Our great sponsors
-
-
-
WorkOS
The modern identity platform for B2B SaaS. The APIs are flexible and easy-to-use, supporting authentication, user identity, and complex enterprise features like SSO and SCIM provisioning.
-
Ah so more like a Rbldnsd used in the RBL/RSL servers. I've run rbldnsd [1] as both a lookup for spammers/malware and as an authoritative server.
[1] - https://github.com/spamhaus/rbldnsd
I'm not 100% certain if I'm understanding the requirement correctly - but would something like this help?
https://github.com/ctxis/SnitchDNS
Hello, author here. That's a database driven DNS server alright. (Bonus: it's got a web admin interface.) There are DNS implementations with various backends; that's kind of the point.
I'm not sure that a mainstream SQL database is really a good target for network telemetry and e.g. access logging artifacts. Not talking about a time stream database either. There is an architecture here, and it's predicated on not collecting "all the things" in a central place.
Example: In this model, a service / server you're monitoring might have a couple of Redis keys which get incremented every time there's a successful or unsuccessful login. Maybe there's a redis hashkey with fails for individual accounts too.
There might be a graph somewhere of the login / attempt rates. It would query the summary redis keys (via the DNS) once a minute (doing whatever it needs to keep historical datapoints for however long they're needed).
If the rate skyrockets, maybe the hashkey with account-level granularity is consulted but most of the time it wouldn't be.
There might also be a Zabbix alarm somewhere querying the same keys, and if a threshold setting is exceeded, then an alarm is sent.
It's pull, not push. It's easy enough to write something to make the periodic queries and post them to e.g. ElasticSearch and graph it with Kibana.
So the question concerns the SIEM part. Something like Splunk is married to its database (their pricing model is based on how much data you want to put into that database). Something like the Yeti Threat Intelligence Platform (TIP) (https://github.com/yeti-platform/yeti) comes with the ability to manage and orchestrate a large number of periodic or event-driven tasks and therefore has the capability to generate the periodic DNS queries; it's been a few years, but its graphing capabilities didn't compare to ELK when I looked at it.
There's a lot of overlap with SCADA as well. All of the necessary features I've mentioned can be assembled from open source projects.
Is there some SIEM, TIP or Ops product out there, with an active userbase, which has the periodic task capability, alarming, and graphing?