFS zvols')
# print ret, err
f.write('\n\n')
ret, err = generate_command_based_section(f, 'zpool status -v',
'ZFS pool status')
# print ret, err
f.write('\n\n')
ret, err = generate_audits_section(f, start_time, past_x_days)
# print ret, err
f.write('\n\n')
ret, err = generate_alerts_section(f, start_time, past_x_days)
# print ret, err
f.write('\n\n')
try:
os.makedirs(status_reports_dir)
except:
pass
final_file_name_with_path = '%s/%s' % (status_reports_dir,
tmp_file_name)
shutil.move(tmp_file_name_with_path, final_file_name_with_path)
d, err = mail.load_email_settings()
if not err and d and 'support_email_addresses' in d and d[
'support_email_addresses']:
# Email settings present so send it out to the support email
# address
email_header = '%s - IntegralSTOR system status report' % socket.getfqdn(
)
email_body = 'Please find the latest IntegralSTOR system status report'
processed_successfully, err = mail.enqueue(
d['support_email_addresses'],
email_header,
email_body,
attachment_file_location=final_file_name_with_path,
delete_attachment_file=False)
if err:
raise Exception(err)
except Exception, e:
# print str(e)
lock.release_lock('generate_system_status_report')
logger.log_or_print('Error generating system status report : %s' % e,
lg,
level='critical')
return -1, 'Error generating system status report : %s' % e
def export_old_audits(min_to_export=1000, export_count=500):
"""Export the oldest export_count audits if the total number of audits exceeds min_to_export
"""
try:
# print min_to_export, export_count
db_path, err = config.get_db_path()
if err:
raise Exception(err)
ret, err = db.get_single_row(
db_path, 'select count(*) as count from audit')
if err:
raise Exception(err)
if ret['count'] > int(min_to_export):
query = "select * from audit order by audit_id limit %d;" % int(
export_count)
full_audit_list, err = db.get_multiple_rows(db_path, query)
if err:
raise Exception(err)
# print full_audit_list
audit_list = []
for a in full_audit_list:
# If it is still in the holding table then skip it so it can
# get processed..
query = "select * from event_notifications_holding where event_id=%d;" % int(
a['audit_id'])
ret, err = db.get_single_row(db_path, query)
# print ret, err
if err:
raise Exception(err)
if not ret:
audit_list.append(a)
# print audit_list
if audit_list:
delete_commands = []
export_dir_name, err = config.get_exported_logs_dir_path()
if not os.path.exists(export_dir_name):
os.makedirs(export_dir_name)
now, err = datetime_utils.get_epoch(
when='now', num_previous_days=0)
if err:
raise Exception(err)
now_str, err = datetime_utils.convert_from_epoch(
now, return_format='str', str_format='%Y_%m_%d_%H_%M', to='local')
export_filename = 'audits_%s' % now_str
with open('%s/%s' % (export_dir_name, export_filename), 'w') as f:
f.write(
'Audit time(UTC) | Audit type | Performed by | Performed from | Audit message\n')
f.write(
'-------------------------------------------------------------------------------------------\n')
for entry in audit_list:
# print entry
aud, err = _parse_audit_entry(entry)
if err:
raise Exception(err)
# print aud, err
f.write('%s | %s | %s | %s | %s\n\n' % (
aud['time'], aud['action'], aud['username'], aud['ip'], aud['action_str']))
delete_commands.append(
['delete from audit where audit_id="%d"' % int(aud['audit_id'])])
# print delete_commands
ret, err = db.execute_iud(db_path, delete_commands)
if err:
raise Exception(err)
except Exception, e:
return False, 'Error exporting old audits : %s' % str(e)
def view_system_info(request):
return_dict = {}
try:
if "ack" in request.GET:
if request.GET["ack"] == "system_time_set":
return_dict['ack_message'] = "Time successfully updated"
elif request.GET["ack"] == "system_date_set":
return_dict['ack_message'] = "Date successfully updated"
elif request.GET["ack"] == "system_datetime_set":
return_dict[
'ack_message'] = "Date and time successfully updated"
elif request.GET["ack"] == 'system_timezone_set':
return_dict['ack_message'] = "Timezone successfully updated"
elif request.GET['ack'] == 'system_date_timezone_set':
return_dict[
'ack_message'] = 'Date and timezone successfully updated'
elif request.GET['ack'] == 'system_time_timezone_set':
return_dict[
'ack_message'] = 'Time and timezone successfully updated'
elif request.GET['ack'] == 'system_datetimetz_set':
return_dict[
'ack_message'] = 'Date, time and timezone successfully updated'
elif request.GET['ack'] == 'config_uploaded':
return_dict[
'ack_message'] = 'Configuration information successfully uploaded'
elif request.GET['ack'] == 'update_org_info_ok':
return_dict[
'ack_message'] = 'Updated orgnazation information successfully'
si, err = system_info.load_system_config()
if err:
raise Exception(err)
org_info, err = system_info.get_org_info()
if err:
raise Exception(err)
return_dict['org_info'] = org_info
now_epoch, err = datetime_utils.get_epoch(when='now',
num_previous_days=0)
if err:
raise Exception(err)
now, err = datetime_utils.convert_from_epoch(now_epoch,
return_format='datetime',
to='local')
if err:
raise Exception(err)
milliseconds = int(now_epoch * 1000)
if err:
raise Exception(err)
system_timezone, err = datetime_utils.get_system_timezone()
if err:
raise Exception(err)
return_dict['date_str'] = now.strftime("%A %d %B %Y")
return_dict['time'] = now
return_dict['milliseconds'] = milliseconds
return_dict['system_timezone'] = system_timezone['system_timezone']
