[Bro] 5 node cluster
dwaters at bioteam.net
Sun Oct 9 08:01:21 PDT 2016
# Serial MAC ProductCode Driver Version
0 482741 00:60:dd:43:84:4a 10G-PCIE2-8C2-2S-SYNC myri_snf 18.104.22.168782 Valid
1 482741 00:60:dd:43:84:4b 10G-PCIE2-8C2-2S-SYNC myri_snf 22.214.171.124782 Valid
It may not make any difference see below. I will try eth2, and that would
reduce it to the time source default settings on the cards.
myri_snf INFO: myriC0: my ether interface name is eth2
myri_snf INFO: eth2: Will use skbuf frags (4096 bytes, order=0)
myri_snf INFO: Enabling host timestamping.
On Sat, Oct 8, 2016 at 3:56 PM, Azoff, Justin S <jazoff at illinois.edu> wrote:
> > On Oct 8, 2016, at 1:44 PM, Darrain Waters <dwaters at bioteam.net> wrote:
> > Turns out it was a simple config issue (like most times & RTFM), and
> traffic is flowing to snf0. My workers were not using the snf0 interface as
> you must if you compiled using the myricom sniffer.
> Interesting.. which cards and which version of the snf drivers are you
> I use interface=p1p1 on our clusters and have never had an issue. As I
> understood things using snf0 was just an alias for 'the first myricom card'
> - Justin Azoff
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Bro