X-Git-Url: https://git.decadent.org.uk/gitweb/?p=nfs-utils.git;a=blobdiff_plain;f=utils%2Fmountd%2Fcache.c;h=d2ae4563d31a1ebe855e34053ac99cda6c56cc09;hp=a14f4f2987369fe8ae299e99b3a6d118c1cf42ef;hb=014e00dfaea0efc92150e2aedc5ca43aa337545e;hpb=5fb04a376e6d5ba940e66507e4a615f4e94116e6

diff --git a/utils/mountd/cache.c b/utils/mountd/cache.c
index a14f4f2..d2ae456 100644
--- a/utils/mountd/cache.c
+++ b/utils/mountd/cache.c
@@ -14,6 +14,7 @@
 #include <sys/types.h>
 #include <sys/select.h>
 #include <sys/stat.h>
+#include <sys/vfs.h>
 #include <time.h>
 #include <netinet/in.h>
 #include <arpa/inet.h>
@@ -23,15 +24,24 @@
 #include <ctype.h>
 #include <pwd.h>
 #include <grp.h>
+#include <mntent.h>
 #include "misc.h"
 #include "nfslib.h"
 #include "exportfs.h"
 #include "mountd.h"
 #include "xmalloc.h"
 #include "fsloc.h"
+#include "pseudoflavors.h"
 
+#ifdef USE_BLKID
 #include "blkid/blkid.h"
+#endif
 
+/*
+ * Invoked by RPC service loop
+ */
+void	cache_set_fds(fd_set *fdset);
+int	cache_process_req(fd_set *readfds);
 
 enum nfsd_fsid {
 	FSID_DEV = 0,
@@ -52,13 +62,15 @@ enum nfsd_fsid {
  * Record is terminated with newline.
  *
  */
-int cache_export_ent(char *domain, struct exportent *exp);
+static int cache_export_ent(char *domain, struct exportent *exp, char *p);
 
+#define INITIAL_MANAGED_GROUPS 100
 
 char *lbuf  = NULL;
 int lbuflen = 0;
+extern int use_ipaddr;
 
-void auth_unix_ip(FILE *f)
+static void auth_unix_ip(FILE *f)
 {
 	/* requests are
 	 *  class IP-ADDR
@@ -69,60 +81,82 @@ void auth_unix_ip(FILE *f)
 	 */
 	char *cp;
 	char class[20];
-	char ipaddr[20];
-	char *client;
-	struct in_addr addr;
+	char ipaddr[INET6_ADDRSTRLEN];
+	char *client = NULL;
+	struct addrinfo *tmp = NULL;
+	struct addrinfo *ai = NULL;
 	if (readline(fileno(f), &lbuf, &lbuflen) != 1)
 		return;
 
+	xlog(D_CALL, "auth_unix_ip: inbuf '%s'", lbuf);
+
 	cp = lbuf;
 
 	if (qword_get(&cp, class, 20) <= 0 ||
 	    strcmp(class, "nfsd") != 0)
 		return;
 
-	if (qword_get(&cp, ipaddr, 20) <= 0)
+	if (qword_get(&cp, ipaddr, sizeof(ipaddr)) <= 0)
 		return;
 
-	if (inet_aton(ipaddr, &addr)==0)
+	tmp = host_pton(ipaddr);
+	if (tmp == NULL)
 		return;
 
 	auth_reload();
 
 	/* addr is a valid, interesting address, find the domain name... */
-	client = client_compose(addr);
+	if (!use_ipaddr) {
+		ai = client_resolve(tmp->ai_addr);
+		client = client_compose(ai);
+		freeaddrinfo(ai);
+	}
+	freeaddrinfo(tmp);
 
-	
 	qword_print(f, "nfsd");
 	qword_print(f, ipaddr);
-	qword_printint(f, time(0)+30*60);
-	if (client)
+	qword_printuint(f, time(0) + DEFAULT_TTL);
+	if (use_ipaddr)
+		qword_print(f, ipaddr);
+	else if (client)
 		qword_print(f, *client?client:"DEFAULT");
 	qword_eol(f);
+	xlog(D_CALL, "auth_unix_ip: client %p '%s'", client, client?client: "DEFAULT");
 
-	if (client) free(client);
-	
+	free(client);
 }
 
-void auth_unix_gid(FILE *f)
+static void auth_unix_gid(FILE *f)
 {
 	/* Request are
 	 *  uid
 	 * reply is
 	 *  uid expiry count list of group ids
 	 */
-	int uid;
+	uid_t uid;
 	struct passwd *pw;
-	gid_t glist[100], *groups = glist;
-	int ngroups = 100;
+	static gid_t *groups = NULL;
+	static int groups_len = 0;
+	gid_t *more_groups;
+	int ngroups;
 	int rv, i;
 	char *cp;
 
