Add linux events

This topic branch leverages the Solaris style FMA call points
in ZFS to create a user space visible event notification system
under Linux.  This new system is called zevent and it unifies
all previous Solaris style ereports and sysevent notifications.

Under this Linux specific scheme when a sysevent or ereport event
occurs an nvlist describing the event is created which looks almost
exactly like a Solaris ereport.  These events are queued up in the
kernel when they occur and conditionally logged to the console.
It is then up to a user space application to consume the events
and do whatever it likes with them.

To make this possible the existing /dev/zfs ABI has been extended
with two new ioctls which behave as follows.

* ZFS_IOC_EVENTS_NEXT
Get the next pending event.  The kernel will keep track of the last
event consumed by the file descriptor and provide the next one if
available.  If no new events are available the ioctl() will block
waiting for the next event.  This ioctl may also be called in a
non-blocking mode by setting zc.zc_guid = ZEVENT_NONBLOCK.  In the
non-blocking case if no events are available ENOENT will be returned.
It is possible that ESHUTDOWN will be returned if the ioctl() is
called while module unloading is in progress.  And finally ENOMEM
may occur if the provided nvlist buffer is not large enough to
contain the entire event.

* ZFS_IOC_EVENTS_CLEAR
Clear are events queued by the kernel.  The kernel will keep a fairly
large number of recent events queued, use this ioctl to clear the
in kernel list.  This will effect all user space processes consuming
events.

The zpool command has been extended to use this events ABI with the
'events' subcommand.  You may run 'zpool events -v' to output a
verbose log of all recent events.  This is very similar to the
Solaris 'fmdump -ev' command with the key difference being it also
includes what would be considered sysevents under Solaris.  You
may also run in follow mode with the '-f' option.  To clear the
in kernel event queue use the '-c' option.

$ sudo cmd/zpool/zpool events -fv
TIME                        CLASS
May 13 2010 16:31:15.777711000 ereport.fs.zfs.config.sync
        class = "ereport.fs.zfs.config.sync"
        ena = 0x40982b7897700001
        detector = (embedded nvlist)
                version = 0x0
                scheme = "zfs"
                pool = 0xed976600de75dfa6
        (end detector)

        time = 0x4bec8bc3 0x2e5aed98
        pool = "zpios"
        pool_guid = 0xed976600de75dfa6
        pool_context = 0x0

While the 'zpool events' command is handy for interactive debugging
it is not expected to be the primary consumer of zevents.  This ABI
was primarily added to facilitate the addition of a user space
monitoring daemon.  This daemon would consume all events posted by
the kernel and based on the type of event perform an action.  For
most events simply forwarding them on to syslog is likely enough.
But this interface also cleanly allows for more sophisticated
actions to be taken such as generating an email for a failed drive.

Signed-off-by: Brian Behlendorf <[email protected]>

1 files changed, 333 insertions, 0 deletions

diff --git a/cmd/zpool/zpool_main.c b/cmd/zpool/zpool_main.c
index 3d2852326..074f76e81 100644
--- a/cmd/zpool/zpool_main.c
+++ b/cmd/zpool/zpool_main.c
@@ -42,6 +42,8 @@
 #include <zone.h>
 #include <sys/fs/zfs.h>
 #include <sys/stat.h>
+#include <sys/fm/util.h>
+#include <sys/fm/protocol.h>
 
 #include <libzfs.h>
 
@@ -77,6 +79,7 @@ static int zpool_do_export(int, char **);
 static int zpool_do_upgrade(int, char **);
 
 static int zpool_do_history(int, char **);
+static int zpool_do_events(int, char **);
 
