Can someone aim me at some RTFM that describes the sensor release and patching process, please? I'm lost trying to understand: When a new version 'n' of the sen

Can someone aim me at some RTFM that describes the sensor release and patching process, please? I'm lost trying to understand: When a new version 'n' of the sensor is released, we upgrade a selected batch of machines and do some tests (mostly waiting around :-)) to see that all is well. Then we upgrade the rest of the fleet by OU. However, 'cause we're scaredy cats, we leave some critical kit on n-1 for longer. And some really critical kit even on n-2. (Yeah, there's a risk in not applying patches I know but there are other outage-related risks that we balance; forget that for now) Our assumption is that n-1, n-2, etc are old, stable releases, and so when fan and shit collided yesterday, we just hopped on the console and did a policy update to revert to n-2 and assumed we'd dodged the bullet. But of course, that failed... you know what they say about assumptions :-) So in a long-winded way that leads to my three questions: Why did the 'content update' take out not just n but n-whatever sensors equally as effectively? Are the n-whatever versions not actually stable? And if the n-whatever versions are not actually stable and are being patched, what's the point of the versioning? Cheers!