1
0
mirror of https://github.com/samba-team/samba.git synced 2025-01-11 05:18:09 +03:00
samba-mirror/ctdb/common/ctdb_util.c
Ronnie Sahlberg 8d12313d6b ouch, the ordering of the constants and the strings must be kept in sync
manually   and ther eis no check for errors.     should fix this later

(This used to be ctdb commit e824af1a41f8ceec1edf6b3d1d6e1758fa00deb2)
2010-08-30 19:43:35 +10:00

691 lines
15 KiB
C

/*
ctdb utility code
Copyright (C) Andrew Tridgell 2006
This program is free software; you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation; either version 3 of the License, or
(at your option) any later version.
This program is distributed in the hope that 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, see <http://www.gnu.org/licenses/>.
*/
#include "includes.h"
#include "lib/tevent/tevent.h"
#include "lib/tdb/include/tdb.h"
#include "system/network.h"
#include "system/filesys.h"
#include "system/wait.h"
#include "system/shmem.h"
#include "../include/ctdb_private.h"
int LogLevel = DEBUG_NOTICE;
int this_log_level = 0;
/*
return error string for last error
*/
const char *ctdb_errstr(struct ctdb_context *ctdb)
{
return ctdb->err_msg;
}
/*
remember an error message
*/
void ctdb_set_error(struct ctdb_context *ctdb, const char *fmt, ...)
{
va_list ap;
talloc_free(ctdb->err_msg);
va_start(ap, fmt);
ctdb->err_msg = talloc_vasprintf(ctdb, fmt, ap);
DEBUG(DEBUG_ERR,("ctdb error: %s\n", ctdb->err_msg));
va_end(ap);
}
/*
a fatal internal error occurred - no hope for recovery
*/
void ctdb_fatal(struct ctdb_context *ctdb, const char *msg)
{
DEBUG(DEBUG_ALERT,("ctdb fatal error: %s\n", msg));
abort();
}
/*
parse a IP:port pair
*/
int ctdb_parse_address(struct ctdb_context *ctdb,
TALLOC_CTX *mem_ctx, const char *str,
struct ctdb_address *address)
{
struct servent *se;
setservent(0);
se = getservbyname("ctdb", "tcp");
endservent();
address->address = talloc_strdup(mem_ctx, str);
CTDB_NO_MEMORY(ctdb, address->address);
if (se == NULL) {
address->port = CTDB_PORT;
} else {
address->port = ntohs(se->s_port);
}
return 0;
}
/*
check if two addresses are the same
*/
bool ctdb_same_address(struct ctdb_address *a1, struct ctdb_address *a2)
{
return strcmp(a1->address, a2->address) == 0 && a1->port == a2->port;
}
/*
hash function for mapping data to a VNN - taken from tdb
*/
uint32_t ctdb_hash(const TDB_DATA *key)
{
uint32_t value; /* Used to compute the hash value. */
uint32_t i; /* Used to cycle through random values. */
/* Set the initial value from the key size. */
for (value = 0x238F13AF * key->dsize, i=0; i < key->dsize; i++)
value = (value + (key->dptr[i] << (i*5 % 24)));
return (1103515243 * value + 12345);
}
/*
a type checking varient of idr_find
*/
static void *_idr_find_type(struct idr_context *idp, int id, const char *type, const char *location)
{
void *p = idr_find(idp, id);
if (p && talloc_check_name(p, type) == NULL) {
DEBUG(DEBUG_ERR,("%s idr_find_type expected type %s but got %s\n",