 static int zpool_do_get(int, char **);
 static int zpool_do_set(int, char **);
@@ -119,6 +122,7 @@ typedef enum {
 	HELP_SCRUB,
 	HELP_STATUS,
 	HELP_UPGRADE,
+	HELP_EVENTS,
 	HELP_GET,
 	HELP_SET,
 	HELP_SPLIT
@@ -167,6 +171,8 @@ static zpool_command_t command_table[] = {
 	{ "upgrade",	zpool_do_upgrade,	HELP_UPGRADE		},
 	{ NULL },
 	{ "history",	zpool_do_history,	HELP_HISTORY		},
+	{ "events",	zpool_do_events,	HELP_EVENTS		},
+	{ NULL },
 	{ "get",	zpool_do_get,		HELP_GET		},
 	{ "set",	zpool_do_set,		HELP_SET		},
 };
@@ -234,6 +240,8 @@ get_usage(zpool_help_t idx) {
 		return (gettext("\tupgrade\n"
 		    "\tupgrade -v\n"
 		    "\tupgrade [-V version] <-a | pool ...>\n"));
+	case HELP_EVENTS:
+		return (gettext("\tevents [-vfc]\n"));
 	case HELP_GET:
 		return (gettext("\tget <\"all\" | property[,...]> "
 		    "<pool> ...\n"));
@@ -4210,6 +4218,331 @@ zpool_do_history(int argc, char **argv)
 	return (ret);
 }
 
