From 5d354e525a56955ae7f68062e283dda85ab07794 Mon Sep 17 00:00:00 2001 From: Yu Watanabe Date: Tue, 15 Mar 2022 13:50:06 +0900 Subject: [PATCH] udev: requeue event when the corresponding block device is locked by another process Previously, if a block device is locked by another process, then the corresponding worker skip to process the corresponding event, and does not broadcast the uevent to libudev listners. This causes several issues: - During a period of a device being locked by a process, if a user trigger an event with `udevadm trigger --settle`, then it never returned. - When there is a delay between close and unlock in a process, then the synthesized events triggered by inotify may not be processed. This can happens easily by wrapping mkfs with flock. This causes severe issues e.g. new devlinks are not created, or old devlinks are not removed. This commit makes events are requeued with a tiny delay when the corresponding block devices are locked by other processes. With this way, the triggered uevent may be delayed but is always processed by udevd. Hence, the above issues can be solved. Also, it is not necessary to watch a block device unconditionally when it is already locked. Hence, the logic is dropped. --- src/udev/udevd.c | 154 +++++++++++++++++++++++++++++------------------ 1 file changed, 97 insertions(+), 57 deletions(-) diff --git a/src/udev/udevd.c b/src/udev/udevd.c index d153b03a38e..973727375b6 100644 --- a/src/udev/udevd.c +++ b/src/udev/udevd.c @@ -70,6 +70,8 @@ #include "version.h" #define WORKER_NUM_MAX 2048U +#define EVENT_RETRY_INTERVAL_USEC (200 * USEC_PER_MSEC) +#define EVENT_RETRY_TIMEOUT_USEC (3 * USEC_PER_MINUTE) static bool arg_debug = false; static int arg_daemonize = false; @@ -128,6 +130,8 @@ typedef struct Event { sd_device_action_t action; uint64_t seqnum; uint64_t blocker_seqnum; + usec_t retry_again_next_usec; + usec_t retry_again_timeout_usec; sd_event_source *timeout_warning_event; sd_event_source *timeout_event; @@ -152,8 +156,13 @@ typedef struct Worker { } Worker; /* passed from worker to main process */ -typedef struct WorkerMessage { -} WorkerMessage; +typedef enum EventResult { + EVENT_RESULT_SUCCESS, + EVENT_RESULT_FAILED, + EVENT_RESULT_TRY_AGAIN, /* when the block device is locked by another process. */ + _EVENT_RESULT_MAX, + _EVENT_RESULT_INVALID = -EINVAL, +} EventResult; static Event *event_free(Event *event) { if (!event) @@ -360,10 +369,11 @@ static void device_broadcast(sd_device_monitor *monitor, sd_device *dev) { "Failed to broadcast event to libudev listeners, ignoring: %m"); } -static int worker_send_message(int fd) { - WorkerMessage message = {}; +static int worker_send_result(Manager *manager, EventResult result) { + assert(manager); + assert(manager->worker_watch[WRITE_END] >= 0); - return loop_write(fd, &message, sizeof(message), false); + return loop_write(manager->worker_watch[WRITE_END], &result, sizeof(result), false); } static int worker_lock_block_device(sd_device *dev, int *ret_fd) { @@ -490,44 +500,12 @@ static int worker_process_device(Manager *manager, sd_device *dev) { if (!udev_event) return -ENOMEM; + /* If this is a block device and the device is locked currently via the BSD advisory locks, + * someone else is using it exclusively. We don't run our udev rules now to not interfere. + * Instead of processing the event, we requeue the event and will try again after a delay. + * + * The user-facing side of this: https://systemd.io/BLOCK_DEVICE_LOCKING */ r = worker_lock_block_device(dev, &fd_lock); - if (r == -EAGAIN) { - /* So this is a block device and the device is locked currently via the BSD advisory locks — - * someone else is exclusively using it. This means we don't run our udev rules now, to not - * interfere. However we want to know when the device is unlocked again, and