# print return_dict['system_timezone']
return_dict['system_info'] = si
if "from" in request.GET:
frm = request.GET["from"]
return_dict['frm'] = frm
return_dict['node'] = si
return django.shortcuts.render_to_response(
"view_system_info.html",
return_dict,
context_instance=django.template.context.RequestContext(request))
except Exception, e:
return_dict['base_template'] = "system_base.html"
return_dict["page_title"] = 'System configuration'
return_dict['tab'] = 'node_info_tab'
return_dict["error"] = 'Error loading system configuration'
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 run_task(task_id):
try:
task, err = get_task(task_id)
if err:
raise Exception(err)
now, err = datetime_utils.get_epoch(when='now')
if err:
raise Exception(err)
db_path, err = config.get_db_path()
if err:
raise Exception(err)
if task['last_run_time']:
seconds_since_last_run = (now - task['last_run_time'])
# retry_interval is in minutes!
if seconds_since_last_run < task['retry_interval'] * 60:
raise Exception("Too young to attempt")
# Mark the task as running
cmd = "update tasks set status = 'running', last_run_time=%d where task_id = '%d'" % (
now, task['task_id'])
status, err = db.execute_iud(db_path, [
[cmd],
], get_rowid=True)
if err:
raise Exception(err)
audit_str = "%s" % task['description']
audit.audit("task_start", audit_str, None, system_initiated=True)
attempts = task['attempts']
run_as_user_name = task['run_as_user_name']
# Now process subtasks for the task
subtasks_query = "select * from subtasks where task_id == '%d' and (status == 'error-retrying' or status == 'queued') order by subtask_id" % task[
'task_id']
subtasks, err = db.get_multiple_rows(db_path, subtasks_query)
if err:
raise Exception(err)
# Assume task is complete unless proven otherwise
task_completed = True
# Iteriate through all the unfinished subtasks related to the
# task
for subtask in subtasks:
subtask_id = subtask["subtask_id"]
status_update = "update subtasks set status = 'running' where subtask_id = '%d' and status is not 'cancelled'" % subtask_id
status, err = db.execute_iud(db_path, [
[status_update],
],
get_rowid=True)
if err:
task_completed = False
break
# Now actually execute the command
# This task is not meant to be executed by the current user
# so switch to that user
(out, return_code), err = command.execute_with_rc(
subtask["command"],
shell=True,
run_as_user_name=run_as_user_name)
if out[0]:
output = re.sub("'", "", ''.join(out[0]))
else:
output = None
if out[1]:
error = re.sub("'", "", ''.join(out[1]))
else:
error = None
if return_code == 0:
# This means the command was successful. So update to
# completed
status_update = "update subtasks set status = 'completed', return_code='%d' where subtask_id = '%d' and status is not 'cancelled';" % (
return_code, subtask_id)
status, err = db.execute_iud(db_path, [
[status_update],
],
get_rowid=True)
if err:
task_completed = False
break
else:
continue
else:
# Subtask command failed
if attempts > 1 or attempts == -2:
status_update = 'update subtasks set status = "error-retrying", return_code="%d" where subtask_id = "%d" and status is not "cancelled";' % (
return_code, subtask_id)
elif attempts in [0, 1]:
status_update = 'update subtasks set status = "failed", return_code="%d" where subtask_id = "%d" and status is not "cancelled";' % (
return_code, subtask_id)
execute, err = db.execute_iud(db_path, [
[status_update],
],
get_rowid=True)
task_completed = False
break
# Capture end time
end_time, err = datetime_utils.get_epoch(when='now')
if err:
raise Exception(err)
# Update status
if task_completed:
status_update = "update tasks set status = 'completed' where task_id = '%d'" % task[
'task_id']
else:
if attempts > 1:
status_update = "update tasks set status = 'error-retrying', attempts = %d where task_id = '%d' and status is not 'cancelled'" % (
attempts - 1, task['task_id'])
elif attempts == -2:
status_update = "update tasks set status = 'error-retrying', attempts = %d where task_id = '%d' and status is not 'cancelled'" % (
-2, task['task_id'])
else:
status_update = "update tasks set status = 'failed', attempts = '%d' where task_id = '%d' and status is not 'cancelled'" % (
0, task['task_id'])
status, err = db.execute_iud(db_path, [
[status_update],
],
get_rowid=True)
if err:
raise Exception(err)
# Update task's end time
end_time_update = "update tasks set end_time=%d where task_id = '%d'" % (
end_time, task['task_id'])
ret, err = db.execute_iud(db_path, [
[end_time_update],
],
get_rowid=True)
if err:
pass
# raise Exception(err)
if task_completed:
audit.audit("task_complete",
audit_str,
None,
system_initiated=True)
else:
audit.audit("task_fail", audit_str, None, system_initiated=True)
except Exception as e:
return False, 'Error processing task: %s' % e
else:
return True, None
def export_old_audits(min_to_export=1000, export_count=500):
"""Export the oldest export_count audits if the total number of audits exceeds min_to_export
"""
try:
# print min_to_export, export_count
db_path, err = config.get_db_path()
if err:
raise Exception(err)
ret, err = db.get_single_row(
db_path, 'select count(*) as count from audit')
if err:
raise Exception(err)
if ret['count'] > int(min_to_export):
query = "select * from audit order by audit_id limit %d;" % int(
export_count)
full_audit_list, err = db.get_multiple_rows(db_path, query)
if err:
raise Exception(err)
# print full_audit_list
audit_list = []
for a in full_audit_list:
# If it is still in the holding table then skip it so it can
# get processed..
query = "select * from event_notifications_holding where event_id=%d;" % int(
a['audit_id'])
ret, err = db.get_single_row(db_path, query)
# print ret, err
if err:
raise Exception(err)
if not ret:
audit_list.append(a)
# print audit_list
if audit_list:
delete_commands = []
export_dir_name, err = config.get_exported_logs_dir_path()
if not os.path.exists(export_dir_name):
os.makedirs(export_dir_name)
now, err = datetime_utils.get_epoch(
when='now', num_previous_days=0)
if err:
raise Exception(err)
now_str, err = datetime_utils.convert_from_epoch(
now, return_format='str', str_format='%Y_%m_%d_%H_%M', to='local')
export_filename = 'audits_%s' % now_str
with open('%s/%s' % (export_dir_name, export_filename), 'w') as f:
f.write(
'Audit time(UTC) | Audit type | Performed by | Performed from | Audit message\n')
f.write(
'-------------------------------------------------------------------------------------------\n')
for entry in audit_list:
# print entry
aud, err = _parse_audit_entry(entry)
if err:
raise Exception(err)
# print aud, err
f.write('%s | %s | %s | %s | %s\n\n' % (
aud['time'], aud['action'], aud['username'], aud['ip'], aud['action_str']))
delete_commands.append(
['delete from audit where audit_id="%d"' % int(aud['audit_id'])])
# print delete_commands
ret, err = db.execute_iud(db_path, delete_commands)
if err:
raise Exception(err)
except Exception, e:
return False, 'Error exporting old audits : %s' % str(e)
def export_old_alerts(older_than_days=1):
"""Move all alerts older than the older_than_days into a file in
/var/log/integralstor/logs/exported dir
"""
try:
cutoff_seconds, err = datetime_utils.get_epoch(
when='now', num_previous_days=older_than_days)
if err:
raise Exception(err)
#query = "select * from alerts where last_update_time < Datetime('now', '-%d days') order by alert_id;"%older_than_days
query = "select * from alerts where last_update_time < %d order by alert_id;" % cutoff_seconds
full_alerts_list, err = _get_and_parse_alerts(query)
if err:
raise Exception(err)
alerts_list = []
# print 'full', full_alerts_list
db_path, err = config.get_db_path()
if err:
raise Exception(err)
for a in full_alerts_list:
