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]>

5 files changed, 50 insertions, 31 deletions

diff --git a/module/zfs/include/sys/fm/fs/zfs.h b/module/zfs/include/sys/fm/fs/zfs.h
index c752edc99..d5c71d174 100644
--- a/module/zfs/include/sys/fm/fs/zfs.h
+++ b/module/zfs/include/sys/fm/fs/zfs.h
@@ -35,7 +35,9 @@ extern "C" {
 #define	FM_EREPORT_ZFS_CHECKSUM			"checksum"
 #define	FM_EREPORT_ZFS_IO			"io"
 #define	FM_EREPORT_ZFS_DATA			"data"
+#define	FM_EREPORT_ZFS_CONFIG_SYNC		"config.sync"
 #define	FM_EREPORT_ZFS_POOL			"zpool"
+#define	FM_EREPORT_ZFS_POOL_DESTROY		"zpool.destroy"
 #define	FM_EREPORT_ZFS_DEVICE_UNKNOWN		"vdev.unknown"
 #define	FM_EREPORT_ZFS_DEVICE_OPEN_FAILED	"vdev.open_failed"
 #define	FM_EREPORT_ZFS_DEVICE_CORRUPT_DATA	"vdev.corrupt_data"
@@ -43,9 +45,19 @@ extern "C" {
 #define	FM_EREPORT_ZFS_DEVICE_BAD_GUID_SUM	"vdev.bad_guid_sum"
 #define	FM_EREPORT_ZFS_DEVICE_TOO_SMALL		"vdev.too_small"
 #define	FM_EREPORT_ZFS_DEVICE_BAD_LABEL		"vdev.bad_label"
+#define	FM_EREPORT_ZFS_DEVICE_REMOVE		"vdev.remove"
+#define	FM_EREPORT_ZFS_DEVICE_CLEAR		"vdev.clear"
+#define	FM_EREPORT_ZFS_DEVICE_CHECK		"vdev.check"
+#define	FM_EREPORT_ZFS_DEVICE_SPARE		"vdev.spare"
+#define	FM_EREPORT_ZFS_DEVICE_AUTOEXPAND	"vdev.autoexpand"
 #define	FM_EREPORT_ZFS_IO_FAILURE		"io_failure"
 #define	FM_EREPORT_ZFS_PROBE_FAILURE		"probe_failure"
 #define	FM_EREPORT_ZFS_LOG_REPLAY		"log_replay"
+#define	FM_EREPORT_ZFS_RESILVER_START		"resilver.start"
+#define	FM_EREPORT_ZFS_RESILVER_FINISH		"resilver.finish"
+#define	FM_EREPORT_ZFS_SCRUB_START		"scrub.start"
+#define	FM_EREPORT_ZFS_SCRUB_FINISH		"scrub.finish"
+#define	FM_EREPORT_ZFS_BOOTFS_VDEV_ATTACH	"bootfs.vdev.attach"
 
 #define	FM_EREPORT_PAYLOAD_ZFS_POOL		"pool"
 #define	FM_EREPORT_PAYLOAD_ZFS_POOL_FAILMODE	"pool_failmode"
@@ -56,6 +68,7 @@ extern "C" {
 #define	FM_EREPORT_PAYLOAD_ZFS_VDEV_PATH	"vdev_path"
 #define	FM_EREPORT_PAYLOAD_ZFS_VDEV_DEVID	"vdev_devid"
 #define	FM_EREPORT_PAYLOAD_ZFS_VDEV_FRU		"vdev_fru"
+#define	FM_EREPORT_PAYLOAD_ZFS_VDEV_STATE	"vdev_state"
 #define	FM_EREPORT_PAYLOAD_ZFS_PARENT_GUID	"parent_guid"
 #define	FM_EREPORT_PAYLOAD_ZFS_PARENT_TYPE	"parent_type"
 #define	FM_EREPORT_PAYLOAD_ZFS_PARENT_PATH	"parent_path"
@@ -85,9 +98,9 @@ extern "C" {
 #define	FM_EREPORT_FAILMODE_CONTINUE		"continue"
 #define	FM_EREPORT_FAILMODE_PANIC		"panic"
 
