[ofa-general] Re: IB Bonding errors with recent kernel

[Or Gerlitz]

Pradeep Satyanarayana wrote:
> I looked through the mailing list going back more than a year. The closest I can find to this issue (-EINVAL) was when you reported problems with junk MGID on a child interface (and that works properly now).

Yes, this was observation way back, since then, I realized that this 
happens also regardless of vlans, bonding, anything.

> I agree that the -EAGAIN problem has been known for some time now. However, this issue with IPoIB bonding is new. My recollections are that it all worked properly around end October. I had not tested since then, so this is something that must have cropped in the interregnum.
Moni just sent you a note on the matter.

> Thanks for alerting me to the fact that the ib-bond script was 
> deprecated. Again this seemed to all work about 6 weeks ago. Is that 
> (ib-bond is deprecated) documented somewhere?
I don't know about whether this is documented (Moni?). Anyway, under all 
aspects: design, implementation, main line kernel maintainer and distro 
deployment, bonding is bonding is bonding, where you just happen to use 
it over IPoIB NICs and not Eth NICs.

I wouldn't recommend that the fact of you using a short term solution 
provided by the ofed ib-bonding package make you take that to the extent 
of relying on proprietary helper script to be well maintained, etc.

Or.