Using apache kafka in SSL mode - ssl

I'm trying to set up kafka in SSL [1-way] mode. I've gone through the official documentation and successfully generated the certificates. I'll note down the behavior for 2 different cases. This setup has only one broker and one zookeeper.
Case-1: Inter-broker communication - Plaintext
Relevant entries in my server.properties file are as follows:
listeners=PLAINTEXT://localhost:9092, SSL://localhost:9093
ssl.keystore.location=/Users/xyz/home/ssl/server.keystore.jks
ssl.keystore.password=****
ssl.key.password=****
I've added a client-ssl.properties in kafka config dir with following entries:
security.protocol=SSL
ssl.truststore.location=/Users/xyz/home/ssl/client.truststore.jks
ssl.truststore.password=****
If I put bootstrap.servers=localhost:9093 or bootstrap.servers=localhost:9092 in my config/producer.properties file, my console-producers/consumers work fine. Is that the intended behavior? If yes, then why? Because I'm specifically trying to connect to localhost:9093 from producer/consumer in SSL mode.
Case-2: Inter-broker communication - SSL
Relevant entries in my server.properties file are as follows:
security.inter.broker.protocol=SSL
listeners=SSL://localhost:9093
ssl.keystore.location=/Users/xyz/home/ssl/server.keystore.jks
ssl.keystore.password=****
ssl.key.password=****
My client-ssl.properties file remains the same. I put bootstrap.servers=localhost:9093 in producer.properties file. Now, none of my producer/consumer can connect to kafka. I get the following msg:
WARN Error while fetching metadata with correlation id 0 : {test=LEADER_NOT_AVAILABLE} (org.apache.kafka.clients.NetworkClient)
What am I doing wrong?
In all these cases I'm using the following commands to start producers/consumers:
./kafka-console-producer.sh --broker-list localhost:9093 --topic test --producer.config ../config/client-ssl.properties
./kafka-console-consumer.sh --bootstrap-server localhost:9093 --topic test --consumer.config ../config/client-ssl.properties

Make sure that the common names (CN) in your certificates match your hostname.
SSL protocol verify CN against hostname. I guess here you should have CN=localhost.
I had a similar issue and that's how I fixed it.

One important information regarding this: The behavior where the CN has to be equal to the hostname can be deactivated, by adding the following line to server.properties:
ssl.endpoint.identification.algorithm=
The default value for this setting is set to https, which ultimately activates the host to CN verification. This is the default since Kafka 2.0.
I've successfully tested a SSL setup (just on the broker side though) with the following properties:
############################ SSL Config #################################
ssl.truststore.location=/path/to/kafka.truststore.jks
ssl.truststore.password=TrustStorePassword
ssl.keystore.location=/path/to/kafka.server.keystore.jks
ssl.keystore.password=KeyStorePassword
ssl.key.password=PrivateKeyPassword
security.inter.broker.protocol=SSL
listeners=SSL://localhost:9093
advertised.listeners=SSL://127.0.0.1:9093
ssl.client.auth=required
ssl.endpoint.identification.algorithm=
You can also find a Shell script to generate SSL certificates (with key- and truststores) alongside some documentation in this github project: https://github.com/confluentinc/confluent-platform-security-tools

Well, both the given answers point out to the right direction, but some more details need to be added to end this confusion.
I generated the certs using this bash script from confluent, and when I looked inside the file, it made sense. I'm pasting the relevant section here:
echo " NOTE: currently in Kafka, the Common Name (CN) does not need to be the FQDN of"
echo " this host. However, at some point, this may change. As such, make the CN"
echo " the FQDN. Some operating systems call the CN prompt 'first / last name'"
There you go. When you're generating the certs, make sure to put localhost (or FQDN) when it asks for first / last name. Do remember that you need to use the same endpoint to expose the broker.

Related

How to fix ActiveMQ Artemis connection timeouts when sslEnabled=true

I'm trying to enable SSL on a Artemis broker and always get this exception when trying to connect:
Exception in thread "main" ActiveMQConnectionTimedOutException[errorType=CONNECTION_TIMEDOUT message=AMQ219013: Timed out waiting to receive cluster topology. Group:null]
at org.apache.activemq.artemis.core.client.impl.ServerLocatorImpl.createSessionFactory(ServerLocatorImpl.java:743)
The code I use to connect is just this:
ActiveMQClient.createServerLocator("tcp://localhost:5500").createSessionFactory();
This is from a fresh install of Artemis 2.23.1 and the only thing I changed from the default broker configuration was to add this acceptor in broker.xml:
<acceptor name="netty-ssl-acceptor">tcp://localhost:5500?sslEnabled=true;keyStorePath=server-keystore.jks;keyStorePassword=securepass</acceptor>
I generated the keystore and truststore using the script provided in this example.
I had first tried a keystore with a cert that is valid for my domain (using a domain-qualified host name in createServerLocator()) but that also gave me the timeout. That is when I went back to fresh installs and tried going through the SSL example.
Various attempts with invalid paths/passwords/certs threw exceptions that led me to what to fix, but so far haven't been able to see what I did wrong with a generic timeout discovering cluster topology.
Anybody have ideas?
You need to specify sslEnabled=true on the client's URL as well so it knows to use SSL, e.g.:
ActiveMQClient.createServerLocator("tcp://localhost:5500?sslEnabled=true").createSessionFactory();
This is done for the JMS connection in the ssl-enabled example which you cited here.
Also, if you're using self-signed certificates then you'll need a truststore for your client as well and you'll need to configure those settings on the client's URL (just like in the example).

