view src/os/unix/ngx_file_aio_read.c @ 4667:d05ab8793a69 stable-1.2

Merge of r4622, r4623: balancing changes. *) Upstream: smooth weighted round-robin balancing. For edge case weights like { 5, 1, 1 } we now produce { a, a, b, a, c, a, a } sequence instead of { c, b, a, a, a, a, a } produced previously. Algorithm is as follows: on each peer selection we increase current_weight of each eligible peer by its weight, select peer with greatest current_weight and reduce its current_weight by total number of weight points distributed among peers. In case of { 5, 1, 1 } weights this gives the following sequence of current_weight's: a b c 0 0 0 (initial state) 5 1 1 (a selected) -2 1 1 3 2 2 (a selected) -4 2 2 1 3 3 (b selected) 1 -4 3 6 -3 4 (a selected) -1 -3 4 4 -2 5 (c selected) 4 -2 -2 9 -1 -1 (a selected) 2 -1 -1 7 0 0 (a selected) 0 0 0 To preserve weight reduction in case of failures the effective_weight variable was introduced, which usually matches peer's weight, but is reduced temporarily on peer failures. This change also fixes loop with backup servers and proxy_next_upstream http_404 (ticket #47), and skipping alive upstreams in some cases if there are multiple dead ones (ticket #64). *) Upstream: fixed ip_hash rebalancing with the "down" flag. Due to weight being set to 0 for down peers, order of peers after sorting wasn't the same as without the "down" flag (with down peers at the end), resulting in client rebalancing for clients on other servers. The only rebalancing which should happen after adding "down" to a server is one for clients on the server. The problem was introduced in r1377 (which fixed endless loop by setting weight to 0 for down servers). The loop is no longer possible with new smooth algorithm, so preserving original weight is safe.
author Maxim Dounin <mdounin@mdounin.ru>
date Mon, 04 Jun 2012 11:21:58 +0000
parents 30eff7580d0c
children ccad84a174e0
line wrap: on
line source


/*
 * Copyright (C) Igor Sysoev
 * Copyright (C) Nginx, Inc.
 */


#include <ngx_config.h>
#include <ngx_core.h>
#include <ngx_event.h>


/*
 * FreeBSD file AIO features and quirks:
 *
 *    if an asked data are already in VM cache, then aio_error() returns 0,
 *    and the data are already copied in buffer;
 *
 *    aio_read() preread in VM cache as minimum 16K (probably BKVASIZE);
 *    the first AIO preload may be up to 128K;
 *
 *    aio_read/aio_error() may return EINPROGRESS for just written data;
 *
 *    kqueue EVFILT_AIO filter is level triggered only: an event repeats
 *    until aio_return() will be called;
 *
 *    aio_cancel() cannot cancel file AIO: it returns AIO_NOTCANCELED always.
 */


extern int  ngx_kqueue;


static ssize_t ngx_file_aio_result(ngx_file_t *file, ngx_event_aio_t *aio,
    ngx_event_t *ev);
static void ngx_file_aio_event_handler(ngx_event_t *ev);


ssize_t
ngx_file_aio_read(ngx_file_t *file, u_char *buf, size_t size, off_t offset,
    ngx_pool_t *pool)
{
    int               n;
    ngx_event_t      *ev;
    ngx_event_aio_t  *aio;

    if (!ngx_file_aio) {
        return ngx_read_file(file, buf, size, offset);
    }

    aio = file->aio;

    if (aio == NULL) {
        aio = ngx_pcalloc(pool, sizeof(ngx_event_aio_t));
        if (aio == NULL) {
            return NGX_ERROR;
        }

        aio->file = file;
        aio->fd = file->fd;
        aio->event.data = aio;
        aio->event.ready = 1;
        aio->event.log = file->log;
#if (NGX_HAVE_AIO_SENDFILE)
        aio->last_offset = -1;
#endif
        file->aio = aio;
    }

    ev = &aio->event;

    if (!ev->ready) {
        ngx_log_error(NGX_LOG_ALERT, file->log, 0,
                      "second aio post for \"%V\"", &file->name);
        return NGX_AGAIN;
    }

