I’ve learned not to end a sensor just because of a noisy line. Once early in my CGM life I just rode it out for a day or so, the line eventually went smooth again, and I got another few weeks out of the sensor.
Nor do I end it when it goes crazy up-and-down and then goes blank. I just wait an hour or three and it comes back all nice and friendly. Like a cat.
This is my strategy, too. I once had a sensor that gave no useful data for a couple days despite a few restarts, but then came back and lasted another 10 days.
Doing things like this is a big part of why I’m able to get a month out of each sensor.
During the little to no data periods I go back to resting g 12x a day (or more).
I am not sure how well it works for everyone though. We always wait for a while after the first time it stops. For us, very quickly (a day typically) it goes to missing 8 hours or more, at which stage we quit. The problem for us is that, without a CGM, it is really easy to miss large spikes that wreak havoc on an A1c. My son has a hard time knowing he is high unless he is really high.
And—the problem for us is that these days we often get this first miss on day 4 or 5.
I typically will restart the sensor if it doesn’t come back within 24 hours. And then restart two or three more times before giving up on it. 
But I pay out of pocket. So I can’t say I’d necessarily go through this much effort if my sensors were covered.
Without a CGM, I end up testing every hour at least, and sometimes more often than that. I also don’t feel highs or lows unless they get extreme. It’s not ideal, but I managed for more than 20 years (including adolescence) without a CGM. So I see it more as an extra bonus than an absolute necessity to survival. I would not want to give up my CGM permanently, but a few days of no data is not going to harm me as long as I’m testing a lot to stay on top of things. I can see how this view may be different for those who have had access to CGM from diagnosis.
Interestingly, our CGM is covered, and my son will quite often take 2-5 days off the CGM when one dies. We feel like being able to function without a CGM is a good thing to practice. But we really wouldn’t want to be without this data for a long period of time.
Does he correct less aggressively for his peaks then?
Yes, I wouldn’t say he is timid, because he is not, but he certainly doesn’t add a temp basal onto a peak. He will frequently stack on the big hormone peaks, but I would say he is quite a bit less aggressive without CGM. He also has to eat a lot more carbs, because his Basal IQ lands 80+% of his peaks with no carbs.