+typedef struct ev_opts {
+	int verbose;
+	int follow;
+	int clear;
+} ev_opts_t;
+
+static void
+zpool_do_events_short(nvlist_t *nvl)
+{
+	char ctime_str[26], str[32], *ptr;
+	int64_t *tv;
+	uint_t n;
+
+	verify(nvlist_lookup_int64_array(nvl, FM_EREPORT_TIME, &tv, &n) == 0);
+	memset(str, ' ', 32);
+	(void) ctime_r((const time_t *)&tv[0], ctime_str);
+	(void) strncpy(str,    ctime_str+4,  6);             /* 'Jun 30'     */
+	(void) strncpy(str+7,  ctime_str+20, 4);             /* '1993'       */
+	(void) strncpy(str+12, ctime_str+11, 8);             /* '21:49:08'   */
+	(void) sprintf(str+20, ".%09lld", (longlong_t)tv[1]);/* '.123456789' */
+	(void) printf(gettext("%s "), str);
+
+	verify(nvlist_lookup_string(nvl, FM_CLASS, &ptr) == 0);
+	(void) printf(gettext("%s\n"), ptr);
+}
+
+static void
+zpool_do_events_nvprint(nvlist_t *nvl, int depth)
+{
+	nvpair_t *nvp;
+
+	for (nvp = nvlist_next_nvpair(nvl, NULL);
+	    nvp != NULL; nvp = nvlist_next_nvpair(nvl, nvp)) {
+
+		data_type_t type = nvpair_type(nvp);
+		const char *name = nvpair_name(nvp);
+
+		boolean_t b;
+		uint8_t i8;
+		uint16_t i16;
+		uint32_t i32;
+		uint64_t i64;
+		char *str;
+		nvlist_t *cnv;
+
+		printf(gettext("%*s%s = "), depth, "", name);
+
+		switch (type) {
+		case DATA_TYPE_BOOLEAN:
+			printf(gettext("%s"), "1");
+			break;
+
+		case DATA_TYPE_BOOLEAN_VALUE:
+			(void) nvpair_value_boolean_value(nvp, &b);
+			printf(gettext("%s"), b ? "1" : "0");
+			break;
+
+		case DATA_TYPE_BYTE:
+			(void) nvpair_value_byte(nvp, &i8);
+			printf(gettext("0x%x"), i8);
+			break;
+
+		case DATA_TYPE_INT8:
+			(void) nvpair_value_int8(nvp, (void *)&i8);
+			printf(gettext("0x%x"), i8);
+			break;
+
+		case DATA_TYPE_UINT8:
+			(void) nvpair_value_uint8(nvp, &i8);
+			printf(gettext("0x%x"), i8);
+			break;
+
+		case DATA_TYPE_INT16:
+			(void) nvpair_value_int16(nvp, (void *)&i16);
+			printf(gettext("0x%x"), i16);
+			break;
+
+		case DATA_TYPE_UINT16:
+			(void) nvpair_value_uint16(nvp, &i16);
+			printf(gettext("0x%x"), i16);
+			break;
+
+		case DATA_TYPE_INT32:
+			(void) nvpair_value_int32(nvp, (void *)&i32);
+			printf(gettext("0x%x"), i32);
+			break;
+
+		case DATA_TYPE_UINT32:
+			(void) nvpair_value_uint32(nvp, &i32);
+			printf(gettext("0x%x"), i32);
+			break;
+
+		case DATA_TYPE_INT64:
+			(void) nvpair_value_int64(nvp, (void *)&i64);
+			printf(gettext("0x%llx"), (u_longlong_t)i64);
+			break;
+
+		case DATA_TYPE_UINT64:
+			(void) nvpair_value_uint64(nvp, &i64);
+			printf(gettext("0x%llx"), (u_longlong_t)i64);
+			break;
+
+		case DATA_TYPE_HRTIME:
+			(void) nvpair_value_hrtime(nvp, (void *)&i64);
+			printf(gettext("0x%llx"), (u_longlong_t)i64);
+			break;
+
+		case DATA_TYPE_STRING:
+			(void) nvpair_value_string(nvp, &str);
+			printf(gettext("\"%s\""), str ? str : "<NULL>");
+			break;
+
+		case DATA_TYPE_NVLIST:
+			printf(gettext("(embedded nvlist)\n"));
+			(void) nvpair_value_nvlist(nvp, &cnv);
+			zpool_do_events_nvprint(cnv, depth + 8);
+			printf(gettext("%*s(end %s)\n"), depth, "", name);
+			break;
+
+		case DATA_TYPE_NVLIST_ARRAY: {
+			nvlist_t **val;
+			uint_t i, nelem;
+
+			(void) nvpair_value_nvlist_array(nvp, &val, &nelem);
+			printf(gettext("(%d embedded nvlists)\n"), nelem);
+			for (i = 0; i < nelem; i++) {
+				printf(gettext("%*s%s[%d] = %s\n"),
+				       depth, "", name, i, "(embedded nvlist)");
+				zpool_do_events_nvprint(val[i], depth + 8);
+				printf(gettext("%*s(end %s[%i])\n"),
+				       depth, "", name, i);
+			}
+			printf(gettext("%*s(end %s)\n"), depth, "", name);
+			}
+			break;
+
+		case DATA_TYPE_INT8_ARRAY: {
+			int8_t *val;