Influxdb over SSL connection

I'm a little bit confused about https communication with influxdb. I am running a 1.8 Influxdb instance on a virtual machine with a public IP. It is an Apache2 server but for now I am not willing to use it as a webserver to display web pages to clients. I want to use it as a database server for influxdb.
I obtained a valid certificate from Let's Encrypt, indeed the welcome page https://datavm.bo.cnr.it works properly over encrypted connection.
Then I followed all the instructions in the docs in order to enable https: I put the fullchain.pem file in /etc/ssl directory, I set the file permissions (not sure about the meaning of this step though), I edited influxdb.conf with https-enabled = true and set the path for https-certificate and https-private.key (fullchain.pem for both, is it right?). Then, systemctl restart influxdb. When I run influxdb -ssl -host datavm.bo.cnr.it I get the following:
Failed to connect to https://datavm.bo.cnr.it:8086: Get https://datavm.bo.cnr.it:8086/ping: http: server gave HTTP response to HTTPS client
Please check your connection settings and ensure 'influxd' is running.
Any help in understanding what I am doing wrong is very appreciated! Thank you
I figured out at least a part of the problem. It was a problem related to permissions on the *.pem files. This thing looks weird because if I tip the following, as documentation says, it does not connect.
sudo chmod 644 /etc/ssl/<CA-certificate-file>
sudo chmod 600 /etc/ssl/<private-key-file>
If, instead, I tip the second line with 644 all works perfectly. But this way I'm giving to anyone the permission to read the private key! I'm not able to figure out this point.
UPDATE
If I put inside /etc/ssl/ the symlinks that point to the .pem files that live inside /etc/letsencrypt/live/hostname the connection is refused. Only if I put a copy of the files the ssl connection starts.
The reason I am willing to put the links inside /etc/ssl/ is the automatic renew of the certificates.
Anyone can help?

How do I create a tls cert for a three node server domain that covers the parent domain as well?

I'm not even sure I asked the question right...
I have three servers running minio in distributed mode. I need all three servers to run with TLS enabled. It's easy enough to run certbot, generate a cert for each node, drop said certs into /etc/minio/certs/ and go! but here's where I start running into issues.
The servers are thus:
node1.files.example.com
node2.files.example.com
node3.files.example.com
I'm launching minio using the following command:
MINIO_ACCESS_KEY=minio \
MINIO_SECRET_KEY=secret \
/usr/local/bin/minio server \
-C /etc/minio --address ":443" \
https://node{1...3}.files.example.com:443/volume/{1...4}/
This works and I am able to connect to all three servers from a webbrowser using https with good certs. however, users will connect to the server using the parent domain "files.example.com" (using distributed DNS)
I already ran certbot and generated the certs for the parent domain... and I copied the certs into /etc/minio/certs/ as well as /etc/minio/certs/CAs/ (calling both files "files.example.com-public.crt" and "files.example.com-public.key" respectively)... this did not work. when I try to open the parent domain "files.example.com" I get a cert error (chich I can bypass) indicating the certificate is for the node in which I have connected and not for the parent domain.
I'm pretty sure this is just a matter of putting the cert in the right place and naming it correctly... right? does anyone know how to do that? I also have an idea there might be a way to issue a cert that covers multiple domains... is that how I'm supposed to do this? how?
I already hit up minio's slack channel and posted on their github, but no ones replying to me. not even, "this won't work."
any ideas?
I gave up and ran certbot in manual mode. it had to install apache on one of the nodes, then certbot had me jump through a couple of minor hoops (namely it had me create a new txt record with my DNS provider, and then create a file with a text string on the server for verification). I then copied the created certs into my minio config directory (/etc/minio/certs/) on all three nodes. that's it.
to be honest, I'd rather use the plugin as it allows for an automated cert renewal, but I'll live with this for now.
You could also run all of them behind a reverse proxy to handle the TLS termination using a wildcard domain certificate (ie. *.files.example.com). The reverse proxy would centralize the certificates, DNS, and certbot script if you prefer, etc to a single node, essentially load balancing the TLS and DNS for the minio nodes. The performance hit of "load-balancing" TLS like this may be acceptable depending on your workload, considering the simplification to your current DNS and TLS cert setup.
[Digital Ocean example using nginx and certbot plugins] https://www.digitalocean.com/community/tutorials/how-to-create-let-s-encrypt-wildcard-certificates-with-certbot

