1/48
Looks like no tags are added yet.
Gaurdudty
detects malicoius activity in AWS acounts, comropmised accounts, data exfiltration, crpto minig
30 days free trail, instant install
analyze cloudtrail management events, VPC flow logs, route 53 dns query logs
extended threat protection - detect multi stage detects (auto enabled)
compromised instance anomlaous traffic, aws account resources IAM roles/API sus calls
Gaurduty protection plans
extended capabilits
EC2 - scan EBS volumes
EKS - potnetila bad activies
runtime monitory - OS level events runtime threats
lambda - lamda logs
S3 protection - risks in S3 bucket
malware protection S3/backup - scan newly uploaded things
TD/Uaroro - potential access threats
automated response gaurd duty
all gothrough event bridge and then into SQS/SNS or lambda
create rule in eventbridge
Gaurd Duty Trusted and Threat IP Lists
can give CIDR you trust or ones you do not
Suprrsed findings
are not sent to security hub s3 dtecvi or eventbridge
but can be viewed in the archive
Troubleshoot gaurd duty no DNS findings
MUST USE DEFAULT DNS RESOLVE - otherwise no DNS based findings
AWS security hubs
across several accounts and aggregate evetythiings
gaurdudty inspector macie, IAM analyser, systems manger, firewall manager, helath, parntner network osolution
30 day free trail charges per ingetsion
security hub cross region and organiation integration
MUST ENABLE AWS CONFIG TO WORK NO MATTER WHAT ENABLE ON ALL ACCOUNTS
aggregate findings all to one region
manage all accounts in organoziaotn, automatically detect new account
by default theo rg manatent is the security hub adminsterator, but you can delegate a new one
Security hub Standards
continous checks against standard rules, PCI, best practices, etc etc
securtiy hub integration with gaurd duty
findings are sent in AWS security finding format (ASFF)
sent within 5 minutes
archiving in garud duty will not arhcive in security tower
security hub services integration
to deal with those findings you can use audit manager, aws chat bot, amazon detectvie, dtrusted advisor, ssm explorer,ops center
security hub 3rd part integariotn
must accept finding from other aws marketplace
you can send findings and both recieve findings
security hub findings
security hub will auotmatically update and delete fidings
example, IAM users key not rotated, password not strong configuration, etc etc
findings past 90 days are deleted
Security Hub insights
identify what requires intervtion/attention
EC2 instances have lots of poor security
Group by filter and optoinal filter
both built in managed insights, and you can also create your own insights
Security Hub - custom actions
automate security hub with event bridge
create action for response and then send to event bridge as a custom action
security hub detects in member account which elevates to security acount and then security accont will intervene and sotp
Amazon detective
investigate the root cause of security findings, machine leanring, graphs thery, statisical analysis
VPC flow logs, EKS audi tlogs, secuity hub etc etc
1 year analysis,
who did it what ddid they do etc etc
give you a complete image and graph
Amazon Inpsector
automated security assessments on EC2 instance leveraging the SSM agent, for container images on ECR, lmabda functions,
will then report in aws security hub and can send into event bridge
ONLY EC2 CONTAINER IAMGES, AND LAMDBA FUNCTIONS
Logging compliance to secure logs
can put all in S3 and then encrypt, control with IAM and bucket polices and then also MFA
Glacier cost savings
Unified Cloudwatch agent
ollect very speciifc data like RAM, used disk space, process etc etc
congiure in the SSM paratmer store
collect logs to sent to cloudwatch
namespace for metris is CWagent
ensure your IAM perma re right
procstat plugin on cloudwatch agent
procstat gets the system utilization of indivudal processes for both linxu and window, exact amount of CPU memory and etc,
all procstat prefix
iniviidual processes
Troubleshoot issues with the cloudwatch agent
Fail to start - problem with configuraiton file /opt/agent/ogs/configuration-validation.log
cant find metrix - check you use right nameapce and check amazon-cloudwatch-agent.json poliy
not puhsing log events - update to latest version, test connectiving to cloudwatch logs endpint, review account and log group configuration check IAm permissions
CloudWatch Logs
best place to store aplication and log stream logs. within applications ocntainer
ecnrypted by default can send logs ot anything
cloudwatch logs S3 export
exported one time, NOT REAL TIME, batch export take up to 12 hours CreateExport Tascl
cloudwatch Logs Subsecpriotns
real-time stream of log events,s ent to lambda, kinesises
subscpriotn filter- filter which logs are devilerd to your destinaiton
aggregate across all accounts using knesis data stream , use a cross accounts ubstripction
attach deisntaion policy and IAM role cross account
Cloudwatch alarms
OK INSUFFICIENT DATA ALARM
high reolsution and cna set a time range
CloudWatch alarm targest
stop terminate or recover EC2 instance, scale in or scale out, send notification toa SNS service
Cloudwatch alarms want multiple metrics
use composite alarm by monitoring state of other alarms (Combining alarms)
cloudwatch alarmt instance recovery
check the EC2 VM, chcek te hardware, check the EBS volume, and then go recover
recovery si same private public elastic metadat and palacment group ALL THE SAME
Cloudwatch alarm good to know
put on top of logs metrics filter
test by doing aws cloudwatch set-alarm-state (CLI)
Cloudwatch conirbutor insights
find the bad hosts or what is making all of theh noise
Amazon eventbridge (formerly cloudwatch events)
schdule cron jobs or react to a ne vent pattern
intercepts any API call from other services
filter events and then send destioation and do actions etc
Amazon event brdige defaul event bus vs partner event bus
default event bus - services from AWS that sends events into default event bus
partner event bus - zendesk/datadog
custom event bus - custom apps
you can arhvie events or indentlfy, can also replay the archived events so you can retest your uatomations
EventBridge schema registry
analyze evnets and infer the schema, lets you register an application and know how the event data wil be strucutred
Athena
SERVERLESS - 5$ per TB
search S3 bucket
athena performance improvemtn
Columnar data - less scan (Apache Parquet or ORC format) - must use service like Glue to convert your data (ETL)
compress data for smaller retrievels
partition datasets into S3 for easier querying
cloud trail what events
holds both mangemnet events
DOES NOT LOG DATA EVENTS (TO HIGH VOLUME)
CloudTrail Insights
detect unusualy activity, consioutly analyzes write events to detect unusual patterns
cloudtrail default retntion
default is 90 days
store past them you need to use S3 and Athena
Cloud trail Lake
centralize all data store and use SQL
immutable storge for 10 years
query results validation that results have not been validated
can make dashboards and have AWS Curated dashboards
Cloud trail log file integrity validation
use SHA-256 and RSA, refence log files for last hour and contains a hash
is cloud trail real time?
NO takes 15 minutes
How to monitor accoutn activity
AWS Config, cloud trail event history, cloudwatch log insihgts and athena quries
Macie data Identifiers
Managed data identifier - pre built in SSN/ credit card
custom data identifier - your own custom sesntive data (employee IDs)
Macie fidnings
sesntive data discovery - log results
suppression rules - attribute base dfilter to arhcive finidngs auotmatically
AWS macie multi-account strategy
associate member account with admin account, administrator can access to all data and add/remove
VPC flow logs
subnet flow logs ENI logs, go into S3, cloudwatch logs,
ELB RDS< leasticache, redshift, NATGW etc
vpc flow logs syntax
version - account id - interace id - src addr - dest addr- src port - dst port - packets -start -action - log status
DNS resolution logs
Cloudwatch logs only
only for public hosted zones