# If it is still in the holding table then skip it so it can get
# processed..
query = "select * from event_notifications_holding where event_id=%d;" % int(
a['alert_id'])
ret, err = db.get_single_row(db_path, query)
# print ret, err
if err:
raise Exception(err)
if not ret:
alerts_list.append(a)
# print 'to export', alerts_list
if alerts_list:
delete_commands = []
export_dir_name, err = config.get_exported_logs_dir_path()
if err:
raise Exception(err)
if not os.path.exists(export_dir_name):
os.makedirs(export_dir_name)
now, err = datetime_utils.get_epoch(when='now',
num_previous_days=0)
if err:
raise Exception(err)
now_str, err = datetime_utils.convert_from_epoch(
now,
return_format='str',
str_format='%Y_%m_%d_%H_%M',
to='local')
export_filename = 'alerts_%s' % now_str
# print export_filename
with open('%s/%s' % (export_dir_name, export_filename), 'w') as f:
f.write(
'First alert time(UTC) | Last update time(UTC) | Repeat count | Subsystem | Severity | Alert message\n'
)
f.write(
'-------------------------------------------------------------------------------------------\n'
)
for al in alerts_list:
f.write('%s | %s | %d | %s | %s | %s\n\n' %
(al['first_alert_time'], al['last_update_time'],
al['repeat_count'], al['subsystem'],
al['severity'], al['alert_str']))
delete_commands.append([
'delete from alerts where alert_id="%d"' %
int(al['alert_id'])
])
# print delete_commands
db_path, err = config.get_db_path()
if err:
raise Exception(err)
ret, err = db.execute_iud(db_path, delete_commands)
if err:
raise Exception(err)
except Exception, e:
return False, 'Error exporting old alerts : %s' % str(e)
def view_system_info(request):
return_dict = {}
try:
if "ack" in request.GET:
if request.GET["ack"] == "system_time_set":
return_dict['ack_message'] = "Time successfully updated"
elif request.GET["ack"] == "system_date_set":
return_dict['ack_message'] = "Date successfully updated"
elif request.GET["ack"] == "system_datetime_set":
return_dict['ack_message'] = "Date and time successfully updated"
elif request.GET["ack"] == 'system_timezone_set':
return_dict['ack_message'] = "Timezone successfully updated"
elif request.GET['ack'] == 'system_date_timezone_set':
return_dict['ack_message'] = 'Date and timezone successfully updated'
elif request.GET['ack'] == 'system_time_timezone_set':
return_dict['ack_message'] = 'Time and timezone successfully updated'
elif request.GET['ack'] == 'system_datetimetz_set':
return_dict['ack_message'] = 'Date, time and timezone successfully updated'
elif request.GET['ack'] == 'config_uploaded':
return_dict['ack_message'] = 'Configuration information successfully uploaded'
elif request.GET['ack'] == 'update_org_info_ok':
return_dict['ack_message'] = 'Updated orgnazation information successfully'
si, err = system_info.load_system_config()
if err:
raise Exception(err)
org_info, err = system_info.get_org_info()
if err:
raise Exception(err)
return_dict['org_info'] = org_info
now_epoch, err = datetime_utils.get_epoch(
when='now', num_previous_days=0)
if err:
raise Exception(err)
now, err = datetime_utils.convert_from_epoch(
now_epoch, return_format='datetime', to='local')
if err:
raise Exception(err)
milliseconds = int(now_epoch * 1000)
if err:
raise Exception(err)
system_timezone, err = datetime_utils.get_system_timezone()
if err:
raise Exception(err)
return_dict['date_str'] = now.strftime("%A %d %B %Y")
return_dict['time'] = now
return_dict['milliseconds'] = milliseconds
return_dict['system_timezone'] = system_timezone['system_timezone']
# print return_dict['system_timezone']
return_dict['system_info'] = si
if "from" in request.GET:
frm = request.GET["from"]
return_dict['frm'] = frm
return_dict['node'] = si
return django.shortcuts.render_to_response("view_system_info.html", return_dict, context_instance=django.template.context.RequestContext(request))
except Exception, e:
return_dict['base_template'] = "system_base.html"
return_dict["page_title"] = 'System configuration'
return_dict['tab'] = 'node_info_tab'
return_dict["error"] = 'Error loading system configuration'
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 export_old_alerts(older_than_days=1):
"""Move all alerts older than the older_than_days into a file in
/var/log/integralstor/logs/exported dir
"""
try:
cutoff_seconds, err = datetime_utils.get_epoch(
when='now', num_previous_days=older_than_days)
if err:
raise Exception(err)
#query = "select * from alerts where last_update_time < Datetime('now', '-%d days') order by alert_id;"%older_than_days
query = "select * from alerts where last_update_time < %d order by alert_id;" % cutoff_seconds
full_alerts_list, err = _get_and_parse_alerts(query)
if err:
raise Exception(err)
alerts_list = []
# print 'full', full_alerts_list
db_path, err = config.get_db_path()
if err:
raise Exception(err)
for a in full_alerts_list:
# If it is still in the holding table then skip it so it can get
# processed..
query = "select * from event_notifications_holding where event_id=%d;" % int(
a['alert_id'])
ret, err = db.get_single_row(db_path, query)
# print ret, err
if err:
raise Exception(err)
if not ret:
alerts_list.append(a)
# print 'to export', alerts_list
if alerts_list:
delete_commands = []
export_dir_name, err = config.get_exported_logs_dir_path()
if err:
raise Exception(err)
if not os.path.exists(export_dir_name):
os.makedirs(export_dir_name)
now, err = datetime_utils.get_epoch(
when='now', num_previous_days=0)
if err:
raise Exception(err)
now_str, err = datetime_utils.convert_from_epoch(
now, return_format='str', str_format='%Y_%m_%d_%H_%M', to='local')
export_filename = 'alerts_%s' % now_str
# print export_filename
with open('%s/%s' % (export_dir_name, export_filename), 'w') as f:
f.write(
'First alert time(UTC) | Last update time(UTC) | Repeat count | Subsystem | Severity | Alert message\n')
f.write(
'-------------------------------------------------------------------------------------------\n')
for al in alerts_list:
f.write('%s | %s | %d | %s | %s | %s\n\n' % (
al['first_alert_time'], al['last_update_time'], al['repeat_count'], al['subsystem'], al['severity'], al['alert_str']))
delete_commands.append(
['delete from alerts where alert_id="%d"' % int(al['alert_id'])])
# print delete_commands
db_path, err = config.get_db_path()
if err:
raise Exception(err)
ret, err = db.execute_iud(db_path, delete_commands)
if err:
raise Exception(err)
except Exception, e:
return False, 'Error exporting old alerts : %s' % str(e)
def record_alerts(alerts_list):
"""Record a set of passed list containing alert dictionaries into the DB."""
