Probability of simultaneous death of disks in RAID1?

Question

Hard disks

Given: two disks of the same model M in RAID1 (let's call them S0 and S1). The new ones were installed at the same time.
After working for two years, S0 gave his soul to God. Almost immediately, an S2 drive, the same model M, was ordered and installed and delivered to the RAID (it took 2 weeks to order + delivery). A month after the death of S0, S1 ordered to live long. Ordered and expected S3.
And here the question arises ... Intuitively, it is felt that now losing both S2 and S3 in two years is more likely than it was with S0 and S1, but there is nothing to prove (from the ter. ver. everything is bad ...).
Actually questions:
1. Is it true that the probability of a simultaneous (or rather - with a shorter margin of time to replace the disk) failure of S2 and S3 has increased.
2. What benefit did RAID1 bring to you personally, did you have similar cases and torments? CHADNT?
Home computer, software RAID ZFS, models of all disks - Seagate ST1000DM003, there is an external backup of important data (by the way, a nuance to note - it also died almost simultaneously with S0, I had to order a new external disk along with S2, i.e. actually S1, on snot, allowed the data to survive the demise of the twin brother S0 and the external drive). After all these nervous events, backups also go weekly to the Amazonian "glacier".

Answer 1

The probability is not zero, but no one will tell you the exact value, it has not increased, but it always exists, during my work I lost both disks several times in Raid 1 for half an hour. A raid is not a replacement for a backup, the benefit of a raid in this case is, nevertheless, the probability is lower than the output of one disk, the read speed from the array is higher.

Answer 2

probability 50%, always took discs for a raid of the same series but from different parties to protect yourself from the marriage of the party and the story from the post)

Answer 3

1) There is one such thing that the raid is not about the reliability of data storage, it is high availability of data, you need reliability, store two more copies from different places.
2) Raid 1 brings a simple benefit, it allows data to be available during the death of one disk. It does not guarantee the reliability of storage at all, practice has shown that the death of the second disk happens quite often and is not equal to zero, this is partly due to the fact that the server is turned off to change the disk (and this is probably the most stressful situation for the server in ordinary life) + mechanically change the disk , which often sometimes leads to an error, for example, they changed the wrong disk (it happened once in our hatzner)

Answer 4

The resource of the disks is quite constant. When one disk flies to others, the load increases. So, yes, they fail together. And it's better to change everything at once.

Answer 5

Sounds like you need to change your wheels...