mdadm/msg.c

476 lines
10 KiB
C
Raw Permalink Normal View History

/*
* Copyright (C) 2008 Intel Corporation
*
* mdmon socket / message handling
*
* This program is free software; you can redistribute it and/or modify it
* under the terms and conditions of the GNU General Public License,
* version 2, as published by the Free Software Foundation.
*
* This program is distributed in the hope it will be useful, but WITHOUT
* ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
* FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
* more details.
*
* You should have received a copy of the GNU General Public License along with
* this program; if not, write to the Free Software Foundation, Inc.,
* 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA.
*/
#ifndef _GNU_SOURCE
#define _GNU_SOURCE
#endif
#include <unistd.h>
#include <stdio.h>
#include <string.h>
#include <stdlib.h>
#include <errno.h>
#include <fcntl.h>
#include <sys/types.h>
#include <sys/socket.h>
#include <sys/un.h>
#include "mdadm.h"
2008-07-12 12:27:40 +02:00
#include "mdmon.h"
2008-07-12 12:27:40 +02:00
static const __u32 start_magic = 0x5a5aa5a5;
static const __u32 end_magic = 0xa5a55a5a;
static int send_buf(int fd, const void* buf, int len, int tmo)
{
2008-07-12 12:27:40 +02:00
fd_set set;
int rv;
struct timeval timeout = {tmo, 0};
struct timeval *ptmo = tmo ? &timeout : NULL;
while (len) {
FD_ZERO(&set);
FD_SET(fd, &set);
rv = select(fd+1, NULL, &set, NULL, ptmo);
if (rv <= 0)
return -1;
rv = write(fd, buf, len);
if (rv <= 0)
return -1;
len -= rv;
buf += rv;
}
return 0;
}
2008-07-12 12:27:40 +02:00
static int recv_buf(int fd, void* buf, int len, int tmo)
{
fd_set set;
int rv;
struct timeval timeout = {tmo, 0};
struct timeval *ptmo = tmo ? &timeout : NULL;
while (len) {
FD_ZERO(&set);
FD_SET(fd, &set);
rv = select(fd+1, &set, NULL, NULL, ptmo);
if (rv <= 0)
return -1;
rv = read(fd, buf, len);
if (rv <= 0)
return -1;
len -= rv;
buf += rv;
}
return 0;
}
2008-07-12 12:27:40 +02:00
int send_message(int fd, struct metadata_update *msg, int tmo)
{
__s32 len = msg->len;
2008-07-12 12:27:40 +02:00
int rv;
rv = send_buf(fd, &start_magic, 4, tmo);
rv = rv ?: send_buf(fd, &len, 4, tmo);
if (len > 0)
2008-07-12 12:27:40 +02:00
rv = rv ?: send_buf(fd, msg->buf, msg->len, tmo);
rv = send_buf(fd, &end_magic, 4, tmo);
return rv;
}
2008-07-12 12:27:40 +02:00
int receive_message(int fd, struct metadata_update *msg, int tmo)
{
2008-07-12 12:27:40 +02:00
__u32 magic;
__s32 len;
2008-07-12 12:27:40 +02:00
int rv;
rv = recv_buf(fd, &magic, 4, tmo);
if (rv < 0 || magic != start_magic)
return -1;
rv = recv_buf(fd, &len, 4, tmo);
if (rv < 0 || len > MSG_MAX_LEN)
return -1;
if (len > 0) {
msg->buf = xmalloc(len);
2008-07-12 12:27:40 +02:00
rv = recv_buf(fd, msg->buf, len, tmo);
if (rv < 0) {
free(msg->buf);
return -1;
}
} else
msg->buf = NULL;
rv = recv_buf(fd, &magic, 4, tmo);
if (rv < 0 || magic != end_magic) {
free(msg->buf);
return -1;
}
msg->len = len;
return 0;
}
2008-07-12 12:27:40 +02:00
int ack(int fd, int tmo)
{
2008-07-12 12:27:40 +02:00
struct metadata_update msg = { .len = 0 };
return send_message(fd, &msg, tmo);
}
2008-07-12 12:27:40 +02:00
int wait_reply(int fd, int tmo)
{
2008-07-12 12:27:40 +02:00
struct metadata_update msg;
int err = receive_message(fd, &msg, tmo);
/* mdmon sent extra data, but caller only cares that we got a
* successful reply
*/
if (err == 0 && msg.len > 0)
free(msg.buf);
return err;
}
int connect_monitor(char *devname)
{
char path[100];
int sfd;
long fl;
struct sockaddr_un addr;
int pos;
char *c;