try:
db_path, err = config.get_db_path()
if err:
raise Exception(err)
if alerts_list:
now, err = datetime_utils.get_epoch(when='now',
num_previous_days=0)
if err:
raise Exception(er)
for alert in alerts_list:
# print alert
command_list = []
result, err = _get_repeat_entry(alert)
# print 'repeat', result, err
if err:
raise Exception(err)
update_alert_id = None
if result:
cmd = [
'update alerts set repeat_count=%d, last_update_time="%d" where alert_id=%d'
% (int(result['repeat_count']) + 1, now,
result['alert_id'])
]
update_alert_id = result['alert_id']
# print 'updating!', cmd
else:
if 'component' not in alert or (not alert['component']):
cmd = [
'insert into alerts(first_alert_time, last_update_time, severity_type_id, subsystem_type_id, alert_str) values (?,?,?,?,?)',
(
now,
now,
alert['severity_type_id'],
alert['subsystem_type_id'],
alert['alert_str'],
)
]
else:
cmd = [
'insert into alerts(first_alert_time, last_update_time, severity_type_id, subsystem_type_id, component, alert_str) values (?,?,?,?,?,?)',
(
now,
now,
alert['severity_type_id'],
alert['subsystem_type_id'],
alert['component'],
alert['alert_str'],
)
]
command_list.append(cmd)
# print command_list
rowid, err = db.execute_iud(db_path,
command_list,
get_rowid=True)
if err:
raise Exception(err)
if update_alert_id:
alert_id = update_alert_id
else:
alert_id = rowid
ret, err = event_notifications.record_event_notification_holding(
event_id=alert_id,
event_type_id=1,
subsystem_type_id=alert['subsystem_type_id'],
severity_type_id=alert['severity_type_id'])
# print ret, err
if err:
raise Exception(err)
except Exception, e:
# print str(e)
return False, 'Error recording alerts : %s' % str(e)
def view_dashboard(request, page = None):
return_dict = {}
try:
return_dict["page_title"] = 'Overall system health'
return_dict['tab'] = 'system_health_tab'
return_dict["error"] = 'Error loading system health data'
if request.method != 'GET':
raise Exception('Invalid access method. Please use the menus')
si, err = system_info.load_system_config()
if err:
raise Exception(err)
if not si:
raise Exception('Error loading system configuration')
#node_name = si.keys()[0]
#node = si[node_name]
return_dict['node'] = si
# print node.keys()
# By default show error page
template = "logged_in_error.html"
# Chart specific declarations
# will return 02, instead of 2.
end_epoch, err = datetime_utils.get_epoch(when='now')
if err:
raise Exception(err)
start_epoch = end_epoch - 3 * 60 * 60
start, err = datetime_utils.convert_from_epoch(
start_epoch, return_format='str', str_format='%H:%M:%S', to='local')
if err:
raise Exception(err)
end, err = datetime_utils.convert_from_epoch(
end_epoch, return_format='str', str_format='%H:%M:%S', to='local')
if err:
raise Exception(err)
todays_date = (datetime.date.today()).strftime('%02d')
value_list = []
time_list = []
num_bad_disks = 0
num_hw_raid_bad_disks = 0
num_hw_raid_ctrl_disks = 0
num_smart_ctrl_disks = 0
num_disks = len(si['disks'])
disks_ok = True
for sn, disk in si['disks'].items():
if 'status' in disk:
if 'hw_raid' in disk:
if not disk['hw_raid']:
num_smart_ctrl_disks += 1
if (disk['status'] is not None and disk['status'].upper() not in ['PASSED', 'OK']):
num_bad_disks += 1
disks_ok = False
else:
num_hw_raid_ctrl_disks += 1
if (disk['status'] is not None and disk['status'].upper() != 'OK'):
num_hw_raid_bad_disks += 1
disks_ok = False
else:
# Assume its a non raid disk
num_smart_ctrl_disks += 1
if (disk['status'] is not None and disk['status'].upper() not in ['PASSED', 'OK']):
num_bad_disks += 1
disks_ok = False
return_dict['num_disks'] = num_disks
return_dict['num_bad_disks'] = num_bad_disks
return_dict['disks_ok'] = disks_ok
return_dict['num_hw_raid_bad_disks'] = num_hw_raid_bad_disks
return_dict['num_hw_raid_ctrl_disks'] = num_hw_raid_ctrl_disks
return_dict['num_smart_ctrl_disks'] = num_smart_ctrl_disks
if 'ipmi_status' in si:
num_sensors = len(si['ipmi_status'])
num_bad_sensors = 0
ipmi_ok = True
for sensor in si['ipmi_status']:
if sensor['status'] in ['ok', 'nr', 'na']:
continue
else:
num_bad_sensors += 1
ipmi_ok = False
return_dict['num_sensors'] = num_sensors
return_dict['num_bad_sensors'] = num_bad_sensors
return_dict['ipmi_ok'] = ipmi_ok
services_dict, err = services_management.get_sysd_services_status()
if err:
raise Exception(err)
num_services = len(services_dict)
num_failed_services = 0
num_active_services = 0
num_inactive_services = 0
services_ok = True
if services_dict:
for service, service_d in services_dict.items():
if service_d["info"]["status"]["status_str"] == "Active":