+	if (groups_len == 0) {
+		groups = malloc(sizeof(gid_t) * INITIAL_MANAGED_GROUPS);
+		if (!groups)
+			return;
+
+		groups_len = INITIAL_MANAGED_GROUPS;
+	}
+
+	ngroups = groups_len;
+
 	if (readline(fileno(f), &lbuf, &lbuflen) != 1)
 		return;
 
 	cp = lbuf;
-	if (qword_get_int(&cp, &uid) != 0)
+	if (qword_get_uint(&cp, &uid) != 0)
 		return;
 
 	pw = getpwuid(uid);
@@ -130,69 +164,81 @@ void auth_unix_gid(FILE *f)
 		rv = -1;
 	else {
 		rv = getgrouplist(pw->pw_name, pw->pw_gid, groups, &ngroups);
-		if (rv == -1 && ngroups >= 100) {
-			groups = malloc(sizeof(gid_t)*ngroups);
-			if (!groups)
+		if (rv == -1 && ngroups >= groups_len) {
+			more_groups = realloc(groups, sizeof(gid_t)*ngroups);
+			if (!more_groups)
 				rv = -1;
-			else
+			else {
+				groups = more_groups;
+				groups_len = ngroups;
 				rv = getgrouplist(pw->pw_name, pw->pw_gid,
 						  groups, &ngroups);
+			}
 		}
 	}
-	qword_printint(f, uid);
-	qword_printint(f, time(0)+30*60);
+	qword_printuint(f, uid);
+	qword_printuint(f, time(0) + DEFAULT_TTL);
 	if (rv >= 0) {
-		qword_printint(f, ngroups);
+		qword_printuint(f, ngroups);
 		for (i=0; i<ngroups; i++)
-			qword_printint(f, groups[i]);
-	}
+			qword_printuint(f, groups[i]);
+	} else
+		qword_printuint(f, 0);
 	qword_eol(f);
-	if (groups != glist)
-		free(groups);
 }
 
-int get_uuid(char *path, char *uuid, int uuidlen, char *u)
+#if USE_BLKID
+static const char *get_uuid_blkdev(char *path)
+{
+	/* We set *safe if we know that we need the
+	 * fsid from statfs too.
+	 */
+	static blkid_cache cache = NULL;
+	struct stat stb;
+	char *devname;
+	blkid_tag_iterate iter;
+	blkid_dev dev;
+	const char *type;
+	const char *val, *uuid = NULL;
+
+	if (cache == NULL)
+		blkid_get_cache(&cache, NULL);
+
+	if (stat(path, &stb) != 0)
+		return NULL;
+	devname = blkid_devno_to_devname(stb.st_dev);
+	if (!devname)
+		return NULL;
+	dev = blkid_get_dev(cache, devname, BLKID_DEV_NORMAL);
+	free(devname);
+	if (!dev)
+		return NULL;
+	iter = blkid_tag_iterate_begin(dev);
+	if (!iter)
+		return NULL;
+	while (blkid_tag_next(iter, &type, &val) == 0) {
+		if (strcmp(type, "UUID") == 0)
+			uuid = val;
+		if (strcmp(type, "TYPE") == 0 &&
+		    strcmp(val, "btrfs") == 0) {
+			uuid = NULL;
+			break;
+		}
+	}
+	blkid_tag_iterate_end(iter);
+	return uuid;
+}
+#else
+#define get_uuid_blkdev(path) (NULL)
+#endif
+
+static int get_uuid(const char *val, int uuidlen, char *u)
 {
 	/* extract hex digits from uuidstr and compose a uuid
 	 * of the given length (max 16), xoring bytes to make
-	 * a smaller uuid.  Then compare with uuid
+	 * a smaller uuid.
 	 */
 	int i = 0;
-	const char *val;
-
-	if (path) {
-		static blkid_cache cache = NULL;
-		struct stat stb;
-		char *devname;
-		blkid_tag_iterate iter;
-		blkid_dev dev;
-		const char *type;
-		if (cache == NULL)
-			blkid_get_cache(&cache, NULL);
-
-		blkid_probe_all_new(cache);
-
-		if (stat(path, &stb) != 0)
-			return 0;
-		devname = blkid_devno_to_devname(stb.st_dev);
-		if (!devname)
-			return 0;
-		dev = blkid_get_dev(cache, devname, BLKID_DEV_NORMAL);
-		free(devname);
-		if (!dev)
-			return 0;
-		iter = blkid_tag_iterate_begin(dev);
-		if (!iter)
-			return 0;
-		while (blkid_tag_next(iter, &type, &val) == 0)
-			if (strcmp(type, "UUID") == 0)
-				break;
-		blkid_tag_iterate_end(iter);
-		if (!type)
-			return 0;
-	} else {
-		val = uuid;
-	}
 	
 	memset(u, 0, uuidlen);
 	for ( ; *val ; val++) {
@@ -215,61 +261,129 @@ int get_uuid(char *path, char *uuid, int uuidlen, char *u)
 	}
 	return 1;
 }
-	
 
-void nfsd_fh(FILE *f)
+static int uuid_by_path(char *path, int type, int uuidlen, char *uuid)
 {
-	/* request are:
-	 *  domain fsidtype fsid
-	 * interpret fsid, find export point and options, and write:
-	 *  domain fsidtype fsid expiry path
+	/* get a uuid for the filesystem found at 'path'.
+	 * There are several possible ways of generating the
+	 * uuids (types).
+	 * Type 0 is used for new filehandles, while other types
+	 * may be used to interpret old filehandle - to ensure smooth
+	 * forward migration.
+	 * We return 1 if a uuid was found (and it might be worth 
+	 * trying the next type) or 0 if no more uuid types can be
+	 * extracted.
 	 */
-	char *cp;
-	char *dom;
-	int fsidtype;
-	int fsidlen;
-	unsigned int dev, major=0, minor=0;
-	unsigned int inode=0;
-	unsigned long long inode64;
-	unsigned int fsidnum=0;
-	char fsid[32];
-	struct exportent *found = NULL;
-	nfs_export *exp;
-	int i;
-	int dev_missing = 0;
-	int uuidlen = 0;
-	char *fhuuid = NULL;
 
