On Apr 16 we noticed a big drop in crash-free sessions.
Most of the frequent unhandled errors in this period are related to failed network requests (401,403,404).
Did we change how/when we mark sessions as crashed?
This is a screenshot of a recent release that I would consider fine/healthy.
As you can see, the crash-free rate is 58% which looks really scary. Again, most of the frequent unhandled errors are network-related. The current situation makes it hard to spot spikes and actual bugs/errors that require an engineering action.
On Apr 16 we noticed a big drop in crash-free sessions.
Most of the frequent unhandled errors in this period are related to failed network requests (401,403,404).
Did we change how/when we mark sessions as crashed?
This is a screenshot of a recent release that I would consider fine/healthy.
As you can see, the crash-free rate is 58% which looks really scary. Again, most of the frequent unhandled errors are network-related. The current situation makes it hard to spot spikes and actual bugs/errors that require an engineering action.