13 KiB
SSL Certificate Exporter
The blackbox_exporter allows you to test the expiry date of a certificate as part of its HTTP(S) probe - which is great. It doesn't, however, tell you which certificate in the chain is nearing expiry or give you any other information that might be useful when sending alerts.
For instance, there's a definite value in knowing, upon first receiving an alert, if it's a certificate you manage directly or one further up the chain. It's also not always necessarily clear from the address you're polling what kind of certificate renewal you're looking at. Is it a Let's Encrypt, in which case it should be handled by automation? Or your organisation's wildcard? Maybe the domain is managed by a third-party and you need to submit a ticket to get it renewed.
Whatever it is, the SSL exporter gives you visibility over those dimensions at the point at which you receive an alert. It also allows you to produce more meaningful visualisations and consoles.
Table of Contents
Created by gh-md-toc
Building
make
./ssl_exporter <flags>
Similarly to the blackbox_exporter, visiting
http://localhost:9219/probe?target=example.com:443
will return certificate metrics for example.com. The ssl_probe_success
metric indicates if the probe has been successful.
Docker
docker pull ribbybibby/ssl-exporter
docker run -p 9219:9219 ribbybibby/ssl-exporter:latest <flags>
Release process
- Update the
VERSION
file in this repository and commit to master - This github action will add a changelog and upload binaries in response to a release being created in Github
- Dockerhub will build and tag a new container image in response to tags of the
format
/^v[0-9.]+$/
Usage
usage: ssl_exporter [<flags>]
Flags:
-h, --help Show context-sensitive help (also try --help-long and
--help-man).
--web.listen-address=":9219"
Address to listen on for web interface and telemetry.
--web.metrics-path="/metrics"
Path under which to expose metrics
--web.probe-path="/probe" Path under which to expose the probe endpoint
--config.file="" SSL exporter configuration file
--log.level="info" Only log messages with the given severity or above. Valid
levels: [debug, info, warn, error, fatal]
--log.format="logger:stderr"
Set the log target and format. Example:
"logger:syslog?appname=bob&local=7" or
"logger:stdout?json=true"
--version Show application version.
Metrics
Metric | Meaning | Labels |
---|---|---|
ssl_cert_not_after | The date after which a peer certificate expires. Expressed as a Unix Epoch Time. | serial_no, issuer_cn, cn, dnsnames, ips, emails, ou |
ssl_cert_not_before | The date before which a peer certificate is not valid. Expressed as a Unix Epoch Time. | serial_no, issuer_cn, cn, dnsnames, ips, emails, ou |
ssl_file_cert_not_after | The date after which a certificate found by the file prober expires. Expressed as a Unix Epoch Time. | file, serial_no, issuer_cn, cn, dnsnames, ips, emails, ou |
ssl_file_cert_not_before | The date before which a certificate found by the file prober is not valid. Expressed as a Unix Epoch Time. | file, serial_no, issuer_cn, cn, dnsnames, ips, emails, ou |
ssl_ocsp_response_next_update | The nextUpdate value in the OCSP response. Expressed as a Unix Epoch Time | |
ssl_ocsp_response_produced_at | The producedAt value in the OCSP response. Expressed as a Unix Epoch Time | |
ssl_ocsp_response_revoked_at | The revocationTime value in the OCSP response. Expressed as a Unix Epoch Time | |
ssl_ocsp_response_status | The status in the OCSP response. 0=Good 1=Revoked 2=Unknown | |
ssl_ocsp_response_stapled | Does the connection state contain a stapled OCSP response? Boolean. | |
ssl_ocsp_response_this_update | The thisUpdate value in the OCSP response. Expressed as a Unix Epoch Time | |
ssl_probe_success | Was the probe successful? Boolean. | |
ssl_prober | The prober used by the exporter to connect to the target. Boolean. | prober |
ssl_tls_version_info | The TLS version used. Always 1. | version |
ssl_verified_cert_not_after | The date after which a certificate in the verified chain expires. Expressed as a Unix Epoch Time. | chain_no, serial_no, issuer_cn, cn, dnsnames, ips, emails, ou |
ssl_verified_cert_not_before | The date before which a certificate in the verified chain is not valid. Expressed as a Unix Epoch Time. | chain_no, serial_no, issuer_cn, cn, dnsnames, ips, emails, ou |
Configuration
TCP
Just like with the blackbox_exporter, you should pass the targets to a single instance of the exporter in a scrape config with a clever bit of relabelling. This allows you to leverage service discovery and keeps configuration centralised to your Prometheus config.
scrape_configs:
- job_name: "ssl"
metrics_path: /probe
static_configs:
- targets:
- example.com:443
- prometheus.io:443
relabel_configs:
- source_labels: [__address__]
target_label: __param_target
- source_labels: [__param_target]
target_label: instance
- target_label: __address__
replacement: 127.0.0.1:9219 # SSL exporter.