    ngx_log_debug4(NGX_LOG_DEBUG_CORE, file->log, 0,
                   "aio complete:%d @%O:%z %V",
                   ev->complete, offset, size, &file->name);

    if (ev->complete) {
        ev->complete = 0;
        ngx_set_errno(aio->err);

        if (aio->err == 0) {
            return aio->nbytes;
        }

        ngx_log_error(NGX_LOG_CRIT, file->log, ngx_errno,
                      "aio read \"%s\" failed", file->name.data);

        return NGX_ERROR;
    }

    ngx_memzero(&aio->aiocb, sizeof(struct aiocb));

    aio->aiocb.aio_fildes = file->fd;
    aio->aiocb.aio_offset = offset;
    aio->aiocb.aio_buf = buf;
    aio->aiocb.aio_nbytes = size;
#if (NGX_HAVE_KQUEUE)
    aio->aiocb.aio_sigevent.sigev_notify_kqueue = ngx_kqueue;
    aio->aiocb.aio_sigevent.sigev_notify = SIGEV_KEVENT;
    aio->aiocb.aio_sigevent.sigev_value.sigval_ptr = ev;
#endif
    ev->handler = ngx_file_aio_event_handler;

    n = aio_read(&aio->aiocb);

    if (n == -1) {
        n = ngx_errno;

        if (n == NGX_EAGAIN) {
            return ngx_read_file(file, buf, size, offset);
        }

        ngx_log_error(NGX_LOG_CRIT, file->log, n,
                      "aio_read(\"%V\") failed", &file->name);

        if (n == NGX_ENOSYS) {
            ngx_file_aio = 0;
            return ngx_read_file(file, buf, size, offset);
        }

        return NGX_ERROR;
    }

    ngx_log_debug2(NGX_LOG_DEBUG_CORE, file->log, 0,
                   "aio_read: fd:%d %d", file->fd, n);

    ev->active = 1;
    ev->ready = 0;
    ev->complete = 0;

    return ngx_file_aio_result(aio->file, aio, ev);
}


static ssize_t
ngx_file_aio_result(ngx_file_t *file, ngx_event_aio_t *aio, ngx_event_t *ev)
{
    int        n;
    ngx_err_t  err;

    n = aio_error(&aio->aiocb);

    ngx_log_debug2(NGX_LOG_DEBUG_CORE, file->log, 0,
                   "aio_error: fd:%d %d", file->fd, n);

    if (n == -1) {
        err = ngx_errno;
        aio->err = err;

        ngx_log_error(NGX_LOG_ALERT, file->log, err,
                      "aio_error(\"%V\") failed", &file->name);
        return NGX_ERROR;
    }

    if (n == NGX_EINPROGRESS) {
        if (ev->ready) {
            ev->ready = 0;
            ngx_log_error(NGX_LOG_ALERT, file->log, n,
                          "aio_read(\"%V\") still in progress",
                          &file->name);
        }

        return NGX_AGAIN;
    }

    n = aio_return(&aio->aiocb);

    if (n == -1) {
        err = ngx_errno;
        aio->err = err;
        ev->ready = 1;

        ngx_log_error(NGX_LOG_CRIT, file->log, err,
                      "aio_return(\"%V\") failed", &file->name);
        return NGX_ERROR;
    }

    aio->err = 0;
    aio->nbytes = n;
    ev->ready = 1;
    ev->active = 0;

    ngx_log_debug2(NGX_LOG_DEBUG_CORE, file->log, 0,
                   "aio_return: fd:%d %d", file->fd, n);

    return n;
}


static void
ngx_file_aio_event_handler(ngx_event_t *ev)
{
    ngx_event_aio_t  *aio;

    aio = ev->data;

    ngx_log_debug2(NGX_LOG_DEBUG_CORE, ev->log, 0,
                   "aio event handler fd:%d %V", aio->fd, &aio->file->name);

    if (ngx_file_aio_result(aio->file, aio, ev) != NGX_AGAIN) {
        aio->handler(ev);
    }
}