mbuono
09/18/2023, 7:39 PMfleetctl from a file managed in git. Currently our packs each have a target of some class of hosts, eg Linux or Mac, and im looking to replicate the functionality of "Run these queries on these hosts". We don't have access to Teams or anything, but im assuming this is all possible using something like fleetctl apply for queries. Where can i find docs on what config keys each "query" object in a file that fleetctl will eventually apply has? I ran fleetctl upgrade-packs just to get a sense, and i see apiVersion , kind , and spec (with some subkeys there), but just wondering if theres an official place i can look through what options i have when configuring queriesmbuono
09/18/2023, 9:45 PMfleetctl , but now i see no result messages being pushed to the output topic (we use pubsub, specified in an env variable) and when i go to hosts and look at Schedule for one of them, i see nothing. Am i missing some step to ensure that queries are actually scheduled and are pushed to the correct output stream? Fleet also seems to think (when i click "manage automations" to just look around) that the log destination is "filesystem", but when i dump its own config i correctly see
osquery:
...
status_log_plugin: pubsub
result_log_plugin: pubsub
so i dont know why the UI would say that the output plugin is filesystemmbuono
09/19/2023, 2:50 PMautomations_enabled: true effectively the way to schedule a query in versions after 4.35? So you set an interval, an optional platform, and then enable automations, and your query will be run on the specified interval and logged to your output plugin that you specified in the fleet config? Or are automations not related to query scheduling, and simply setting interval and platform is enough to schedule a query?Kathy Satterlee
09/19/2023, 4:46 PMinterval set and automations enabled.