100% @lucasbradstreet , thank you!

metrics fix is working

Great :)

cool

It looks like all your latency is batch timeout latency now. Are you using 100ms?

100ms yes

can very likely be removed so that we use your 50ms default now

i’ll be shunting backfill events through a core.async input soon, to catch all the missing work from the various downtimes. we should see some nice activity on the graphs then

Ah cool.

Looks like it's all under control now

yup, it is. thank you for being so responsive, Lucas. it’s tremendously helpful

No worries. Glad you're getting good results out of it now

yeah thanks so much Lucas, it's great to have a really stable system now (touch wood :P)