location, type, talloc_get_name(p)));
return NULL;
}
return p;
}
/*
update a max latency number
*/
void ctdb_latency(struct ctdb_db_context *ctdb_db, const char *name, double *latency, struct timeval t)
{
double l = timeval_elapsed(&t);
if (l > *latency) {
*latency = l;
}
if (ctdb_db->ctdb->tunable.log_latency_ms !=0) {
if (l*1000 > ctdb_db->ctdb->tunable.log_latency_ms) {
DEBUG(DEBUG_WARNING, ("High latency %.6fs for operation %s on database %s\n", l, name, ctdb_db->db_name));
}
}
}
/*
update a reclock latency number
*/
void ctdb_reclock_latency(struct ctdb_context *ctdb, const char *name, double *latency, double l)
{
if (l > *latency) {
*latency = l;
}
if (ctdb->tunable.reclock_latency_ms !=0) {
if (l*1000 > ctdb->tunable.reclock_latency_ms) {
DEBUG(DEBUG_ERR, ("High RECLOCK latency %fs for operation %s\n", l, name));
}
}
}
uint32_t ctdb_reqid_new(struct ctdb_context *ctdb, void *state)
{
int id = idr_get_new_above(ctdb->idr, state, ctdb->lastid+1, INT_MAX);
if (id < 0) {
DEBUG(DEBUG_NOTICE, ("Reqid wrap!\n"));
id = idr_get_new(ctdb->idr, state, INT_MAX);
}
ctdb->lastid = id;
return id;
}
void *_ctdb_reqid_find(struct ctdb_context *ctdb, uint32_t reqid, const char *type, const char *location)
{
void *p;
p = _idr_find_type(ctdb->idr, reqid, type, location);
if (p == NULL) {
DEBUG(DEBUG_WARNING, ("Could not find idr:%u\n",reqid));
}
return p;
}
void ctdb_reqid_remove(struct ctdb_context *ctdb, uint32_t reqid)
{
int ret;
ret = idr_remove(ctdb->idr, reqid);
if (ret != 0) {
DEBUG(DEBUG_ERR, ("Removing idr that does not exist\n"));
}
}
/*
form a ctdb_rec_data record from a key/data pair
note that header may be NULL. If not NULL then it is included in the data portion
of the record
*/
struct ctdb_rec_data *ctdb_marshall_record(TALLOC_CTX *mem_ctx, uint32_t reqid,
TDB_DATA key,
struct ctdb_ltdb_header *header,
TDB_DATA data)
{
size_t length;
struct ctdb_rec_data *d;
length = offsetof(struct ctdb_rec_data, data) + key.dsize +
data.dsize + (header?sizeof(*header):0);
d = (struct ctdb_rec_data *)talloc_size(mem_ctx, length);
if (d == NULL) {
return NULL;
}
d->length = length;
d->reqid = reqid;
d->keylen = key.dsize;
memcpy(&d->data[0], key.dptr, key.dsize);
if (header) {
d->datalen = data.dsize + sizeof(*header);
memcpy(&d->data[key.dsize], header, sizeof(*header));
memcpy(&d->data[key.dsize+sizeof(*header)], data.dptr, data.dsize);
} else {
d->datalen = data.dsize;
memcpy(&d->data[key.dsize], data.dptr, data.dsize);
}
return d;
}
/* helper function for marshalling multiple records */
struct ctdb_marshall_buffer *ctdb_marshall_add(TALLOC_CTX *mem_ctx,
struct ctdb_marshall_buffer *m,
uint64_t db_id,
uint32_t reqid,
TDB_DATA key,
struct ctdb_ltdb_header *header,
TDB_DATA data)
{
struct ctdb_rec_data *r;
size_t m_size, r_size;
struct ctdb_marshall_buffer *m2;
r = ctdb_marshall_record(mem_ctx, reqid, key, header, data);
if (r == NULL) {
talloc_free(m);
return NULL;
}
if (m == NULL) {