-#define	FM_RESOURCE_REMOVED			"removed"
-#define	FM_RESOURCE_AUTOREPLACE			"autoreplace"
-#define	FM_RESOURCE_STATECHANGE			"statechange"
+#define	FM_EREPORT_RESOURCE_REMOVED		"removed"
+#define	FM_EREPORT_RESOURCE_AUTOREPLACE		"autoreplace"
+#define	FM_EREPORT_RESOURCE_STATECHANGE		"statechange"
 
 #ifdef	__cplusplus
 }
diff --git a/module/zfs/include/sys/fm/protocol.h b/module/zfs/include/sys/fm/protocol.h
index 5eca760da..1ee221286 100644
--- a/module/zfs/include/sys/fm/protocol.h
+++ b/module/zfs/include/sys/fm/protocol.h
@@ -69,6 +69,7 @@ extern "C" {
 /* ereport payload member names */
 #define	FM_EREPORT_DETECTOR		"detector"
 #define	FM_EREPORT_ENA			"ena"
+#define	FM_EREPORT_TIME			"time"
 
 /* list.* event payload member names */
 #define	FM_LIST_EVENT_SIZE		"list-sz"
@@ -327,16 +328,13 @@ extern "C" {
 #define	FM_FMRI_SW_CTXT_ZONE			"zone"
 #define	FM_FMRI_SW_CTXT_CTID			"ctid"
 #define	FM_FMRI_SW_CTXT_STACK			"stack"
+#define	FM_NVA_FREE		0	/* free allocator on nvlist_destroy */
+#define	FM_NVA_RETAIN		1	/* keep allocator on nvlist_destroy */
 
 extern nv_alloc_t *fm_nva_xcreate(char *, size_t);
 extern void fm_nva_xdestroy(nv_alloc_t *);
-
 extern nvlist_t *fm_nvlist_create(nv_alloc_t *);
 extern void fm_nvlist_destroy(nvlist_t *, int);
-
-#define	FM_NVA_FREE	0		/* free allocator on nvlist_destroy */
-#define	FM_NVA_RETAIN	1		/* keep allocator on nvlist_destroy */
-
 extern void fm_ereport_set(nvlist_t *, int, const char *, uint64_t,
     const nvlist_t *, ...);
 extern void fm_payload_set(nvlist_t *, ...);
@@ -350,8 +348,6 @@ extern void fm_fmri_cpu_set(nvlist_t *, int, const nvlist_t *, uint32_t,
     uint8_t *, const char *);
 extern void fm_fmri_mem_set(nvlist_t *, int, const nvlist_t *, const char *,
     const char *, uint64_t);
-extern void fm_authority_set(nvlist_t *, int, const char *, const char *,
-    const char *, const char *);
 extern void fm_fmri_zfs_set(nvlist_t *, int, uint64_t, uint64_t);
 extern void fm_fmri_hc_create(nvlist_t *, int, const nvlist_t *, nvlist_t *,
     nvlist_t *, int, ...);
diff --git a/module/zfs/include/sys/fm/util.h b/module/zfs/include/sys/fm/util.h
index 37334101b..94947d67c 100644
--- a/module/zfs/include/sys/fm/util.h
+++ b/module/zfs/include/sys/fm/util.h
@@ -31,7 +31,6 @@ extern "C" {
 #endif
 
 #include <sys/nvpair.h>
-#include <sys/errorq.h>
 
 /*
  * Shared user/kernel definitions for class length, error channel name,
@@ -71,29 +70,42 @@ typedef struct erpt_dump {
 } erpt_dump_t;
 
