Hello. On Sat, Aug 14, 2021 at 08:15:16AM +0800, brookxu <brookxu.cn@xxxxxxxxx> wrote: > 1. If a large number of containers are created and deleted, > there will be a lot of error logs. > 2. According to an error log, we cannot better understand > the actual pressure of resources. > > Therefore, perhaps we should use a failcnt counter to count > the number of failures, so that we can easily understand the > actual pressure of resources and avoid too many error log.. This is an understandable use case and generally the implementation via the counter is good as well. However, the non-hierarchical failcnt interface looks like v1ism to me (I think new features should come with v2 first in mind). What about exposing this in misc.events file with max.$res_name entries? Or if the hierarchical reporting is unnecessary now, there can be just misc.events.local for starters. (That reminds me the forgotten pids.events[.local] rework [1], oops.) Michal https://lore.kernel.org/lkml/20191128172612.10259-1-mkoutny@xxxxxxxx/#t
