Hi, thanks for looking into this!
On Wed, 22 Jan 2020 at 18:07, Tomas Vondra <tomas.vondra@2ndquadrant.com> wrote:
>
> So what was the actual rate of table creation? How long did it take to
> get to this state?
Currently we create tables at about 1.8 Hz. We drop them at the same
rate, almost all tables are short-lived (between 1d and 1h). We
started seeing this error roughly 1.5 years after we started using
this system, but in the beginning the rate was not as high as it is
now.
I don’t have all historical occurrences at hand, but in the past 14
days, the error has occurred twice.
> What kernel/filesystem was used? What did lsof say?
We use kernel 4.15.0-1036-gcp and the file system is ext4. Lsof does
not currently list the file in the most recent error message, but I
don't know what the state was at the time of the error.
Best,
Ruud