-	if (readline(fileno(f), &lbuf, &lbuflen) != 1)
-		return;
+	/* Possible sources of uuid are
+	 * - blkid uuid
+	 * - statfs64 uuid
+	 *
+	 * On some filesystems (e.g. vfat) the statfs64 uuid is simply an
+	 * encoding of the device that the filesystem is mounted from, so
+	 * it we be very bad to use that (as device numbers change).  blkid
+	 * must be preferred.
+	 * On other filesystems (e.g. btrfs) the statfs64 uuid contains
+	 * important info that the blkid uuid cannot contain:  This happens
+	 * when multiple subvolumes are exported (they have the same
+	 * blkid uuid but different statfs64 uuids).
+	 * We rely on get_uuid_blkdev *knowing* which is which and not returning
+	 * a uuid for filesystems where the statfs64 uuid is better.
+	 *
+	 */
+	struct statfs64 st;
+	char fsid_val[17];
+	const char *blkid_val;
+	const char *val;
 
-	cp = lbuf;
+	blkid_val = get_uuid_blkdev(path);
 
-	dom = malloc(strlen(cp));
-	if (dom == NULL)
-		return;
-	if (qword_get(&cp, dom, strlen(cp)) <= 0)
-		goto out;
-	if (qword_get_int(&cp, &fsidtype) != 0)
-		goto out;
-	if (fsidtype < 0 || fsidtype > 7)
-		goto out; /* unknown type */
-	if ((fsidlen = qword_get(&cp, fsid, 32)) <= 0)
-		goto out;
+	if (statfs64(path, &st) == 0 &&
+	    (st.f_fsid.__val[0] || st.f_fsid.__val[1]))
+		snprintf(fsid_val, 17, "%08x%08x",
+			 st.f_fsid.__val[0], st.f_fsid.__val[1]);
+	else
+		fsid_val[0] = 0;
+
+	if (blkid_val && (type--) == 0)
+		val = blkid_val;
+	else if (fsid_val[0] && (type--) == 0)
+		val = fsid_val;
+	else
+		return 0;
+
+	get_uuid(val, uuidlen, uuid);
+	return 1;
+}
+
+/* Iterate through /etc/mtab, finding mountpoints
+ * at or below a given path
+ */
+static char *next_mnt(void **v, char *p)
+{
+	FILE *f;
+	struct mntent *me;
+	int l = strlen(p);
+	if (*v == NULL) {
+		f = setmntent("/etc/mtab", "r");
+		*v = f;
+	} else
+		f = *v;
+	while ((me = getmntent(f)) != NULL &&
+	       (strncmp(me->mnt_dir, p, l) != 0 ||
+		me->mnt_dir[l] != '/'))
+		;
+	if (me == NULL) {
+		endmntent(f);
+		*v = NULL;
+		return NULL;
+	}
+	return me->mnt_dir;
+}
+
+/* True iff e1 is a child of e2 and e2 has crossmnt set: */
+static bool subexport(struct exportent *e1, struct exportent *e2)
+{
+	char *p1 = e1->e_path, *p2 = e2->e_path;
+	int l2 = strlen(p2);
+
+	return e2->e_flags & NFSEXP_CROSSMOUNT
+	       && strncmp(p1, p2, l2) == 0
+	       && p1[l2] == '/';
+}
+
+struct parsed_fsid {
+	int fsidtype;
+	/* We could use a union for this, but it would be more
+	 * complicated; why bother? */
+	unsigned int inode;
+	unsigned int minor;
+	unsigned int major;
+	unsigned int fsidnum;
+	int uuidlen;
+	char *fhuuid;
+};
+
+int parse_fsid(int fsidtype, int fsidlen, char *fsid, struct parsed_fsid *parsed)
+{
+	unsigned int dev;
+	unsigned long long inode64;
+
+	parsed->fsidtype = fsidtype;
 	switch(fsidtype) {
 	case FSID_DEV: /* 4 bytes: 2 major, 2 minor, 4 inode */
 		if (fsidlen != 8)
-			goto out;
+			return -1;
 		memcpy(&dev, fsid, 4);
-		memcpy(&inode, fsid+4, 4);
-		major = ntohl(dev)>>16;
-		minor = ntohl(dev) & 0xFFFF;
+		memcpy(&parsed->inode, fsid+4, 4);
+		parsed->major = ntohl(dev)>>16;
+		parsed->minor = ntohl(dev) & 0xFFFF;
 		break;
 
 	case FSID_NUM: /* 4 bytes - fsid */
 		if (fsidlen != 4)
-			goto out;
-		memcpy(&fsidnum, fsid, 4);
+			return -1;
+		memcpy(&parsed->fsidnum, fsid, 4);
 		break;
 
