[Buildbot] #3183: fix network "spottiness" on new infra
Buildbot trac
trac at buildbot.net
Wed Feb 11 00:09:14 UTC 2015
#3183: fix network "spottiness" on new infra
---------------------+--------------------------
Reporter: dustin | Owner: verm
Type: defect | Status: accepted
Priority: critical | Milestone: sys - other
Version: | Resolution:
Keywords: |
---------------------+--------------------------
Comment (by dustin):
OK, so let's reason through this. All of these hosts are showing up on
Po33, but there are only two physical ports in Po33. So there's a bridge
somewhere.
The only thing plugged into Po33 is vm1. Does vm1 have a bridge? Yup.
{{{
bridge0: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> metric 0 mtu
1500
ether 02:fe:61:83:ab:00
nd6 options=1<PERFORMNUD>
id 00:00:00:00:00:00 priority 32768 hellotime 2 fwddelay 15
maxage 20 holdcnt 6 proto rstp maxaddr 2000 timeout 1200
root id 00:00:00:00:00:00 priority 32768 ifcost 0 port 0
member: lagg0 flags=143<LEARNING,DISCOVER,AUTOEDGE,AUTOPTP>
ifmaxaddr 0 port 6 priority 128 path cost 20000
member: tap0 flags=143<LEARNING,DISCOVER,AUTOEDGE,AUTOPTP>
ifmaxaddr 0 port 9 priority 128 path cost 2000000
member: vlan0 flags=143<LEARNING,DISCOVER,AUTOEDGE,AUTOPTP>
ifmaxaddr 0 port 7 priority 128 path cost 2000000
}}}
I'll wager that putting a lagg port into a bridge allows traffic to bridge
in one port and out the other.
I took igb1 down on that host, just to be sure. From what I can tell it
was already disabled by the switch. Even before doing so, I wasn't able
to replicate the network error, although I've gotten various emails today
indicating failure.
I'm not sure how to fix this. I know that LACP is, in part, intended to
replace STP, so it doesn't make sense to run STP across those two links.
Maybe this is a FreeBSD bug?
--
Ticket URL: <http://trac.buildbot.net/ticket/3183#comment:36>
Buildbot <http://buildbot.net/>
Buildbot: build/test automation
More information about the bugs
mailing list