Fix all the confusion over directories once and for all. We now have 3 directory definitions: mdmon directory for its pid and sock files (compile time define, not changable at run time), mdmonitor directory which is for the mdadm monitor mode pid file (can only be passed in via command line at the time mdadm is invoked in monitor mode), and the directory for the mdadm incremental assembly map file (compile time define, not changable at run time). Only the mdadm map file still hunts multiple locations, and the number of locations has been reduced to /var/run and the compile time specified location. Re-use of similar sounding defines that actually didn't denote their actual usage at compile time made it more difficult for a person to know what affect changing the compile time defines would have on the resulting programs. This patch renames the various defines to clearly identify which item the define affects. It also reduces the number of various directories which will be searched for these files as this has lead to confusion in mdadm and mdmon in terms of which files should take precedence when files exist in multiple locations, etc. It's best if the person compiling the program intentionally and with planning selects the right directories to be used for the various purposes. Which directory is right depends on which items you are talking about and what boot loader your system uses and what initramfs generation program your system uses. Because of the inter-dependency of all these items it would typically be up to the distribution that mdadm is being integrated into to select the correct values for these defines. Signed-off-by: Doug Ledford <dledford@redhat.com>
2010-07-20 18:15:37 +02:00
pos = sprintf(path, "%s/", MDMON_DIR);
if (is_subarray(devname)) {
devname++;
c = strchr(devname, '/');
if (!c)
return -1;
snprintf(&path[pos], c - devname + 1, "%s", devname);
pos += c - devname;
} else
pos += sprintf(&path[pos], "%s", devname);
sprintf(&path[pos], ".sock");
sfd = socket(PF_LOCAL, SOCK_STREAM, 0);
if (sfd < 0)
return -1;
addr.sun_family = PF_LOCAL;
strcpy(addr.sun_path, path);
if (connect(sfd, (struct sockaddr*)&addr, sizeof(addr)) < 0) {
close(sfd);
return -1;
}
fl = fcntl(sfd, F_GETFL, 0);
fl |= O_NONBLOCK;
fcntl(sfd, F_SETFL, fl);
return sfd;
}
int fping_monitor(int sfd)
{
int err = 0;
if (sfd < 0)
return sfd;
/* try to ping existing socket */
if (ack(sfd, 20) != 0)
err = -1;
/* check the reply */
if (!err && wait_reply(sfd, 20) != 0)
err = -1;
return err;
}
/* give the monitor a chance to update the metadata */
int ping_monitor(char *devname)
{
int sfd = connect_monitor(devname);
int err;
if (sfd >= 0) {
err = fping_monitor(sfd);
close(sfd);
} else
err = -1;
return err;
}
static char *ping_monitor_version(char *devname)
{
int sfd = connect_monitor(devname);
struct metadata_update msg;
int err = 0;
if (sfd < 0)
return NULL;
if (ack(sfd, 20) != 0)
err = -1;
if (!err && receive_message(sfd, &msg, 20) != 0)
err = -1;
close(sfd);
if (err || !msg.len || !msg.buf)
return NULL;
return msg.buf;
}
int unblock_subarray(struct mdinfo *sra, const int unfreeze)
{
char buf[64];
int rc = 0;
if (sra) {
sprintf(buf, "external:%s\n", sra->text_version);
buf[9] = '/';
} else
buf[9] = '-';
if (buf[9] == '-' ||
sysfs_set_str(sra, NULL, "metadata_version", buf) ||
(unfreeze &&
sysfs_attribute_available(sra, NULL, "sync_action") &&
sysfs_set_str(sra, NULL, "sync_action", "idle")))
rc = -1;
return rc;
}
int block_subarray(struct mdinfo *sra)
{
char buf[64];
int rc = 0;
sprintf(buf, "external:%s\n", sra->text_version);
buf[9] = '-';
if (sysfs_set_str(sra, NULL, "metadata_version", buf))
rc = -1;
return rc;
}
/* check mdmon version if it supports
* array blocking mechanism
*/
int check_mdmon_version(char *container)
{
char *version = NULL;
if (!mdmon_running(container)) {
/* if mdmon is not active we assume that any instance that is
* later started will match the current mdadm version, if this
* assumption is violated we may inadvertantly rebuild an array
* that was meant for reshape, or start rebuild on a spare that
* was to be moved to another container
*/
/* pass */;
} else {
int ver;
version = ping_monitor_version(container);
ver = version ? mdadm_version(version) : -1;
free(version);
if (ver < 3002000) {
pr_err("mdmon instance for %s cannot be disabled\n",
container);
return -1;
}
}
return 0;
}
/**
* block_monitor - prevent mdmon spare assignment
* @container - container to block
* @freeze - flag to additionally freeze sync_action
*
* This is used by the reshape code to freeze the container, and the
* auto-rebuild implementation to atomically move spares.