 	case FSID_MAJOR_MINOR: /* 12 bytes: 4 major, 4 minor, 4 inode 
@@ -277,10 +391,12 @@ void nfsd_fh(FILE *f)
 		 * an historical accident
 		 */
 		if (fsidlen != 12)
-			goto out;
-		memcpy(&dev, fsid, 4); major = ntohl(dev);
-		memcpy(&dev, fsid+4, 4); minor = ntohl(dev);
-		memcpy(&inode, fsid+8, 4);
+			return -1;
+		memcpy(&dev, fsid, 4);
+		parsed->major = ntohl(dev);
+		memcpy(&dev, fsid+4, 4);
+		parsed->minor = ntohl(dev);
+		memcpy(&parsed->inode, fsid+8, 4);
 		break;
 
 	case FSID_ENCODE_DEV: /* 8 bytes: 4 byte packed device number, 4 inode */
@@ -288,102 +404,203 @@ void nfsd_fh(FILE *f)
 		 * no-one outside this host has any business interpreting it
 		 */
 		if (fsidlen != 8)
-			goto out;
+			return -1;
 		memcpy(&dev, fsid, 4);
-		memcpy(&inode, fsid+4, 4);
-		major = (dev & 0xfff00) >> 8;
-		minor = (dev & 0xff) | ((dev >> 12) & 0xfff00);
+		memcpy(&parsed->inode, fsid+4, 4);
+		parsed->major = (dev & 0xfff00) >> 8;
+		parsed->minor = (dev & 0xff) | ((dev >> 12) & 0xfff00);
 		break;
 
 	case FSID_UUID4_INUM: /* 4 byte inode number and 4 byte uuid */
 		if (fsidlen != 8)
-			goto out;
-		memcpy(&inode, fsid, 4);
-		uuidlen = 4;
-		fhuuid = fsid+4;
+			return -1;
+		memcpy(&parsed->inode, fsid, 4);
+		parsed->uuidlen = 4;
+		parsed->fhuuid = fsid+4;
 		break;
 	case FSID_UUID8: /* 8 byte uuid */
 		if (fsidlen != 8)
-			goto out;
-		uuidlen = 8;
-		fhuuid = fsid;
+			return -1;
+		parsed->uuidlen = 8;
+		parsed->fhuuid = fsid;
 		break;
 	case FSID_UUID16: /* 16 byte uuid */
 		if (fsidlen != 16)
-			goto out;
-		uuidlen = 16;
-		fhuuid = fsid;
+			return -1;
+		parsed->uuidlen = 16;
+		parsed->fhuuid = fsid;
 		break;
 	case FSID_UUID16_INUM: /* 8 byte inode number and 16 byte uuid */
 		if (fsidlen != 24)
-			goto out;
+			return -1;
 		memcpy(&inode64, fsid, 8);
-		inode = inode64;
-		uuidlen = 16;
-		fhuuid = fsid+8;
+		parsed->inode = inode64;
+		parsed->uuidlen = 16;
+		parsed->fhuuid = fsid+8;
 		break;
 	}
+	return 0;
+}
+
+static bool match_fsid(struct parsed_fsid *parsed, nfs_export *exp, char *path)
+{
+	struct stat stb;
+	int type;
+	char u[16];
+
+	if (stat(path, &stb) != 0)
+		return false;
+	if (!S_ISDIR(stb.st_mode) && !S_ISREG(stb.st_mode))
+		return false;
+
+	switch (parsed->fsidtype) {
+	case FSID_DEV:
+	case FSID_MAJOR_MINOR:
+	case FSID_ENCODE_DEV:
+		if (stb.st_ino != parsed->inode)
+			return false;
+		if (parsed->major != major(stb.st_dev) ||
+		    parsed->minor != minor(stb.st_dev))
+			return false;
+		return true;
+	case FSID_NUM:
+		if (((exp->m_export.e_flags & NFSEXP_FSID) == 0 ||
+		     exp->m_export.e_fsid != parsed->fsidnum))
+			return false;
+		return true;
+	case FSID_UUID4_INUM:
+	case FSID_UUID16_INUM:
+		if (stb.st_ino != parsed->inode)
+			return false;
+		goto check_uuid;
+	case FSID_UUID8:
+	case FSID_UUID16:
+		if (!is_mountpoint(path))
+			return false;
+	check_uuid:
+		if (exp->m_export.e_uuid)
+			get_uuid(exp->m_export.e_uuid, parsed->uuidlen, u);
+		else
+			for (type = 0;
+			     uuid_by_path(path, type, parsed->uuidlen, u);
+			     type++)
+				if (memcmp(u, parsed->fhuuid, parsed->uuidlen) == 0)
+					return true;
+
+		if (memcmp(u, parsed->fhuuid, parsed->uuidlen) != 0)
+			return false;
+		return true;
+	}
+	/* Well, unreachable, actually: */
+	return false;
+}
+
+static void nfsd_fh(FILE *f)
+{
+	/* request are:
+	 *  domain fsidtype fsid
+	 * interpret fsid, find export point and options, and write:
+	 *  domain fsidtype fsid expiry path
+	 */
+	char *cp;
+	char *dom;
+	int fsidtype;
+	int fsidlen;
+	char fsid[32];
+	struct parsed_fsid parsed;
+	struct exportent *found = NULL;
+	struct addrinfo *ai = NULL;
+	char *found_path = NULL;
+	nfs_export *exp;
+	int i;
+	int dev_missing = 0;
+
+	if (readline(fileno(f), &lbuf, &lbuflen) != 1)
+		return;
+
+	xlog(D_CALL, "nfsd_fh: inbuf '%s'", lbuf);
+
+	cp = lbuf;
+
+	dom = malloc(strlen(cp));
+	if (dom == NULL)
+		return;
+	if (qword_get(&cp, dom, strlen(cp)) <= 0)
+		goto out;
+	if (qword_get_int(&cp, &fsidtype) != 0)
+		goto out;
+	if (fsidtype < 0 || fsidtype > 7)
+		goto out; /* unknown type */
+	if ((fsidlen = qword_get(&cp, fsid, 32)) <= 0)
+		goto out;
+	if (parse_fsid(fsidtype, fsidlen, fsid, &parsed))
+		goto out;
 
