| Message ID | 20250417-svc-somaxconn-v1-1-ff5955cc9f45@kernel.org (mailing list archive) |
|---|---|
| State | New |
| Headers | show |
| Series | sunrpc: allow SOMAXCONN backlogged TCP connections | expand |
From: Chuck Lever <chuck.lever@oracle.com> On Thu, 17 Apr 2025 14:54:36 -0400, Jeff Layton wrote: > The connection backlog passed to listen() denotes the number of > connections that are fully established, but that have not yet been > accept()ed. If the amount goes above that level, new connection requests > will be dropped on the floor until the value goes down. If all the knfsd > threads are bogged down in (e.g.) disk I/O, new connection attempts can > stall because of this. > > [...] Applied to nfsd-testing, thanks! [1/1] sunrpc: allow SOMAXCONN backlogged TCP connections commit: 1fe4a78475a5203a9ee1d9ede1bc2043cb505382 -- Chuck Lever
diff --git a/net/sunrpc/svcsock.c b/net/sunrpc/svcsock.c index 72e5a01df3d352582a5c25e0b8081a041e3792ee..60f2883268faf15b8e01a5d12d67a5d01b278a5e 100644 --- a/net/sunrpc/svcsock.c +++ b/net/sunrpc/svcsock.c @@ -1542,7 +1542,7 @@ static struct svc_xprt *svc_create_socket(struct svc_serv *serv, if (protocol == IPPROTO_TCP) { sk_net_refcnt_upgrade(sock->sk); - if ((error = kernel_listen(sock, 64)) < 0) + if ((error = kernel_listen(sock, SOMAXCONN)) < 0) goto bummer; }
The connection backlog passed to listen() denotes the number of connections that are fully established, but that have not yet been accept()ed. If the amount goes above that level, new connection requests will be dropped on the floor until the value goes down. If all the knfsd threads are bogged down in (e.g.) disk I/O, new connection attempts can stall because of this. For the same rationale that Trond points out in the userland patch [1], ensure that svc_xprt sockets created by the kernel allow SOMAXCONN (4096) backlogged connections instead of the 64 that they do today. [1]: https://lore.kernel.org/linux-nfs/20240308180223.2965601-1-trond.myklebust@hammerspace.com/ Cc: Trond Myklebust <trond.myklebust@hammerspace.com> Signed-off-by: Jeff Layton <jlayton@kernel.org> --- The backlog was set at 5 in v2.4.0. Neil changed it to 64 in 2002, and it's been set to that ever since. This is particularly needed for servers that are started via the netlink interface, so we don't regress the behavior that Trond's patch fixed with the sockfd-passing interface. --- net/sunrpc/svcsock.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) --- base-commit: 01bc8a703933a0b7b76e6d6fbc58e4f1d5b64ae5 change-id: 20250417-svc-somaxconn-b6413c1d39bf Best regards,