num_active_services += 1
elif service_d["info"]["status"]["status_str"] == "Inactive":
num_inactive_services += 1
elif service_d["info"]["status"]["status_str"] == "Failed":
num_failed_services += 1
services_ok = False
elif service_d["info"]["status"]["status_str"] == "Unknown State":
num_failed_services += 1
services_ok = False
return_dict['num_services'] = num_services
return_dict['num_active_services'] = num_active_services
return_dict['num_inactive_services'] = num_inactive_services
return_dict['num_failed_services'] = num_failed_services
return_dict['services_ok'] = services_ok
else:
raise Exception('Error retrieving services status')
pools, err = zfs.get_pools()
if err:
raise Exception(err)
num_pools = len(pools)
num_bad_pools = 0
num_degraded_pools = 0
num_high_usage_pools = 0
for pool in pools:
if pool['usage']['used_percent'] > 75:
num_high_usage_pools += 1
if pool['config']['pool']['root']['status']['state'] == 'ONLINE':
pass
elif pool['config']['pool']['root']['status']['state'] == 'DEGRADED':
num_degraded_pools += 1
else:
num_bad_pools += 1
return_dict['num_pools'] = num_pools
return_dict['num_bad_pools'] = num_bad_pools
return_dict['num_degraded_pools'] = num_degraded_pools
return_dict['num_high_usage_pools'] = num_high_usage_pools
load_avg_ok = True
if (si["load_avg"]["5_min"] > si["load_avg"]["cpu_cores"]) or (si["load_avg"]["15_min"] > si["load_avg"]["cpu_cores"]):
load_avg_ok = False
return_dict['load_avg_ok'] = load_avg_ok
shares_list, err = cifs.get_shares_list()
if err:
raise Exception(err)
return_dict['num_cifs_shares'] = len(shares_list)
exports_list, err = nfs.load_exports_list()
if err:
raise Exception(err)
return_dict['num_nfs_exports'] = len(exports_list)
target_list, err = iscsi_stgt.get_targets()
if err:
raise Exception(err)
return_dict['num_iscsi_targets'] = len(target_list)
with open('/proc/uptime', 'r') as f:
uptime_seconds = float(f.readline().split()[0])
uptime_str = '%s hours' % (
':'.join(str(datetime.timedelta(seconds=uptime_seconds)).split(':')[:2]))
return_dict['uptime_str'] = uptime_str
# CPU status
if not page:
page = "sys_health"
if page == "cpu":
return_dict["page_title"] = 'CPU statistics'
return_dict['tab'] = 'cpu_tab'
return_dict["error"] = 'Error loading CPU statistics'
cpu, err = stats.get_system_stats(todays_date, start, end, "cpu")
if err:
raise Exception(err)
value_dict = {}
if cpu:
for key in cpu.keys():
value_list = []
time_list = []
if key == "date":
pass
else:
if cpu[key]:
for a in cpu[key]:
time_list.append(a[0])
value_list.append(a[1])
value_dict[key] = value_list
return_dict["data_dict"] = value_dict
queue, err = stats.get_system_stats(
todays_date, start, end, "queue")
if err:
raise Exception(err)
value_dict = {}
if queue:
for key in queue.keys():
value_list = []
time_list = []
if key == "date":
pass
else:
for a in queue[key]:
time_list.append(a[0])
value_list.append(a[1])
value_dict[key] = value_list
return_dict["data_dict_queue"] = value_dict
return_dict['node'] = si
d = {}
template = "view_cpu_stats.html"
elif page == "sys_health":
return_dict["page_title"] = 'Overall system health'
return_dict['tab'] = 'system_health_tab'
return_dict["error"] = 'Error loading system health data'
template = "view_dashboard.html"
hw_platform, err = config.get_hardware_platform()
if hw_platform:
return_dict['hw_platform'] = hw_platform
if hw_platform == 'dell':
from integralstor.platforms import dell
idrac_url, err = dell.get_idrac_addr()
if idrac_url:
return_dict['idrac_url'] = idrac_url
# Memory
elif page == "memory":
return_dict["page_title"] = 'Memory statistics'
return_dict['tab'] = 'memory_tab'
return_dict["error"] = 'Error loading memory statistics'
mem, err = stats.get_system_stats(
todays_date, start, end, "memory")
if err:
raise Exception(err)
if mem:
for a in mem["memused"]:
time_list.append(a[0])
value_list.append((a[1] / (1024 * 1024)))
return_dict['memory_status'] = si['memory']
template = "view_memory_stats.html"
# Network
elif page == "network":
return_dict["page_title"] = 'Network statistics'
return_dict['tab'] = 'network_tab'
return_dict["error"] = 'Error loading Network statistics'
network, err = stats.get_system_stats(
todays_date, start, end, "network")
if err:
raise Exception(err)
value_dict = {}
if network:
for key in network.keys():
value_list = []
time_list = []
if key == "date" or key == "lo":
pass
else:
for a in network[key]["ifutil-percent"]:
time_list.append(a[0])
value_list.append(a[1])
value_dict[key] = value_list
return_dict["data_dict"] = value_dict
return_dict["network_status"] = si['interfaces']
