Fair enough. That test has more problems as the nightly worst-case is just a total killer time-wise... I really think it should have a wall-clock expiration deadline inside too.
Dawid On Thu, Nov 4, 2021 at 2:26 PM Robert Muir <[email protected]> wrote: > On Thu, Nov 4, 2021 at 9:17 AM Dawid Weiss <[email protected]> wrote: > > > > I think there should be a better solution based on throttling. This test > runs with 4 threads - I can imagine processes where an order of magnitude > more threads will index in parallel (+ merges). So it should just work. I > don't know how to fix it though it's causing additional noise for me when I > scan the mailing list. Isn't switching to a different directory going to be > a similar kind of non-fix workaround? > > > > I don't think of it as a non-fix workaround. the test is supposed to > be testing merge-on-getreader right? It isn't supposed to be testing > throttling of excessive filehandles, although it is nice that it > "discovered" an issue there. > > IMO, ideally we would add a separate test for that that explicitly > tries to max out the file handles and ensures IW does the right thing. > > --------------------------------------------------------------------- > To unsubscribe, e-mail: [email protected] > For additional commands, e-mail: [email protected] > >