+			uint_t i, nelem;
+
+			(void) nvpair_value_int8_array(nvp, &val, &nelem);
+			for (i = 0; i < nelem; i++)
+				printf(gettext("0x%x "), val[i]);
+
+			break;
+			}
+
+		case DATA_TYPE_UINT8_ARRAY: {
+			uint8_t *val;
+			uint_t i, nelem;
+
+			(void) nvpair_value_uint8_array(nvp, &val, &nelem);
+			for (i = 0; i < nelem; i++)
+				printf(gettext("0x%x "), val[i]);
+
+			break;
+			}
+
+		case DATA_TYPE_INT16_ARRAY: {
+			int16_t *val;
+			uint_t i, nelem;
+
+			(void) nvpair_value_int16_array(nvp, &val, &nelem);
+			for (i = 0; i < nelem; i++)
+				printf(gettext("0x%x "), val[i]);
+
+			break;
+			}
+
+		case DATA_TYPE_UINT16_ARRAY: {
+			uint16_t *val;
+			uint_t i, nelem;
+
+			(void) nvpair_value_uint16_array(nvp, &val, &nelem);
+			for (i = 0; i < nelem; i++)
+				printf(gettext("0x%x "), val[i]);
+
+			break;
+			}
+
+		case DATA_TYPE_INT32_ARRAY: {
+			int32_t *val;
+			uint_t i, nelem;
+
+			(void) nvpair_value_int32_array(nvp, &val, &nelem);
+			for (i = 0; i < nelem; i++)
+				printf(gettext("0x%x "), val[i]);
+
+			break;
+			}
+
+		case DATA_TYPE_UINT32_ARRAY: {
+			uint32_t *val;
+			uint_t i, nelem;
+
+			(void) nvpair_value_uint32_array(nvp, &val, &nelem);
+			for (i = 0; i < nelem; i++)
+				printf(gettext("0x%x "), val[i]);
+
+			break;
+			}
+
+		case DATA_TYPE_INT64_ARRAY: {
+			int64_t *val;
+			uint_t i, nelem;
+
+			(void) nvpair_value_int64_array(nvp, &val, &nelem);
+			for (i = 0; i < nelem; i++)
+				printf(gettext("0x%llx "), (u_longlong_t)val[i]);
+
+			break;
+			}
+
+		case DATA_TYPE_UINT64_ARRAY: {
+			uint64_t *val;
+			uint_t i, nelem;
+
+			(void) nvpair_value_uint64_array(nvp, &val, &nelem);
+			for (i = 0; i < nelem; i++)
+				printf(gettext("0x%llx "), (u_longlong_t)val[i]);
+
+			break;
+			}
+
+		case DATA_TYPE_STRING_ARRAY:
+		case DATA_TYPE_BOOLEAN_ARRAY:
+		case DATA_TYPE_BYTE_ARRAY:
+		case DATA_TYPE_DOUBLE:
+		case DATA_TYPE_UNKNOWN:
+			printf(gettext("<unknown>"));
+			break;
+		}
+
+		printf(gettext("\n"));
+	}
+}
+
+static int
+zpool_do_events_next(ev_opts_t *opts)
+{
+	nvlist_t *nvl;
+	int cleanup_fd, ret, dropped;
+
+        cleanup_fd = open(ZFS_DEV, O_RDWR);
+        VERIFY(cleanup_fd >= 0);
+
+	(void) printf(gettext("%-30s %s\n"), "TIME", "CLASS");
+
+	while (1) {
+		ret = zpool_events_next(g_zfs, &nvl, &dropped,
+		    !!opts->follow, cleanup_fd);
+		if (ret || nvl == NULL)
+			break;
+
+		if (dropped > 0)
+			(void) printf(gettext("dropped %d events\n"), dropped);
+
+		zpool_do_events_short(nvl);
+
+		if (opts->verbose) {
+			zpool_do_events_nvprint(nvl, 8);
+			printf(gettext("\n"));
+		}
+
+		nvlist_free(nvl);
+	}
+
+        VERIFY(0 == close(cleanup_fd));
+
+	return (ret);
+}
+
+static int
+zpool_do_events_clear(ev_opts_t *opts)
+{
+	int count, ret;
+
+	ret = zpool_events_clear(g_zfs, &count);
+	if (!ret)
+		(void) printf(gettext("cleared %d events\n"), count);
+
+	return (ret);
+}
+
+/*
+ * zpool events [-vfc]
+ *
+ * Displays events logs by ZFS.
+ */
+int
+zpool_do_events(int argc, char **argv)
+{
+	ev_opts_t opts = { 0 };
+	int ret;
+	int c;
+
+	/* check options */
+	while ((c = getopt(argc, argv, "vfc")) != -1) {
+		switch (c) {
+		case 'v':
+			opts.verbose = 1;
+			break;
+		case 'f':
+			opts.follow = 1;
+			break;
+		case 'c':
+			opts.clear = 1;
+			break;
+		case '?':
+			(void) fprintf(stderr, gettext("invalid option '%c'\n"),
+			    optopt);
+			usage(B_FALSE);
+		}
+	}
+	argc -= optind;
+	argv += optind;
+
+	if (opts.clear)
+		ret = zpool_do_events_clear(&opts);
+	else
+		ret = zpool_do_events_next(&opts);
+
+	return ret;
+}
+
 static int
 get_callback(zpool_handle_t *zhp, void *data)
 {