 #ifdef _KERNEL
-#include <sys/systm.h>
 
-#define	FM_STK_DEPTH	20	/* maximum stack depth */
-#define	FM_SYM_SZ	64	/* maximum symbol size */
-#define	FM_ERR_PIL	2	/* PIL for ereport_errorq drain processing */
+#define ZEVENT_SHUTDOWN	0x1
 
-#define	FM_EREPORT_PAYLOAD_NAME_STACK		"stack"
+typedef void zevent_cb_t(nvlist_t *, nvlist_t *);
 
-extern errorq_t *ereport_errorq;
-extern void *ereport_dumpbuf;
-extern size_t ereport_dumplen;
+typedef struct zevent_s {
+	nvlist_t	*ev_nvl;       /* protected by the zevent_lock */
+	nvlist_t	*ev_detector;  /* " */
+	list_t		ev_ze_list;    /* " */
+	list_node_t	ev_node;       /* " */
+	zevent_cb_t	*ev_cb;        /* " */
+} zevent_t;
+
+typedef struct zfs_zevent {
+	zevent_t	*ze_zevent;    /* protected by the zevent_lock */
+	list_node_t	ze_node;       /* " */
+	uint64_t	ze_dropped;    /* " */
+} zfs_zevent_t;
 
 extern void fm_init(void);
+extern void fm_fini(void);
 extern void fm_nvprint(nvlist_t *);
-extern void fm_panic(const char *, ...);
-extern void fm_banner(void);
+extern void zfs_zevent_post(nvlist_t *, nvlist_t *, zevent_cb_t *);
+extern void zfs_zevent_drain_all(int *);
+extern int zfs_zevent_fd_hold(int, minor_t *, zfs_zevent_t **);
+extern void zfs_zevent_fd_rele(int);
+extern int zfs_zevent_next(zfs_zevent_t *, nvlist_t **, uint64_t *);
+extern int zfs_zevent_wait(zfs_zevent_t *);
+extern void zfs_zevent_init(zfs_zevent_t **);
+extern void zfs_zevent_destroy(zfs_zevent_t *);
 
-extern void fm_ereport_dump(void);
-extern void fm_ereport_post(nvlist_t *, int);
+#else
 
-extern void fm_payload_stack_add(nvlist_t *, const pc_t *, int);
+static inline void fm_init(void) { }
+static inline void fm_fini(void) { }
 
-extern int is_fm_panic();
 #endif  /* _KERNEL */
 
 #ifdef	__cplusplus
diff --git a/module/zfs/include/sys/zfs_context.h b/module/zfs/include/sys/zfs_context.h
index 558e9e188..af9275b19 100644
--- a/module/zfs/include/sys/zfs_context.h
+++ b/module/zfs/include/sys/zfs_context.h
@@ -58,14 +58,9 @@ extern "C" {
 #include <sys/zone.h>
 #include <sys/uio.h>
 #include <sys/zfs_debug.h>
-#include <sys/sysevent.h>
-#include <sys/sysevent/eventdefs.h>
-#include <sys/sysevent/dev.h>
-#include <sys/fm/util.h>
+#include <sys/fm/fs/zfs.h>
 #include <sys/sunddi.h>
 
-#define	CPU_SEQID	(CPU->cpu_seqid)
-
 #ifdef	__cplusplus
 }
 #endif
diff --git a/module/zfs/include/sys/zfs_ioctl.h b/module/zfs/include/sys/zfs_ioctl.h
index 84bf794fe..ad41561ad 100644
--- a/module/zfs/include/sys/zfs_ioctl.h
+++ b/module/zfs/include/sys/zfs_ioctl.h
@@ -236,6 +236,9 @@ typedef struct zinject_record {
 #define	ZINJECT_FLUSH_ARC	0x2
 #define	ZINJECT_UNLOAD_SPA	0x4
 
+#define	ZEVENT_NONBLOCK		0x1
+#define	ZEVENT_SIZE		1024
+
 typedef struct zfs_share {
 	uint64_t	z_exportdata;
 	uint64_t	z_sharedata;