m = talloc_zero_size(mem_ctx, offsetof(struct ctdb_marshall_buffer, data));
if (m == NULL) {
return NULL;
}
m->db_id = db_id;
}
m_size = talloc_get_size(m);
r_size = talloc_get_size(r);
m2 = talloc_realloc_size(mem_ctx, m, m_size + r_size);
if (m2 == NULL) {
talloc_free(m);
return NULL;
}
memcpy(m_size + (uint8_t *)m2, r, r_size);
talloc_free(r);
m2->count++;
return m2;
}
/* we've finished marshalling, return a data blob with the marshalled records */
TDB_DATA ctdb_marshall_finish(struct ctdb_marshall_buffer *m)
{
TDB_DATA data;
data.dptr = (uint8_t *)m;
data.dsize = talloc_get_size(m);
return data;
}
/*
loop over a marshalling buffer
- pass r==NULL to start
- loop the number of times indicated by m->count
*/
struct ctdb_rec_data *ctdb_marshall_loop_next(struct ctdb_marshall_buffer *m, struct ctdb_rec_data *r,
uint32_t *reqid,
struct ctdb_ltdb_header *header,
TDB_DATA *key, TDB_DATA *data)
{
if (r == NULL) {
r = (struct ctdb_rec_data *)&m->data[0];
} else {
r = (struct ctdb_rec_data *)(r->length + (uint8_t *)r);
}
if (reqid != NULL) {
*reqid = r->reqid;
}
if (key != NULL) {
key->dptr = &r->data[0];
key->dsize = r->keylen;
}
if (data != NULL) {
data->dptr = &r->data[r->keylen];
data->dsize = r->datalen;
if (header != NULL) {
data->dptr += sizeof(*header);
data->dsize -= sizeof(*header);
}
}
if (header != NULL) {
if (r->datalen < sizeof(*header)) {
return NULL;
}
*header = *(struct ctdb_ltdb_header *)&r->data[r->keylen];
}
return r;
}
/*
if possible, make this task very high priority
*/
void ctdb_high_priority(struct ctdb_context *ctdb)
{
errno = 0;
if (nice(-20) == -1 && errno != 0) {
DEBUG(DEBUG_WARNING,("Unable to renice self: %s\n",
strerror(errno)));
} else {
DEBUG(DEBUG_NOTICE,("Scheduler says I'm nice: %i\n",
getpriority(PRIO_PROCESS, getpid())));
}
}
/*
make ourselves slightly nicer: eg. a ctdb child.
*/
void ctdb_reduce_priority(struct ctdb_context *ctdb)
{
errno = 0;
if (nice(10) == -1 && errno != 0) {
DEBUG(DEBUG_WARNING,("Unable to lower priority: %s\n",
strerror(errno)));
}
}
void set_nonblocking(int fd)
{
unsigned v;
v = fcntl(fd, F_GETFL, 0);
fcntl(fd, F_SETFL, v | O_NONBLOCK);
}
void set_close_on_exec(int fd)
{
unsigned v;
v = fcntl(fd, F_GETFD, 0);
fcntl(fd, F_SETFD, v | FD_CLOEXEC);
}
bool parse_ipv4(const char *s, unsigned port, struct sockaddr_in *sin)
{
sin->sin_family = AF_INET;
sin->sin_port = htons(port);
if (inet_pton(AF_INET, s, &sin->sin_addr) != 1) {
DEBUG(DEBUG_ERR, (__location__ " Failed to translate %s into sin_addr\n", s));
return false;
}
return true;
}
static bool parse_ipv6(const char *s, const char *ifaces, unsigned port, ctdb_sock_addr *saddr)
{
saddr->ip6.sin6_family = AF_INET6;
saddr->ip6.sin6_port = htons(port);
saddr->ip6.sin6_flowinfo = 0;
saddr->ip6.sin6_scope_id = 0;
if (inet_pton(AF_INET6, s, &saddr->ip6.sin6_addr) != 1) {
DEBUG(DEBUG_ERR, (__location__ " Failed to translate %s into sin6_addr\n", s));
return false;
}