template = "view_network_stats.html"
return_dict["labels"] = time_list
return_dict["data"] = value_list
return django.shortcuts.render_to_response(template, return_dict, context_instance=django.template.context.RequestContext(request))
except Exception, e:
return_dict['base_template'] = "monitoring_base.html"
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 download_log(request):
""" Download the system log of the type specified in log_type POST param
This calls the /sys_log via an http request on that node to get the info"""
return_dict = {}
try:
hw_platform, err = config.get_hardware_platform()
if err:
raise Exception(err)
if hw_platform and hw_platform != 'dell':
raise Exception('Unknown hardware platform')
return_dict['hw_platform'] = hw_platform
form = log_management_forms.DownloadLogsForm(request.POST or None)
if request.method == 'POST':
if form.is_valid():
cd = form.cleaned_data
log_type = cd['log_type']
if log_type in ['alerts', 'audit', 'hardware']:
response = django.http.HttpResponse()
if log_type == 'alerts':
response['Content-disposition'] = 'attachment; filename=alerts_log.txt'
all_alerts, err = alerts.get_alerts()
if err:
raise Exception(err)
for alert in all_alerts:
if int(alert['repeat_count']) > 1:
response.write('Last alert time %s\nAlert message: %s\nRepeated count: %d\n\n' %
(alert['last_update_time'], alert['alert_str'], int(alert['repeat_count'])))
else:
response.write('Last alert time %s\nAlert message: %s\n\n' %
(alert['last_update_time'], alert['alert_str']))
response.flush()
elif log_type == 'audit':
response['Content-disposition'] = 'attachment; filename=audit_log.txt'
all_audits, err = audit.get_entries()
if err:
raise Exception(err)
for audit_info in all_audits:
response.write('Time : %s \n' % audit_info['time'])
response.write('Source IP : %s \n' %
audit_info['ip'])
response.write('Action : %s \n' %
audit_info['action_str'])
response.write('\n')
response.flush()
elif log_type == 'hardware':
response['Content-disposition'] = 'attachment; filename=hardware_logs.txt'
hw_platform, err = config.get_hardware_platform()
if not hw_platform or hw_platform != 'dell':
raise Exception('Unknown hardware platform')
if hw_platform == 'dell':
from integralstor.platforms import dell
logs_dict, err = dell.get_alert_logs()
if err:
raise Exception(err)
if not logs_dict:
raise Exception('No logs detected!')
for timestamp, log_list in logs_dict.items():
for log in log_list:
response.write('Time : %s\n' %
log['date_time'])
response.write(
'Severity : %s\n' % log['Severity'])
response.write(
'Description : %s\n' % log['description'])
response.write('\n')
response.flush()
else:
raise Exception('Unknown platform')
else:
scripts_log, err = config.get_scripts_log_path()
if err:
raise Exception(err)
system_logs = [('/var/log/boot.log', 'boot.log'), ('/var/log/dmesg', 'dmesg'), ('/var/log/messages', 'messages'),
('/var/log/smblog.vfs', 'samba'), ('/var/log/samba/log.winbindd', 'winbind'), (scripts_log, 'scripts')]
now_local_epoch, err = datetime_utils.get_epoch(when='now')
if err:
raise Exception(err)
now_local_str, err = datetime_utils.convert_from_epoch(
now_local_epoch, return_format='str', str_format='%Y_%m_%d_%H_%M', to='local')
if err:
raise Exception(err)
zf_name = 'IntegralSTOR_system_logs_%s.zip' % now_local_str
try:
out = io.BytesIO()
zf = zipfile.ZipFile(out, 'w')
for entry in system_logs:
zf.write(entry[0], arcname=entry[1])
#zf.write(file_name, arcname=display_name)
zf.close()
except Exception as e:
raise Exception(
"Error compressing log file : %s" % str(e))
response = django.http.HttpResponse(
out.getvalue(), content_type='application/x-compressed')
response['Content-disposition'] = 'attachment; filename=%s' % (
zf_name)
return response
# either a get or an invalid form so send back form
return_dict['form'] = form
return django.shortcuts.render_to_response('download_log_form.html', return_dict, context_instance=django.template.context.RequestContext(request))
except Exception, e:
return_dict['base_template'] = "system_base.html"
return_dict["page_title"] = 'Download system logs'
return_dict['tab'] = 'logs_tab'
return_dict["error"] = 'Error downloading system logs'
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 run_task(task_id):
try:
task, err = get_task(task_id)
if err:
raise Exception(err)
now, err = datetime_utils.get_epoch(when='now')
if err:
raise Exception(err)
db_path, err = config.get_db_path()
if err:
raise Exception(err)
if task['last_run_time']:
seconds_since_last_run = (now - task['last_run_time'])
