Donate to e Foundation | Murena handsets with /e/OS | Own a part of Murena! Learn more

Commit 3dedbb5c authored by Jason Baron's avatar Jason Baron Committed by Trond Myklebust
Browse files

rpc: Add -EPERM processing for xs_udp_send_request() 



If an iptables drop rule is added for an nfs server, the client can end up in
a softlockup. Because of the way that xs_sendpages() is structured, the -EPERM
is ignored since the prior bits of the packet may have been successfully queued
and thus xs_sendpages() returns a non-zero value. Then, xs_udp_send_request()
thinks that because some bits were queued it should return -EAGAIN. We then try
the request again and again, resulting in cpu spinning. Reproducer:

1) open a file on the nfs server '/nfs/foo' (mounted using udp)
2) iptables -A OUTPUT -d <nfs server ip> -j DROP
3) write to /nfs/foo
4) close /nfs/foo
5) iptables -D OUTPUT -d <nfs server ip> -j DROP

The softlockup occurs in step 4 above.

The previous patch, allows xs_sendpages() to return both a sent count and
any error values that may have occurred. Thus, if we get an -EPERM, return
that to the higher level code.

With this patch in place we can successfully abort the above sequence and
avoid the softlockup.

I also tried the above test case on an nfs mount on tcp and although the system
does not softlockup, I still ended up with the 'hung_task' firing after 120
seconds, due to the i/o being stuck. The tcp case appears a bit harder to fix,
since -EPERM appears to get ignored much lower down in the stack and does not
propogate up to xs_sendpages(). This case is not quite as insidious as the
softlockup and it is not addressed here.

Reported-by: default avatarYigong Lou <ylou@akamai.com>
Signed-off-by: default avatarJason Baron <jbaron@akamai.com>
Signed-off-by: default avatarTrond Myklebust <trond.myklebust@primarydata.com>
parent f279cd00

net/sunrpc/clnt.c

+2 −0
Original line number Original line Diff line number Diff line
@@ -1913,6 +1913,7 @@ call_transmit_status(struct rpc_task *task) 
	case -EHOSTDOWN:

	case -EHOSTDOWN:

	case -EHOSTUNREACH:

	case -EHOSTUNREACH:

	case -ENETUNREACH:

	case -ENETUNREACH:

	case -EPERM:

		if (RPC_IS_SOFTCONN(task)) {

		if (RPC_IS_SOFTCONN(task)) {

			xprt_end_transmit(task);

			xprt_end_transmit(task);

			rpc_exit(task, task->tk_status);

			rpc_exit(task, task->tk_status);
@@ -2018,6 +2019,7 @@ call_status(struct rpc_task *task) 
	case -EHOSTDOWN:

	case -EHOSTDOWN:

	case -EHOSTUNREACH:

	case -EHOSTUNREACH:

	case -ENETUNREACH:

	case -ENETUNREACH:

	case -EPERM:

		if (RPC_IS_SOFTCONN(task)) {

		if (RPC_IS_SOFTCONN(task)) {

			rpc_exit(task, status);

			rpc_exit(task, status);

			break;

			break;

net/sunrpc/xprtsock.c

+6 −0
Original line number Original line Diff line number Diff line
@@ -644,6 +644,10 @@ static int xs_udp_send_request(struct rpc_task *task) 
	dprintk("RPC:       xs_udp_send_request(%u) = %d\n",

	dprintk("RPC:       xs_udp_send_request(%u) = %d\n",

			xdr->len - req->rq_bytes_sent, status);

			xdr->len - req->rq_bytes_sent, status);





	/* firewall is blocking us, don't return -EAGAIN or we end up looping */

	if (status == -EPERM)

		goto process_status;



	if (sent > 0 || status == 0) {

	if (sent > 0 || status == 0) {

		req->rq_xmit_bytes_sent += sent;

		req->rq_xmit_bytes_sent += sent;

		if (sent >= req->rq_slen)

		if (sent >= req->rq_slen)
@@ -652,6 +656,7 @@ static int xs_udp_send_request(struct rpc_task *task) 
		status = -EAGAIN;

		status = -EAGAIN;

	}

	}





process_status:

	switch (status) {

	switch (status) {

	case -ENOTSOCK:

	case -ENOTSOCK:

		status = -ENOTCONN;

		status = -ENOTCONN;
@@ -667,6 +672,7 @@ static int xs_udp_send_request(struct rpc_task *task) 
	case -ENOBUFS:

	case -ENOBUFS:

	case -EPIPE:

	case -EPIPE:

	case -ECONNREFUSED:

	case -ECONNREFUSED:

	case -EPERM:

		/* When the server has died, an ICMP port unreachable message

		/* When the server has died, an ICMP port unreachable message

		 * prompts ECONNREFUSED. */

		 * prompts ECONNREFUSED. */

		clear_bit(SOCK_ASYNC_NOSPACE, &transport->sock->flags);

		clear_bit(SOCK_ASYNC_NOSPACE, &transport->sock->flags);