if (ifaces && IN6_IS_ADDR_LINKLOCAL(&saddr->ip6.sin6_addr)) {
if (strchr(ifaces, ',')) {
DEBUG(DEBUG_ERR, (__location__ " Link local address %s "
"is specified for multiple ifaces %s\n",
s, ifaces));
return false;
}
saddr->ip6.sin6_scope_id = if_nametoindex(ifaces);
}
return true;
}
/*
parse a ip:port pair
*/
bool parse_ip_port(const char *addr, ctdb_sock_addr *saddr)
{
TALLOC_CTX *tmp_ctx = talloc_new(NULL);
char *s, *p;
unsigned port;
char *endp = NULL;
bool ret;
s = talloc_strdup(tmp_ctx, addr);
if (s == NULL) {
DEBUG(DEBUG_ERR, (__location__ " Failed strdup()\n"));
talloc_free(tmp_ctx);
return false;
}
p = rindex(s, ':');
if (p == NULL) {
DEBUG(DEBUG_ERR, (__location__ " This addr: %s does not contain a port number\n", s));
talloc_free(tmp_ctx);
return false;
}
port = strtoul(p+1, &endp, 10);
if (endp == NULL || *endp != 0) {
/* trailing garbage */
DEBUG(DEBUG_ERR, (__location__ " Trailing garbage after the port in %s\n", s));
talloc_free(tmp_ctx);
return false;
}
*p = 0;
/* now is this a ipv4 or ipv6 address ?*/
ret = parse_ip(s, NULL, port, saddr);
talloc_free(tmp_ctx);
return ret;
}
/*
parse an ip
*/
bool parse_ip(const char *addr, const char *ifaces, unsigned port, ctdb_sock_addr *saddr)
{
char *p;
bool ret;
/* now is this a ipv4 or ipv6 address ?*/
p = index(addr, ':');
if (p == NULL) {
ret = parse_ipv4(addr, port, &saddr->ip);
} else {
ret = parse_ipv6(addr, ifaces, port, saddr);
}
return ret;
}
/*
parse a ip/mask pair
*/
bool parse_ip_mask(const char *str, const char *ifaces, ctdb_sock_addr *addr, unsigned *mask)
{
TALLOC_CTX *tmp_ctx = talloc_new(NULL);
char *s, *p;
char *endp = NULL;
bool ret;
ZERO_STRUCT(*addr);
s = talloc_strdup(tmp_ctx, str);
if (s == NULL) {
DEBUG(DEBUG_ERR, (__location__ " Failed strdup()\n"));
talloc_free(tmp_ctx);
return false;
}
p = rindex(s, '/');
if (p == NULL) {
DEBUG(DEBUG_ERR, (__location__ " This addr: %s does not contain a mask\n", s));
talloc_free(tmp_ctx);
return false;
}
*mask = strtoul(p+1, &endp, 10);
if (endp == NULL || *endp != 0) {
/* trailing garbage */
DEBUG(DEBUG_ERR, (__location__ " Trailing garbage after the mask in %s\n", s));
talloc_free(tmp_ctx);
return false;
}
*p = 0;
/* now is this a ipv4 or ipv6 address ?*/
ret = parse_ip(s, ifaces, 0, addr);
talloc_free(tmp_ctx);
return ret;
}
/*
This is used to canonicalize a ctdb_sock_addr structure.
*/
void ctdb_canonicalize_ip(const ctdb_sock_addr *ip, ctdb_sock_addr *cip)
{
char prefix[12] = { 0,0,0,0,0,0,0,0,0,0,0xff,0xff };
memcpy(cip, ip, sizeof (*cip));
if ( (ip->sa.sa_family == AF_INET6)
&& !memcmp(&ip->ip6.sin6_addr, prefix, 12)) {
memset(cip, 0, sizeof(*cip));
#ifdef HAVE_SOCK_SIN_LEN
cip->ip.sin_len = sizeof(*cip);
#endif
cip->ip.sin_family = AF_INET;
cip->ip.sin_port = ip->ip6.sin6_port;
memcpy(&cip->ip.sin_addr, &ip->ip6.sin6_addr.s6_addr32[3], 4);
}
}
bool ctdb_same_ip(const ctdb_sock_addr *tip1, const ctdb_sock_addr *tip2)
{
ctdb_sock_addr ip1, ip2;
ctdb_canonicalize_ip(tip1, &ip1);
