anyone having a routing loop?
mnathani: we have two SMF runs between the two.  1 Gbps for the network, 40 Gbps for Ceph
although the Arista QSFP+ SMF adaptor we got doesn't seem to want to run at full capacity.  We're getting enough bandwidth out of it though, so it's not a problem
tmarble: yeah, live migrations are still totally impressing me..  I migrated the tmux box that I run IRC on, like 7 times in one day haha .. just to test all the different hosts
I was never disconnected, no packet loss that I could tell (although there probably was a few lost)
when I did it over the Infiniband network, the whole migration took something like 3 or 4 seconds
my 174.136.x.x addresses seem to be unreachable at the moment
seems to have been intermitent throughout the day
acf_: yeah you paste an mtr report
just getting that now
oops, *can you
http://paste.debian.net/1078494/
trace to 174.136.111.132
OK let me look into this further
this is from HE in Vancouver, although it is also unreachable from my Amazon Cloudfront reverse proxy setup
thanks
it's been going on since at least 10:30 or 11 this morning
whenever I posted in the channel earlier is when I noticed it
looks like it's also affecting the HE looking glass default core3.fmt2.he.net in case that helps for testing
acf_: how does it look now?
looks back to normal. thanks!
well, I made a routing change about an hour ago.. or maybe two...  I emailed you to test it again, but I didn't get a reply.  I also didn't get any other tickets, so I kinda assumed it was OK
ahh, sorry about that. I should probably check that email...
So yeah, I think I know what happened
no worries
sorry for the downtime!
haha all good. it's amazing it's as rare as it is
:)
I'm opening up the /24 announcement we did earlier, to the full /20
If we have problems, it better to know now than the last minute
acf_: I can still ping your IP
yeah, sorry for the late response, but everything still looks good from here