Can't connect Filebeat to Logstash

I am new to elasticsearch and I am following the tutorial here:
I have hit a stumbling block as I can connect the servers with the ELK-stack configured with the server that is logging activity to FileBeat.
I have narrowed it down to an issue with the SSL certificates copied from the ELK server as when i check /var/log/messages I get the following error:
usr/bin/filebeat[13730]: transport.go:125: SSL client failed to
connect with: x509: certificate signed by unknown authority (possibly
because of "crypto/rsa: verification error" while trying to verify
candidate authority certificate "serial:16193853809450343771")
How ever, the keys have been copied over and these files are the same on both servers :
cat /etc/pki/tls/certs/logstash-forwarder.crt
When I try to read the syslogs, I get the following message :
sudo tail /var/log/syslog | grep filebeat:
tail: cannot open ‘/var/log/syslog’ for reading: No such file or directory.
I will appreciate any pointers on this
I found a similar issue in the elastic forum in the following link.
In summery, You should add to your FileBeatconfig:
insecure: true
And than see if you manage to connect. If you do, you can use this guidelines for how to configure your ssl connection

Let's encrypt SSL certificate on subdomain

I developed an application for a client which I host on a subdomain, now the problem is that I don't own the main domain/website. They've added a DNS record to point to the IP on which I host that app. Now I want to request a Free & automatic certificate from Let's Encrypt. But when I try the handshake it says
Getting challenge for subdomain.example.com from acme-server...
Error: http://subdomain.example.com/.well-known/acme-challenge/letsencrypt_**** is not reachable. Aborting the script.
dig output for subdomain.example.com:subdomain.example.com
Please make sure /.well-known alias is setup in WWW server.
Which makes sense cause I don't own that domain on my server. But if I try to generate it without the main domain I get:
You must include your main domain: example.com.
Cannot Execute Your Request
Details
Must include your domain example.com in the LetsEncrypt entries.
So I'm curious on how I can just set up a certificate without owning the main domain. I tried googling the issue but I couldn't find any relevant results. Any help would be much appreciated.
First
You don't need to own the domain, you just need to be able to copy a file to the location serving that domain. (You're all set there it sounds like)
Second
What tool are you using? The error message you gave makes me think the client is misconfigured. The challenge name is usually something like https://example.com/.well-known/acme-challenge/jQqx6qlM8u3wpi88N6lwvFd7SA07oK468mB1x4YIk1g. Compare that to your error:
Error: http://example.com/.well-known/acme-challenge/letsencrypt_example.com is not reachable. Aborting the script.
Third
I'm the author of Greenlock, which is compatible with Let's Encrypt. I'm confident that it will work for you.
Install
# Feel free to read the source first
curl -fsS https://get.greenlock.app/ | bash
Usage with existing webserver:
Let's say that:
You're using Apache or Nginx.
You confirm that ping example.com gives the IP of your server
You're exposing http on port 80 (otherwise verification will fail)
Your website is located in /srv/www/example.com
Your email is jon#example.com (must be a real email address)
You want to store your certificate as /etc/acme/live/example.com/fullchain.pem
This is what the command would look like:
sudo greenlock certonly --webroot \
--acme-version draft-11 --acme-url https://acme-v02.api.letsencrypt.org/directory \
--agree-tos --email jon#example.com --domains example.com \
--community-member \
--root /srv/www/example.com \
--config-dir /etc/acme
If that doesn't work on the first try then change out --acme-url https://acme-v02.api.letsencrypt.org/directory for --acme-url https://acme-staging-v02.api.letsencrypt.org/directory while you debug. Otherwise your server could become blocked from Let's Encrypt for too many bad requests. Just know that you'll have to delete the certificates from the staging environment and retry with the production url since the tool cannot tell which certificates are "production" and which ones are "testing".
The --community-member flag is optional, but will provide me with analytics and allow me to contact you about important or mandatory changes as well as other relevant updates.
After you get the success message you can then use those certificates in your webserver config and restart it.
That will work as a cron job as well. You could run it daily and it will only renew the certificate after about 75 days. You could also put a cron job to send the "update configuration" signal to your webserver (normally HUP or USR1) every few days to cause it to start using the new certificates without even restarting (...or just have it restart).
Usage without a web server
If you just want to quickly test without even having a webserver running, this will do it for you:
sudo greenlock certonly --standalone \
--acme-version draft-11 --acme-url https://acme-v02.api.letsencrypt.org/directory \
--agree-tos --email jon#example.com --domains example.com \
--community-member \
--config-dir /etc/acme/
That runs expecting that you DO NOT have a webserver running on port 80, as it will start one temporarily just for the purpose of the certificate.
sudo is required for using port 80 and for writing to root and httpd-owned directories (like /etc and /srv/www). You can run the command as your webserver's user instead if that has the correct permissions.
Use Greenlock as your webserver
We're working on an option to bypass the middleman altogether and simply use greenlock as your webserver, which would probably work great for simple vhosting like it sounds like you're doing. Let me know if that's interesting to you and I'll make sure to update you about it.
Fourth
Let's Encrypt also has an official client called certbot which will likely work just as well, perhaps better, but back in the early days it was easier for me to build my own than to use theirs due to issues which they have long since fixed.
Whats important is the sub domains A record. It should be the IP Address of from where you are trying to request the sub domains certificate.