ctdb_canonicalize_ip(tip2, &ip2);
if (ip1.sa.sa_family != ip2.sa.sa_family) {
return false;
}
switch (ip1.sa.sa_family) {
case AF_INET:
return ip1.ip.sin_addr.s_addr == ip2.ip.sin_addr.s_addr;
case AF_INET6:
return !memcmp(&ip1.ip6.sin6_addr.s6_addr[0],
&ip2.ip6.sin6_addr.s6_addr[0],
16);
default:
DEBUG(DEBUG_ERR, (__location__ " CRITICAL Can not compare sockaddr structures of type %u\n", ip1.sa.sa_family));
return false;
}
return true;
}
/*
compare two ctdb_sock_addr structures
*/
bool ctdb_same_sockaddr(const ctdb_sock_addr *ip1, const ctdb_sock_addr *ip2)
{
return ctdb_same_ip(ip1, ip2) && ip1->ip.sin_port == ip2->ip.sin_port;
}
char *ctdb_addr_to_str(ctdb_sock_addr *addr)
{
static char cip[128] = "";
switch (addr->sa.sa_family) {
case AF_INET:
inet_ntop(addr->ip.sin_family, &addr->ip.sin_addr, cip, sizeof(cip));
break;
case AF_INET6:
inet_ntop(addr->ip6.sin6_family, &addr->ip6.sin6_addr, cip, sizeof(cip));
break;
default:
DEBUG(DEBUG_ERR, (__location__ " ERROR, unknown family %u\n", addr->sa.sa_family));
}
return cip;
}
unsigned ctdb_addr_to_port(ctdb_sock_addr *addr)
{
switch (addr->sa.sa_family) {
case AF_INET:
return ntohs(addr->ip.sin_port);
break;
case AF_INET6:
return ntohs(addr->ip6.sin6_port);
break;
default:
DEBUG(DEBUG_ERR, (__location__ " ERROR, unknown family %u\n", addr->sa.sa_family));
}
return 0;
}
void ctdb_block_signal(int signum)
{
sigset_t set;
sigemptyset(&set);
sigaddset(&set,signum);
sigprocmask(SIG_BLOCK,&set,NULL);
}
void ctdb_unblock_signal(int signum)
{
sigset_t set;
sigemptyset(&set);
sigaddset(&set,signum);
sigprocmask(SIG_UNBLOCK,&set,NULL);
}
struct debug_levels debug_levels[] = {
{DEBUG_EMERG, "EMERG"},
{DEBUG_ALERT, "ALERT"},
{DEBUG_CRIT, "CRIT"},
{DEBUG_ERR, "ERR"},
{DEBUG_WARNING, "WARNING"},
{DEBUG_NOTICE, "NOTICE"},
{DEBUG_INFO, "INFO"},
{DEBUG_DEBUG, "DEBUG"},
{0, NULL}
};
const char *get_debug_by_level(int32_t level)
{
int i;
for (i=0; debug_levels[i].description != NULL; i++) {
if (debug_levels[i].level == level) {
return debug_levels[i].description;
}
}
return "Unknown";
}
int32_t get_debug_by_desc(const char *desc)
{
int i;
for (i=0; debug_levels[i].description != NULL; i++) {
if (!strcmp(debug_levels[i].description, desc)) {
return debug_levels[i].level;
}
}
return DEBUG_ERR;
}
/* we don't lock future pages here; it would increase the chance that
* we'd fail to mmap later on. */
void ctdb_lockdown_memory(struct ctdb_context *ctdb)
{
#ifdef HAVE_MLOCKALL
/* Extra stack, please! */
char dummy[10000];
memset(dummy, 0, sizeof(dummy));
if (ctdb->valgrinding) {
return;
}
/* Avoid compiler optimizing out dummy. */
mlock(dummy, sizeof(dummy));
if (mlockall(MCL_CURRENT) != 0) {
DEBUG(DEBUG_WARNING,("Failed to lock memory: %s'\n",
strerror(errno)));
}
#endif
}
const char *ctdb_eventscript_call_names[] = {
"init",
"setup",
"startup",
"startrecovery",
"recovered",
"takeip",
"releaseip",
"stopped",
"monitor",
"status",
"shutdown",
"reload",
"updateip",
"ipreallocated"
};