retrigger the - * device again then, so that the rules are run eventually. For that we use IN_CLOSE_WRITE - * inotify watches (which isn't exactly the same as waiting for the BSD locks to release, but - * not totally off, as long as unlock+close() is done together, as it usually is). - * - * (The user-facing side of this: https://systemd.io/BLOCK_DEVICE_LOCKING) - * - * There's a bit of a chicken and egg problem here for this however: inotify watching is - * supposed to be enabled via an option set via udev rules (OPTIONS+="watch"). If we skip the - * udev rules here however (as we just said we do), we would thus never see that specific - * udev rule, and thus never turn on inotify watching. But in order to catch up eventually - * and run them we need the inotify watching: hence a classic chicken and egg problem. - * - * Our way out here: if we see the block device locked, unconditionally watch the device via - * inotify, regardless of any explicit request via OPTIONS+="watch". Thus, a device that is - * currently locked via the BSD file locks will be treated as if we ran a single udev rule - * only for it: the one that turns on inotify watching for it. If we eventually see the - * inotify IN_CLOSE_WRITE event, and then run the rules after all and we then realize that - * this wasn't actually requested (i.e. no OPTIONS+="watch" set) we'll simply turn off the - * watching again (see below). Effectively this means: inotify watching is now enabled either - * a) when the udev rules say so, or b) while the device is locked. - * - * Worst case scenario hence: in the (unlikely) case someone locked the device and we clash - * with that we might do inotify watching for a brief moment for a device where we actually - * weren't supposed to. But that shouldn't be too bad, in particular as BSD locks being taken - * on a block device is kinda an indication that the inotify logic is desired too, to some - * degree — they go hand-in-hand after all. */ - - log_device_debug(dev, "Block device is currently locked, installing watch to wait until the lock is released."); - (void) udev_watch_begin(manager->inotify_fd, dev); - - /* Now the watch is installed, let's lock the device again, maybe in the meantime things changed */ - r = worker_lock_block_device(dev, &fd_lock); - } if (r < 0) return r; @@ -560,25 +538,29 @@ static int worker_process_device(Manager *manager, sd_device *dev) { static int worker_device_monitor_handler(sd_device_monitor *monitor, sd_device *dev, void *userdata) { Manager *manager = userdata; + EventResult result; int r; assert(dev); assert(manager); r = worker_process_device(manager, dev); - if (r == -EAGAIN) - /* if we couldn't acquire the flock(), then proceed quietly */ - log_device_debug_errno(dev, r, "Device currently locked, not processing."); - else { - if (r < 0) - log_device_warning_errno(dev, r, "Failed to process device, ignoring: %m"); + if (r == -EAGAIN) { + /* if we couldn't acquire the flock(), then requeue the event */ + result = EVENT_RESULT_TRY_AGAIN; + log_device_debug_errno(dev, r, "Block device is currently locked, requeueing the event."); + } else if (r < 0) { + result = EVENT_RESULT_FAILED; + log_device_warning_errno(dev, r, "Failed to process device, ignoring: %m"); + } else + result = EVENT_RESULT_SUCCESS; + if (result != EVENT_RESULT_TRY_AGAIN) /* send processed event back to libudev listeners */ device_broadcast(monitor, dev); - } /* send udevd the result of the event execution */ - r = worker_send_message(manager->worker_watch[WRITE_END]); + r = worker_send_result(manager, result); if (r < 0) log_device_warning_errno(dev, r, "Failed to send signal to main daemon, ignoring: %m"); @@ -794,6 +776,17 @@ static int event_is_blocked(Event *event) { assert(event->manager); assert(event->blocker_seqnum <= event->seqnum); + if (event->retry_again_next_usec > 0) { + usec_t now_usec; + + r = sd_event_now(event->manager->event, clock_boottime_or_monotonic(), &now_usec); + if (r < 0) + return r; + + if (event->retry_again_next_usec <= now_usec) + return true; + } + if (event->blocker_seqnum == event->seqnum) /* we have checked previously and no blocker found */ return false; @@ -980,6 +973,44 @@ static int event_queue_start(Manager *manager) { return 0; } +static int event_requeue(Event *event) { + usec_t now_usec; + int r; + + assert(event); + assert(event->manager); + assert(event->manager->event); + + event->timeout_warning_event = sd_event_source_disable_unref(event->timeout_warning_event); + event->timeout_event = sd_event_source_disable_unref(event->timeout_event); + + /* add a short delay to suppress busy loop */ + r = sd_event_now(event->manager->event, clock_boottime_or_monotonic(), &now_usec); + if (r < 0) + return log_device_warning_errno(event->dev, r, + "Failed to get current time, " + "skipping event (SEQNUM=%"PRIu64", ACTION=%s): %m", + event->seqnum, strna(device_action_to_string(event->action))); + + if (event->retry_again_timeout_usec > 0 && event->retry_again_timeout_usec <= now_usec) + return log_device_warning_errno(event->dev, SYNTHETIC_ERRNO(ETIMEDOUT), + "The underlying block device is locked by a process more than %s, " + "skipping event (SEQNUM=%"PRIu64", ACTION=%s).", + FORMAT_TIMESPAN(EVENT_RETRY_TIMEOUT_USEC, USEC_PER_MINUTE), + event->seqnum, strna(device_action_to_string(event->action))); + + event->retry_again_next_usec = usec_add(now_usec, EVENT_RETRY_INTERVAL_USEC); + if (event->retry_again_timeout_usec == 0) + event->retry_again_timeout_usec = usec_add(now_usec, EVENT_RETRY_TIMEOUT_USEC); + + if (event->worker && event->worker->event == event) + event->worker->event = NULL; + event->worker = NULL; + + event->state = EVENT_QUEUED; + return 0; +} + static int event_queue_insert(Manager *manager, sd_device *dev) { sd_device_action_t action; uint64_t seqnum; @@ -1054,11 +1085,8 @@ static int on_worker(sd_event_source *s, int fd, uint32_t revents, void *userdat assert(manager); for (;;) { - WorkerMessage msg; - struct iovec iovec = { - .iov_base = &msg, - .iov_len = sizeof(msg), - }; + EventResult result; + struct iovec iovec = IOVEC_MAKE(&result, sizeof(result)); CMSG_BUFFER_TYPE(CMSG_SPACE(sizeof(struct ucred))) control; struct msghdr msghdr = { .msg_iov = &iovec, @@ -1081,7 +1109,7 @@ static int on_worker(sd_event_source *s, int fd, uint32_t revents, void *userdat cmsg_close_all(&msghdr); - if (size != sizeof(WorkerMessage)) { + if (size != sizeof(EventResult)) { log_warning("Ignoring worker message with invalid size %zi bytes", size); continue; } @@ -1106,6 +1134,11 @@ static int on_worker(sd_event_source *s, int fd, uint32_t revents, void *userdat worker->state = WORKER_IDLE; /* worker returned */ + if (result == EVENT_RESULT_TRY_AGAIN && + event_requeue(worker->event) < 0) + device_broadcast(manager->monitor, worker->event->dev); + + /* When event_requeue() succeeds, worker->event is NULL, and event_free() handles NULL gracefully. */ event_free(worker->event); } @@ -1467,8 +1500,15 @@ static int on_post(sd_event_source *s, void *userdata) { assert(manager); - if (!LIST_IS_EMPTY(manager->events)) + if (!LIST_IS_EMPTY(manager->events)) { + /* Try to process pending events if idle workers exist. Why is this necessary? + * When a worker finished an event and became idle, even if there was a pending event, + * the corresponding device might have been locked and the processing of the event + * delayed for a while, preventing the worker from processing the event immediately. + * Now, the device may be unlocked. Let's try again! */ + event_queue_start(manager); return 1; + } /* There are no queued events. Let's remove /run/udev/queue and clean up the idle processes. */