Official Node periodically offline

[wangminqi]

Not fully checked, but this behavior problem seems not exist before the offline & recovery of official nodes around May 10th.

So the 2 official bootnodes will constantly be offline 3 "sessions" out of 10 "sessions" (not producing block).

The behavior is the following:
So currently there are 5 collators: METASPAN, STAKE PLUS, GATOTECH, and two official nodes from parallel team.
For each "session", totally 5 blocks, they will each generate one block as collator.

The problem is, there is a very clear pattern that both two official nodes will stably not generate blocks 3 "session" out of 10 "session".

For us, currently it will make block producing slower (missing 6 blocks out of each 50), which will make calculation of APY biased.