HTTPS
By default the exporter will make a TCP connection to the target. This will be
suitable for most cases but if you want to take advantage of http proxying you
can use a HTTPS client by setting the https
module parameter:
scrape_configs:
- job_name: "ssl"
metrics_path: /probe
params:
module: ["https"] # <-----
static_configs:
- targets:
- example.com:443
- prometheus.io:443
relabel_configs:
- source_labels: [__address__]
target_label: __param_target
- source_labels: [__param_target]
target_label: instance
- target_label: __address__
replacement: 127.0.0.1:9219
This will use proxy servers discovered by the environment variables HTTP_PROXY
,
HTTPS_PROXY
and ALL_PROXY
. Or, you can set the proxy_url
option in the module
configuration.
The latter takes precedence.
File
The file
prober exports ssl_file_cert_not_after
and
ssl_file_cert_not_before
for PEM encoded certificates found in local files.
Files local to the exporter can be scraped by providing them as the target parameter:
curl "localhost:9219/probe?module=file&target=/etc/ssl/cert.pem"
The target parameter supports globbing (as provided by the doublestar package), which allows you to capture multiple files at once:
curl "localhost:9219/probe?module=file&target=/etc/ssl/**/*.pem"
One specific usage of this prober could be to run the exporter as a DaemonSet in Kubernetes and then scrape each instance to check the expiry of certificates on each node:
scrape_configs:
- job_name: "ssl"
metrics_path: /probe
params:
module: ["file"]
target: ["/etc/kubernetes/**/*.crt"]
kubernetes_sd_configs:
- role: node
relabel_configs:
- source_labels: [__address__]
regex: ^(.*):(.*)$
target_label: __address__
replacement: ${1}:9219
Configuration file
You can provide further module configuration by providing the path to a
configuration file with --config.file
. The file is written in yaml format,
defined by the schema below.
modules: [<module>]
<module>
# The type of probe (https, tcp, file)
prober: <prober_string>
# How long the probe will wait before giving up.
[ timeout: <duration> ]
# Configuration for TLS
[ tls_config: <tls_config> ]
# The specific probe configuration
[ https: <https_probe> ]
[ tcp: <tcp_probe> ]
<tls_config>
# Disable target certificate validation.
[ insecure_skip_verify: <boolean> | default = false ]
# The CA cert to use for the targets.
[ ca_file: <filename> ]
# The client cert file for the targets.
[ cert_file: <filename> ]
# The client key file for the targets.
[ key_file: <filename> ]
# Used to verify the hostname for the targets.
[ server_name: <string> ]
<https_probe>
# HTTP proxy server to use to connect to the targets.
[ proxy_url: <string> ]
<tcp_probe>
# Use the STARTTLS command before starting TLS for those protocols that support it (smtp, ftp, imap)
[ starttls: <string> ]
Example Queries
Certificates that expire within 7 days:
ssl_cert_not_after - time() < 86400 * 7
Wildcard certificates that are expiring:
ssl_cert_not_after{cn=~"\*.*"} - time() < 86400 * 7
Certificates that expire within 7 days in the verified chain that expires latest:
ssl_verified_cert_not_after{chain_no="0"} - time() < 86400 * 7
Number of certificates presented by the server:
count(ssl_cert_not_after) by (instance)
Identify failed probes:
ssl_probe_success == 0
Peer Certificates vs Verified Chain Certificates
Metrics are exported for the NotAfter
and NotBefore
fields for peer
certificates as well as for the verified chain that is
constructed by the client.
The former only includes the certificates that are served explicitly by the target, while the latter can contain multiple chains of trust that are constructed from root certificates held by the client to the target's server certificate.
This has important implications when monitoring certificate expiry.
For instance, it may be the case that ssl_cert_not_after
reports that the root
certificate served by the target is expiring soon even though clients can form
another, much longer lived, chain of trust using another valid root certificate
held locally. In this case, you may want to use ssl_verified_cert_not_after
to
alert on expiry instead, as this will contain the chain that the client actually
constructs:
ssl_verified_cert_not_after{chain_no="0"} - time() < 86400 * 7
Each chain is numbered by the exporter in reverse order of expiry, so that
chain_no="0"
is the chain that will expire the latest. Therefore the query
above will only alert when the chain of trust between the exporter and the
target is truly nearing expiry.
It's very important to note that a query of this kind only represents the chain of trust between the exporter and the target. Genuine clients may hold different root certs than the exporter and therefore have different verified chains of trust.
Grafana
You can find a simple dashboard here that tracks certificate expiration dates and target connection errors.