 	auth_reload();
 
 	/* Now determine export point for this fsid/domain */
 	for (i=0 ; i < MCL_MAXTYPES; i++) {
-		for (exp = exportlist[i]; exp; exp = exp->m_next) {
-			struct stat stb;
-			char u[16];			
+		nfs_export *next_exp;
+		for (exp = exportlist[i].p_head; exp; exp = next_exp) {
+			char *path;
+
+			if (exp->m_export.e_flags & NFSEXP_CROSSMOUNT) {
+				static nfs_export *prev = NULL;
+				static void *mnt = NULL;
+				
+				if (prev == exp) {
+					/* try a submount */
+					path = next_mnt(&mnt, exp->m_export.e_path);
+					if (!path) {
+						next_exp = exp->m_next;
+						prev = NULL;
+						continue;
+					}
+					next_exp = exp;
+				} else {
+					prev = exp;
+					mnt = NULL;
+					path = exp->m_export.e_path;
+					next_exp = exp;
+				}
+			} else {
+				path = exp->m_export.e_path;
+				next_exp = exp->m_next;
+			}
 
-			if (!client_member(dom, exp->m_client->m_hostname))
+			if (!use_ipaddr && !client_member(dom, exp->m_client->m_hostname))
 				continue;
 			if (exp->m_export.e_mountpoint &&
 			    !is_mountpoint(exp->m_export.e_mountpoint[0]?
 					   exp->m_export.e_mountpoint:
 					   exp->m_export.e_path))
 				dev_missing ++;
-			if (stat(exp->m_export.e_path, &stb) != 0)
-				continue;
-			switch(fsidtype){
-			case FSID_DEV:
-			case FSID_MAJOR_MINOR:
-			case FSID_ENCODE_DEV:
-				if (stb.st_ino != inode)
-					continue;
-				if (major != major(stb.st_dev) ||
-				    minor != minor(stb.st_dev))
-					continue;
-				break;
-			case FSID_NUM:
-				if (((exp->m_export.e_flags & NFSEXP_FSID) == 0 ||
-				     exp->m_export.e_fsid != fsidnum))
-					continue;
-				break;
-			case FSID_UUID4_INUM:
-			case FSID_UUID16_INUM:
-				if (stb.st_ino != inode)
-					continue;
-				goto check_uuid;
-			case FSID_UUID8:
-			case FSID_UUID16:
-				if (!is_mountpoint(exp->m_export.e_path))
-					continue;
-			check_uuid:
-				if (exp->m_export.e_uuid)
-					get_uuid(NULL, exp->m_export.e_uuid,
-						 uuidlen, u);
-				else if (get_uuid(exp->m_export.e_path, NULL,
-						  uuidlen, u) == 0)
-					continue;
 
-				if (memcmp(u, fhuuid, uuidlen) != 0)
+			if (!match_fsid(&parsed, exp, path))
+				continue;
+			if (use_ipaddr) {
+				if (ai == NULL) {
+					struct addrinfo *tmp;
+					tmp = host_pton(dom);
+					if (tmp == NULL)
+						goto out;
+					ai = client_resolve(tmp->ai_addr);
+					freeaddrinfo(tmp);
+				}
+				if (!client_check(exp->m_client, ai))
 					continue;
-				break;
 			}
-			/* It's a match !! */
-			if (!found)
+			if (!found || subexport(&exp->m_export, found)) {
 				found = &exp->m_export;
-			else if (strcmp(found->e_path, exp->m_export.e_path)!= 0)
+				free(found_path);
+				found_path = strdup(path);
+				if (found_path == NULL)
+					goto out;
+			} else if (strcmp(found->e_path, exp->m_export.e_path)
+				   && !subexport(found, &exp->m_export))
 			{
 				xlog(L_WARNING, "%s and %s have same filehandle for %s, using first",
-				     found->e_path, exp->m_export.e_path, dom);
+				     found_path, path, dom);
 			}
 		}
 	}
@@ -408,28 +625,41 @@ void nfsd_fh(FILE *f)
 	}
 
 	if (found)
-		cache_export_ent(dom, found);
+		if (cache_export_ent(dom, found, found_path) < 0)
+			found = 0;
 
 	qword_print(f, dom);
 	qword_printint(f, fsidtype);
 	qword_printhex(f, fsid, fsidlen);
-	qword_printint(f, time(0)+30*60);
+	/* The fsid -> path lookup can be quite expensive as it
+	 * potentially stats and reads lots of devices, and some of those
+	 * might have spun-down.  The Answer is not likely to
+	 * change underneath us, and an 'exportfs -f' can always
+	 * remove this from the kernel, so use a really log
+	 * timeout.  Maybe this should be configurable on the command
+	 * line.
+	 */
+	qword_printint(f, 0x7fffffff);
 	if (found)
-		qword_print(f, found->e_path);
+		qword_print(f, found_path);
 	qword_eol(f);
  out:
+	if (found_path)
+		free(found_path);
+	freeaddrinfo(ai);
 	free(dom);
+	xlog(D_CALL, "nfsd_fh: found %p path %s", found, found ? found->e_path : NULL);
 	return;		
 }
 