# retry_interval is in minutes!
if seconds_since_last_run < task['retry_interval'] * 60:
raise Exception("Too young to attempt")
# Mark the task as running
cmd = "update tasks set status = 'running', last_run_time=%d where task_id = '%d'" % (
now, task['task_id'])
status, err = db.execute_iud(
db_path, [[cmd], ], get_rowid=True)
if err:
raise Exception(err)
audit_str = "%s" % task['description']
audit.audit("task_start", audit_str,
None, system_initiated=True)
attempts = task['attempts']
run_as_user_name = task['run_as_user_name']
# Now process subtasks for the task
subtasks_query = "select * from subtasks where task_id == '%d' and (status == 'error-retrying' or status == 'queued') order by subtask_id" % task[
'task_id']
subtasks, err = db.get_multiple_rows(db_path, subtasks_query)
if err:
raise Exception(err)
# Assume task is complete unless proven otherwise
task_completed = True
# Iteriate through all the unfinished subtasks related to the
# task
for subtask in subtasks:
subtask_id = subtask["subtask_id"]
status_update = "update subtasks set status = 'running' where subtask_id = '%d' and status is not 'cancelled'" % subtask_id
status, err = db.execute_iud(
db_path, [[status_update], ], get_rowid=True)
if err:
task_completed = False
break
# Now actually execute the command
# This task is not meant to be executed by the current user
# so switch to that user
(out, return_code), err = command.execute_with_rc(
subtask["command"], shell=True, run_as_user_name=run_as_user_name)
if out[0]:
output = re.sub("'", "", ''.join(out[0]))
else:
output = None
if out[1]:
error = re.sub("'", "", ''.join(out[1]))
else:
error = None
if return_code == 0:
# This means the command was successful. So update to
# completed
status_update = "update subtasks set status = 'completed', return_code='%d' where subtask_id = '%d' and status is not 'cancelled';" % (
return_code, subtask_id)
status, err = db.execute_iud(
db_path, [[status_update], ], get_rowid=True)
if err:
task_completed = False
break
else:
continue
else:
# Subtask command failed
if attempts > 1 or attempts == -2:
status_update = 'update subtasks set status = "error-retrying", return_code="%d" where subtask_id = "%d" and status is not "cancelled";' % (
return_code, subtask_id)
elif attempts in [0, 1]:
status_update = 'update subtasks set status = "failed", return_code="%d" where subtask_id = "%d" and status is not "cancelled";' % (
return_code, subtask_id)
execute, err = db.execute_iud(
db_path, [[status_update], ], get_rowid=True)
task_completed = False
break
# Capture end time
end_time, err = datetime_utils.get_epoch(when='now')
if err:
raise Exception(err)
# Update status
if task_completed:
status_update = "update tasks set status = 'completed' where task_id = '%d'" % task[
'task_id']
else:
if attempts > 1:
status_update = "update tasks set status = 'error-retrying', attempts = %d where task_id = '%d' and status is not 'cancelled'" % (
attempts - 1, task['task_id'])
elif attempts == -2:
status_update = "update tasks set status = 'error-retrying', attempts = %d where task_id = '%d' and status is not 'cancelled'" % (
-2, task['task_id'])
else:
status_update = "update tasks set status = 'failed', attempts = '%d' where task_id = '%d' and status is not 'cancelled'" % (
0, task['task_id'])
status, err = db.execute_iud(
db_path, [[status_update], ], get_rowid=True)
if err:
raise Exception(err)
# Update task's end time
end_time_update = "update tasks set end_time=%d where task_id = '%d'" % (
end_time, task['task_id'])
ret, err = db.execute_iud(
db_path, [[end_time_update], ], get_rowid=True)
if err:
pass
# raise Exception(err)
if task_completed:
audit.audit("task_complete", audit_str,
None, system_initiated=True)
else:
audit.audit("task_fail", audit_str,
None, system_initiated=True)
except Exception as e:
return False, 'Error processing task: %s' % e
else:
return True, None