[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Xen-devel] [PATCH v16 2/7] remus: introduce remus device

On Fri, Jul 18, 2014 at 11:58 AM, Ian Campbell <Ian.Campbell@xxxxxxxxxx> wrote:
On Fri, 2014-07-18 at 11:54 -0400, Shriram Rajagopalan wrote:

> if (info->netbuf) {
> Â...
> } else
> Â LOG(WARN, "Network buffering disabled. Failover may not be successful")

If this is the case then why are we offering that option to the user in
the first place?

What is the use case for disabling network buffering given that it
breaks failover?

The keyword above is "may". Without network buffering, ongoing TCPÂ
connections "may" be hung, depending on whether any communicationÂ
happened during the failed checkpoint. Same applies to disk.

If one can control the network interactions or if the application is capable of recovering
from lost TCP connections (or lost UDP packets for that matter), then it stands to benefit
from no-network buffering as it eliminates the latency overhead introduced by buffering every
packet [ 0.5 x checkpoint-interval + RTT between primary-backup].Â

Xen-devel mailing list



Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.