-static void write_fsloc(FILE *f, struct exportent *ep, char *path)
+static void write_fsloc(FILE *f, struct exportent *ep)
 {
 	struct servers *servers;
 
 	if (ep->e_fslocmethod == FSLOC_NONE)
 		return;
 
-	servers = replicas_lookup(ep->e_fslocmethod, ep->e_fslocdata, path);
+	servers = replicas_lookup(ep->e_fslocmethod, ep->e_fslocdata);
 	if (!servers)
 		return;
 	qword_print(f, "fsloc");
@@ -445,32 +675,134 @@ static void write_fsloc(FILE *f, struct exportent *ep, char *path)
 	release_replicas(servers);
 }
 
+static void write_secinfo(FILE *f, struct exportent *ep, int flag_mask)
+{
+	struct sec_entry *p;
+
+	for (p = ep->e_secinfo; p->flav; p++)
+		; /* Do nothing */
+	if (p == ep->e_secinfo) {
+		/* There was no sec= option */
+		return;
+	}
+	qword_print(f, "secinfo");
+	qword_printint(f, p - ep->e_secinfo);
+	for (p = ep->e_secinfo; p->flav; p++) {
+		qword_printint(f, p->flav->fnum);
+		qword_printint(f, p->flags & flag_mask);
+	}
+
+}
+
 static int dump_to_cache(FILE *f, char *domain, char *path, struct exportent *exp)
 {
 	qword_print(f, domain);
 	qword_print(f, path);
-	qword_printint(f, time(0)+30*60);
 	if (exp) {
-		qword_printint(f, exp->e_flags);
+		int different_fs = strcmp(path, exp->e_path) != 0;
+		int flag_mask = different_fs ? ~NFSEXP_FSID : ~0;
+
+		qword_printuint(f, time(0) + exp->e_ttl);
+		qword_printint(f, exp->e_flags & flag_mask);
 		qword_printint(f, exp->e_anonuid);
 		qword_printint(f, exp->e_anongid);
 		qword_printint(f, exp->e_fsid);
- 		if (exp->e_uuid == NULL) {
+		write_fsloc(f, exp);
+		write_secinfo(f, exp, flag_mask);
+ 		if (exp->e_uuid == NULL || different_fs) {
  			char u[16];
- 			if (get_uuid(exp->e_path, NULL, 16, u)) {
+ 			if (uuid_by_path(path, 0, 16, u)) {
  				qword_print(f, "uuid");
  				qword_printhex(f, u, 16);
  			}
- 		} else if (exp->e_uuid) {
+ 		} else {
+ 			char u[16];
+ 			get_uuid(exp->e_uuid, 16, u);
  			qword_print(f, "uuid");
- 			qword_printhex(f, exp->e_uuid, 16);
+ 			qword_printhex(f, u, 16);
  		}
-		write_fsloc(f, &exp, path);
-	}
+	} else
+		qword_printuint(f, time(0) + DEFAULT_TTL);
 	return qword_eol(f);
 }
 
-void nfsd_export(FILE *f)
+static int is_subdirectory(char *child, char *parent)
+{
+	int l = strlen(parent);
+
+	return strcmp(child, parent) == 0
+		|| (strncmp(child, parent, l) == 0 && child[l] == '/');
+}
+
+static int path_matches(nfs_export *exp, char *path)
+{
+	if (exp->m_export.e_flags & NFSEXP_CROSSMOUNT)
+		return is_subdirectory(path, exp->m_export.e_path);
+	return strcmp(path, exp->m_export.e_path) == 0;
+}
+
+static int
+client_matches(nfs_export *exp, char *dom, struct addrinfo *ai)
+{
+	if (use_ipaddr)
+		return client_check(exp->m_client, ai);
+	return client_member(dom, exp->m_client->m_hostname);
+}
+
+static int
+export_matches(nfs_export *exp, char *dom, char *path, struct addrinfo *ai)
+{
+	return path_matches(exp, path) && client_matches(exp, dom, ai);
+}
+
+static nfs_export *
+lookup_export(char *dom, char *path, struct addrinfo *ai)
+{
+	nfs_export *exp;
+	nfs_export *found = NULL;
+	int found_type = 0;
+	int i;
+
+	for (i=0 ; i < MCL_MAXTYPES; i++) {
+		for (exp = exportlist[i].p_head; exp; exp = exp->m_next) {
+			if (!export_matches(exp, dom, path, ai))
+				continue;
+			if (!found) {
+				found = exp;
+				found_type = i;
+				continue;
+			}
+
+			/* Always prefer non-V4ROOT mounts */
+			if (found->m_export.e_flags & NFSEXP_V4ROOT)
+				continue;
+
+			/* If one is a CROSSMOUNT, then prefer the longest path */
+			if (((found->m_export.e_flags & NFSEXP_CROSSMOUNT) ||
+			     (exp->m_export.e_flags & NFSEXP_CROSSMOUNT)) &&
+			    strlen(found->m_export.e_path) !=
+			    strlen(exp->m_export.e_path)) {
+
+				if (strlen(exp->m_export.e_path) >
+				    strlen(found->m_export.e_path)) {
+					found = exp;
+					found_type = i;
+				}
+				continue;
+
+			} else if (found_type == i && found->m_warned == 0) {
+				xlog(L_WARNING, "%s exported to both %s and %s, "
+				     "arbitrarily choosing options from first",
+				     path, found->m_client->m_hostname, exp->m_client->m_hostname,
+				     dom);
+				found->m_warned = 1;
+			}
+		}
+	}
+	return found;
+}
+
+static void nfsd_export(FILE *f)
 {
 	/* requests are:
 	 *  domain path
@@ -479,14 +811,15 @@ void nfsd_export(FILE *f)
 	 */
 
