We had a router failure on Saturday night, which took the perl.org network equipment offline for most of the night. We were able to cobble things back together, but as of now (Sunday afternoon) we're still working out some kinks in our solution. In particular we're not getting the throughput we expect. You may notice that some services are slower than normal.
We'll update this blog post with any significant updates.
Update July 1, 2018 10:00pm: set hw.vtnet.tso_disable=1
is the magic that got our speeds back up and helped with some other weirdness.
Update July 1, 2018 11:23pm: IPv6 is working again too. Time for sleep.
Post a Comment