* In both cases we need to stop mdmon from assigning spares to replace
* failed devices as we might have other plans for the spare.
* For the reshape case we also need to 'freeze' sync_action so that
* no recovery happens until we have fully prepared for the reshape.
*
* We tell mdmon that the array is frozen by marking the 'metadata' name
* with a leading '-'. The previously told mdmon "Don't make this array
* read/write, leave it readonly". Now it means a more general "Don't
* reconfigure this array at all".
* As older versions of mdmon (which might run from initrd) don't understand
* this, we first check that the running mdmon is new enough.
*/
int block_monitor(char *container, const int freeze)
{
struct mdstat_ent *ent, *e, *e2;
struct mdinfo *sra = NULL;
char buf[64];
int rv = 0;
if (check_mdmon_version(container))
return -1;
ent = mdstat_read(0, 0);
if (!ent) {
pr_err("failed to read /proc/mdstat while disabling mdmon\n");
return -1;
}
/* freeze container contents */
for (e = ent; e; e = e->next) {
if (!is_container_member(e, container))
continue;
sysfs_free(sra);
sra = sysfs_read(-1, e->devnm, GET_VERSION);
if (!sra) {
pr_err("failed to read sysfs for subarray%s\n",
to_subarray(e, container));
break;
}
/* can't reshape an array that we can't monitor */
if (sra->text_version[0] == '-')
break;
if (freeze && sysfs_freeze_array(sra) < 1)
break;
/* flag this array to not be modified by mdmon (close race with
* takeover in reshape case and spare reassignment in the
* auto-rebuild case)
*/
if (block_subarray(sra))
break;
ping_monitor(container);
/* check that we did not race with recovery */
if ((freeze &&
!sysfs_attribute_available(sra, NULL, "sync_action")) ||
(freeze &&
sysfs_attribute_available(sra, NULL, "sync_action") &&
sysfs_get_str(sra, NULL, "sync_action", buf, 20) > 0 &&
strcmp(buf, "frozen\n") == 0))
/* pass */;
else {
unblock_subarray(sra, 0);
break;
}
/* Double check against races - there should be no spares
* or part-spares
*/
sysfs_free(sra);
sra = sysfs_read(-1, e->devnm, GET_DEVS | GET_STATE);
if (sra && sra->array.spare_disks > 0) {
unblock_subarray(sra, freeze);
break;
}
}
if (e) {
pr_err("failed to freeze subarray%s\n",
to_subarray(e, container));
/* thaw the partially frozen container */
for (e2 = ent; e2 && e2 != e; e2 = e2->next) {
if (!is_container_member(e2, container))
continue;
sysfs_free(sra);
sra = sysfs_read(-1, e2->devnm, GET_VERSION);
if (unblock_subarray(sra, freeze))
pr_err("Failed to unfreeze %s\n", e2->devnm);
}
ping_monitor(container); /* cleared frozen */
rv = -1;
}
sysfs_free(sra);
free_mdstat(ent);
return rv;
}
void unblock_monitor(char *container, const int unfreeze)
{
struct mdstat_ent *ent, *e;
struct mdinfo *sra = NULL;
int to_ping = 0;
ent = mdstat_read(0, 0);
if (!ent) {
pr_err("failed to read /proc/mdstat while unblocking container\n");
return;
}
/* unfreeze container contents */
for (e = ent; e; e = e->next) {
if (!is_container_member(e, container))
continue;
sysfs_free(sra);
sra = sysfs_read(-1, e->devnm, GET_VERSION|GET_LEVEL);
if (!sra)
continue;
if (sra->array.level > 0)
to_ping++;
if (unblock_subarray(sra, unfreeze))
pr_err("Failed to unfreeze %s\n", e->devnm);
}
if (to_ping)
ping_monitor(container);
sysfs_free(sra);
free_mdstat(ent);
}
/* give the manager a chance to view the updated container state. This
* would naturally happen due to the manager noticing a change in
* /proc/mdstat; however, pinging encourages this detection to happen
* while an exclusive open() on the container is active
*/
int ping_manager(char *devname)
{
int sfd = connect_monitor(devname);
struct metadata_update msg = { .len = -1 };
int err = 0;
if (sfd < 0)
return sfd;
err = send_message(sfd, &msg, 20);
/* check the reply */
if (!err && wait_reply(sfd, 20) != 0)
err = -1;
close(sfd);
return err;
}
/* using takeover operation for grow purposes, mdadm has to be sure
* that mdmon processes all updates, and if necessary it will be closed
* at takeover to raid0 operation
*/
void flush_mdmon(char *container)
{
ping_manager(container);
ping_monitor(container);
}