 	char *cp;
-	int i;
 	char *dom, *path;
-	nfs_export *exp, *found = NULL;
-
+	nfs_export *found = NULL;
+	struct addrinfo *ai = NULL;
 
 	if (readline(fileno(f), &lbuf, &lbuflen) != 1)
 		return;
 
+	xlog(D_CALL, "nfsd_export: inbuf '%s'", lbuf);
+
 	cp = lbuf;
 	dom = malloc(strlen(cp));
 	path = malloc(strlen(cp));
@@ -501,32 +834,33 @@ void nfsd_export(FILE *f)
 
 	auth_reload();
 
-	/* now find flags for this export point in this domain */
-	for (i=0 ; i < MCL_MAXTYPES; i++) {
-		for (exp = exportlist[i]; exp; exp = exp->m_next) {
-			if (!client_member(dom, exp->m_client->m_hostname))
-				continue;
-			if (strcmp(path, exp->m_export.e_path))
-				continue;
-			if (!found)
-				found = exp;
-			else {
-				xlog(L_WARNING, "%s exported to both %s and %s in %s",
-				     path, exp->m_client->m_hostname, found->m_client->m_hostname,
-				     dom);
-			}
-		}
+	if (use_ipaddr) {
+		struct addrinfo *tmp;
+		tmp = host_pton(dom);
+		if (tmp == NULL)
+			goto out;
+		ai = client_resolve(tmp->ai_addr);
+		freeaddrinfo(tmp);
+			goto out;
 	}
 
+	found = lookup_export(dom, path, ai);
+
 	if (found) {
-		dump_to_cache(f, dom, path, &found->m_export);
-		mountlist_add(dom, path);
+		if (dump_to_cache(f, dom, path, &found->m_export) < 0) {
+			xlog(L_WARNING,
+			     "Cannot export %s, possibly unsupported filesystem"
+			     " or fsid= required", path);
+			dump_to_cache(f, dom, path, NULL);
+		}
 	} else {
 		dump_to_cache(f, dom, path, NULL);
 	}
  out:
+	xlog(D_CALL, "nfsd_export: found %p path %s", found, path ? path : NULL);
 	if (dom) free(dom);
 	if (path) free(path);
+	freeaddrinfo(ai);
 }
 
 
@@ -534,27 +868,41 @@ struct {
 	char *cache_name;
 	void (*cache_handle)(FILE *f);
 	FILE *f;
+	char vbuf[RPC_CHAN_BUF_SIZE];
 } cachelist[] = {
-	{ "auth.unix.ip", auth_unix_ip},
-	{ "auth.unix.gid", auth_unix_gid},
-	{ "nfsd.export", nfsd_export},
-	{ "nfsd.fh", nfsd_fh},
-	{ NULL, NULL }
+	{ "auth.unix.ip", auth_unix_ip, NULL, ""},
+	{ "auth.unix.gid", auth_unix_gid, NULL, ""},
+	{ "nfsd.export", nfsd_export, NULL, ""},
+	{ "nfsd.fh", nfsd_fh, NULL, ""},
+	{ NULL, NULL, NULL, ""}
 };
 
 extern int manage_gids;
+
+/**
+ * cache_open - prepare communications channels with kernel RPC caches
+ *
+ */
 void cache_open(void) 
 {
 	int i;
 	for (i=0; cachelist[i].cache_name; i++ ) {
 		char path[100];
-		if (!manage_gids && cachelist[i].f == auth_unix_gid)
+		if (!manage_gids && cachelist[i].cache_handle == auth_unix_gid)
 			continue;
 		sprintf(path, "/proc/net/rpc/%s/channel", cachelist[i].cache_name);
 		cachelist[i].f = fopen(path, "r+");
+		if (cachelist[i].f != NULL) {
+			setvbuf(cachelist[i].f, cachelist[i].vbuf, _IOLBF, 
+				RPC_CHAN_BUF_SIZE);
+		}
 	}
 }
 
