(I sent this before, but the mail wasn't going through. Trying again. Sorry for a possible duplicate.) On 2013-11-13T12:24:49, David Teigland <teigland@xxxxxxxxxx> wrote: > There are significant subtleties involved in using SO_LINGER with tcp. > I do not know whether applying it here would be good or bad in the tcp > case, much less the sctp case. I assume that the default tcp/sctp > behavior exists for some good reasons, and I'd like any change to be > reviewed by experts in the area. That'd be awesome. I wonder if we can find some? ;-) The goal here is that we know the other endpoint is down (we received a node down event and have completed fencing at that stage). Hence, SO_LINGER to speed up the shutdown of the socket seems appropriate. See http://blog.netherlabs.nl/articles/2009/01/18/the-ultimate-so_linger-page-or-why-is-my-tcp-not-reliable And for SCTP it is the same: http://tools.ietf.org/html/draft-ietf-tsvwg-sctpsocket-01#section-7.1.4 (We may actually only want to set SO_LINGER for the node down event case, not generally. On receiving node down, set SO_LINGER as described here. Otherwise, we may hit the corner cases in the first reference; but we're already exposed to that today.) > We do not want to risk harming any common situations for the sake of > this uncommon and avoidable one. I really would love to know how we can avoid it. We have a few customers who can reproduce this. Regards, Lars -- Architect Storage/HA SUSE LINUX Products GmbH, GF: Jeff Hawn, Jennifer Guild, Felix Imendörffer, HRB 21284 (AG Nürnberg) "Experience is the name everyone gives to their mistakes." -- Oscar Wilde -- Linux-cluster mailing list Linux-cluster@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/linux-cluster
