In this answer, I will respond to some of the points raised in the question, however without any expectation of completeness.
Isn't a decent fluency in English the requirement for senior academic positions (regardless of native languages)?
To some extent, it is, although there are probably many different interpretations of what "decent fluency" means around.
If a professor in non-English speaking country is not fluent in English, how does he supervise graduate students?
Maybe I'm misunderstanding this, but ... simply in their non-English language? There is no general requirement that communication with graduate students (including foreign graduate students!) would have to happen in English.
For supervising/conducting cutting-edge research, someone should be able to read technical literature. Research papers are published in English (and there is no translation normally).
This is true, however I have noticed that there can be extremely wide gaps between the ability to read and understand complicated English texts and the ability to express complex ideas in English. Hence, a capable researcher can be perfectly able to understand technical literature written in English, and yet be unable to publish their own ideas in English.
Of course, this is not generally the case in all non-English-speaking places, but it probably depends on how English is trained or learned.
I understand it is easier for native English speakers to read and write in scientific literature, but we don't speak their language, we speak an international language, which is their native language too.
If it is their native language, too, then we do speak their language. Whether our motivation for learning or using English is to communicate with native speakers or with the international community doesn't matter; the result is that, as you say, "it is easier for native English speakers to read and write in scientific literature".
is it reasonable and justifiable to give a senior academic position to someone who cannot read 99% of recent research findings in his field?
As explained above, comprehension is not necessarily the same as expression. And even then, this statement possibly neglects that some fields might rely less on English than what I am used to from CS.
Is it acceptable for a professor or even a PhD student to have no understanding of English language?
No, rather not.
Coming back to your titular question:
Isn't English the common academic language?
Yes, in many fields, it is. But this mere fact has little influence on the effects you allude to. It neither means that English material is inaccessible to someone with low skills at writing in English, nor that ideas and surveys are always developed and conducted in English from the start.