+/**
+ * cache_set_fds - prepare cache file descriptors for one iteration of the service loop
+ * @fdset: pointer to fd_set to prepare
+ */
 void cache_set_fds(fd_set *fdset)
 {
 	int i;
@@ -564,6 +912,10 @@ void cache_set_fds(fd_set *fdset)
 	}
 }
 
+/**
+ * cache_process_req - process any active cache file descriptors during service loop iteration
+ * @fdset: pointer to fd_set to examine for activity
+ */
 int cache_process_req(fd_set *readfds) 
 {
 	int i;
@@ -582,11 +934,11 @@ int cache_process_req(fd_set *readfds)
 
 /*
  * Give IP->domain and domain+path->options to kernel
- * % echo nfsd $IP  $[now+30*60] $domain > /proc/net/rpc/auth.unix.ip/channel
- * % echo $domain $path $[now+30*60] $options $anonuid $anongid $fsid > /proc/net/rpc/nfsd.export/channel
+ * % echo nfsd $IP  $[now+DEFAULT_TTL] $domain > /proc/net/rpc/auth.unix.ip/channel
+ * % echo $domain $path $[now+DEFAULT_TTL] $options $anonuid $anongid $fsid > /proc/net/rpc/nfsd.export/channel
  */
 
-int cache_export_ent(char *domain, struct exportent *exp)
+static int cache_export_ent(char *domain, struct exportent *exp, char *path)
 {
 	int err;
 	FILE *f = fopen("/proc/net/rpc/nfsd.export/channel", "w");
@@ -594,40 +946,94 @@ int cache_export_ent(char *domain, struct exportent *exp)
 		return -1;
 
 	err = dump_to_cache(f, domain, exp->e_path, exp);
+	if (err) {
+		xlog(L_WARNING,
+		     "Cannot export %s, possibly unsupported filesystem or"
+		     " fsid= required", exp->e_path);
+	}
+
+	while (err == 0 && (exp->e_flags & NFSEXP_CROSSMOUNT) && path) {
+		/* really an 'if', but we can break out of
+		 * a 'while' more easily */
+		/* Look along 'path' for other filesystems
+		 * and export them with the same options
+		 */
+		struct stat stb;
+		size_t l = strlen(exp->e_path);
+		__dev_t dev;
+
+		if (strlen(path) <= l || path[l] != '/' ||
+		    strncmp(exp->e_path, path, l) != 0)
+			break;
+		if (stat(exp->e_path, &stb) != 0)
+			break;
+		dev = stb.st_dev;
+		while(path[l] == '/') {
+			char c;
+			/* errors for submount should fail whole filesystem */
+			int err2;
+
+			l++;
+			while (path[l] != '/' && path[l])
+				l++;
+			c = path[l];
+			path[l] = 0;
+			err2 = lstat(path, &stb);
+			path[l] = c;
+			if (err2 < 0)
+				break;
+			if (stb.st_dev == dev)
+				continue;
+			dev = stb.st_dev;
+			path[l] = 0;
+			dump_to_cache(f, domain, path, exp);
+			path[l] = c;
+		}
+		break;
+	}
+
 	fclose(f);
-	mountlist_add(domain, exp->e_path);
 	return err;
 }
 
-int cache_export(nfs_export *exp)
+/**
+ * cache_export - Inform kernel of a new nfs_export
+ * @exp: target nfs_export
+ * @path: NUL-terminated C string containing export path
+ */
+int cache_export(nfs_export *exp, char *path)
 {
+	char buf[INET6_ADDRSTRLEN];
 	int err;
 	FILE *f;
 
-	if (exp->m_export.e_maptype != CLE_MAP_IDENT) {
-		xlog(L_ERROR, "%s: unsupported mapping; kernel supports only 'identity' (default)",
-		    exp->m_export.m_path);
-		return -1;
-	}
-
 	f = fopen("/proc/net/rpc/auth.unix.ip/channel", "w");
 	if (!f)
 		return -1;
 
+
 	qword_print(f, "nfsd");
-	qword_print(f, inet_ntoa(exp->m_client->m_addrlist[0]));
-	qword_printint(f, time(0)+30*60);
+	qword_print(f,
+		host_ntop(get_addrlist(exp->m_client, 0), buf, sizeof(buf)));
+	qword_printuint(f, time(0) + exp->m_export.e_ttl);
 	qword_print(f, exp->m_client->m_hostname);
 	err = qword_eol(f);
 	
 	fclose(f);
 
-	err = cache_export_ent(exp->m_client->m_hostname, &exp->m_export)
+	err = cache_export_ent(exp->m_client->m_hostname, &exp->m_export, path)
 		|| err;
 	return err;
 }
 
-/* Get a filehandle.
+/**
+ * cache_get_filehandle - given an nfs_export, get its root filehandle
+ * @exp: target nfs_export
+ * @len: length of requested file handle
+ * @p: NUL-terminated C string containing export path
+ *
+ * Returns pointer to NFS file handle of root directory of export
+ *
  * { 
  *   echo $domain $path $length 
  *   read filehandle <&0
@@ -661,4 +1067,3 @@ cache_get_filehandle(nfs_export *exp, int len, char *p)
 	fh.fh_size = qword_get(&bp, (char *)fh.fh_handle, NFS3_FHSIZE);
 	return &fh;
 }
-