int
pq_getkeepalivesinterval(Port *port)
{
#ifdef TCP_KEEPINTVL
if (port == NULL || IS_AF_UNIX(port->laddr.addr.ss_family))
return 0;
if (port->keepalives_interval != 0)
return port->keepalives_interval;
if (port->default_keepalives_interval == 0)
{
socklen_t size = sizeof(port->default_keepalives_interval);
if (getsockopt(port->sock, IPPROTO_TCP, TCP_KEEPINTVL,
(char *) &port->default_keepalives_interval,
&size) < 0)
{
elog(LOG, "getsockopt(TCP_KEEPINTVL) failed: %m");
port->default_keepalives_interval = -1; /* don't know */
}
}
return port->default_keepalives_interval;
#else
return 0;
#endif
}
int
pq_getkeepalivescount(Port *port)
{
#ifdef TCP_KEEPCNT
if (port == NULL || IS_AF_UNIX(port->laddr.addr.ss_family))
return 0;
if (port->keepalives_count != 0)
return port->keepalives_count;
if (port->default_keepalives_count == 0)
{
ACCEPT_TYPE_ARG3 size = sizeof(port->default_keepalives_count);
if (getsockopt(port->sock, IPPROTO_TCP, TCP_KEEPCNT,
(char *) &port->default_keepalives_count,
&size) < 0)
{
elog(LOG, "getsockopt(TCP_KEEPCNT) failed: %m");
port->default_keepalives_count = -1; /* don't know */
}
}
return port->default_keepalives_count;
#else
return 0;
#endif
}
int
pq_getkeepalivesinterval(Port *port)
{
#if defined(TCP_KEEPINTVL) || defined(SIO_KEEPALIVE_VALS)
if (port == NULL || IS_AF_UNIX(port->laddr.addr.ss_family))
return 0;
if (port->keepalives_interval != 0)
return port->keepalives_interval;
if (port->default_keepalives_interval == 0)
{
#ifndef WIN32
ACCEPT_TYPE_ARG3 size = sizeof(port->default_keepalives_interval);
if (getsockopt(port->sock, IPPROTO_TCP, TCP_KEEPINTVL,
(char *) &port->default_keepalives_interval,
&size) < 0)
{
elog(LOG, "getsockopt(TCP_KEEPINTVL) failed: %m");
port->default_keepalives_interval = -1; /* don't know */
}
#else
/* We can't get the defaults on Windows, so return "don't know" */
port->default_keepalives_interval = -1;
#endif /* WIN32 */
}
return port->default_keepalives_interval;
#else
return 0;
#endif
}
int
pq_setkeepalivesidle(int idle, Port *port)
{
if (port == NULL || IS_AF_UNIX(port->laddr.addr.ss_family))
return STATUS_OK;
#if defined(TCP_KEEPIDLE) || defined(TCP_KEEPALIVE) || defined(SIO_KEEPALIVE_VALS)
if (idle == port->keepalives_idle)
return STATUS_OK;
#ifndef WIN32
if (port->default_keepalives_idle <= 0)
{
if (pq_getkeepalivesidle(port) < 0)
{
if (idle == 0)
return STATUS_OK; /* default is set but unknown */
else
return STATUS_ERROR;
}
}
if (idle == 0)
idle = port->default_keepalives_idle;
#ifdef TCP_KEEPIDLE
if (setsockopt(port->sock, IPPROTO_TCP, TCP_KEEPIDLE,
(char *) &idle, sizeof(idle)) < 0)
{
elog(LOG, "setsockopt(TCP_KEEPIDLE) failed: %m");
return STATUS_ERROR;
}
#else
if (setsockopt(port->sock, IPPROTO_TCP, TCP_KEEPALIVE,
(char *) &idle, sizeof(idle)) < 0)
{
elog(LOG, "setsockopt(TCP_KEEPALIVE) failed: %m");
return STATUS_ERROR;
}
#endif
port->keepalives_idle = idle;
#else /* WIN32 */
return pq_setkeepaliveswin32(port, idle, port->keepalives_interval);
#endif
#else /* TCP_KEEPIDLE || SIO_KEEPALIVE_VALS */
if (idle != 0)
{
elog(LOG, "setting the keepalive idle time is not supported");
return STATUS_ERROR;
}
#endif
return STATUS_OK;
}
int
pq_setkeepalivesinterval(int interval, Port *port)
{
if (port == NULL || IS_AF_UNIX(port->laddr.addr.ss_family))
return STATUS_OK;
#if defined(TCP_KEEPINTVL) || defined (SIO_KEEPALIVE_VALS)
if (interval == port->keepalives_interval)
return STATUS_OK;
#ifndef WIN32
if (port->default_keepalives_interval <= 0)
{
if (pq_getkeepalivesinterval(port) < 0)
{
if (interval == 0)
return STATUS_OK; /* default is set but unknown */
else
return STATUS_ERROR;
}
}
if (interval == 0)
interval = port->default_keepalives_interval;
if (setsockopt(port->sock, IPPROTO_TCP, TCP_KEEPINTVL,
(char *) &interval, sizeof(interval)) < 0)
{
elog(LOG, "setsockopt(TCP_KEEPINTVL) failed: %m");
return STATUS_ERROR;
}
port->keepalives_interval = interval;
#else /* WIN32 */
return pq_setkeepaliveswin32(port, port->keepalives_idle, interval);
#endif
#else
if (interval != 0)
{
elog(LOG, "setsockopt(TCP_KEEPINTVL) not supported");
return STATUS_ERROR;
}
#endif
return STATUS_OK;
}
int
pq_getkeepalivesidle(Port *port)
{
#if defined(TCP_KEEPIDLE) || defined(TCP_KEEPALIVE) || defined(WIN32)
if (port == NULL || IS_AF_UNIX(port->laddr.addr.ss_family))
return 0;
if (port->keepalives_idle != 0)
return port->keepalives_idle;
if (port->default_keepalives_idle == 0)
{
#ifndef WIN32
ACCEPT_TYPE_ARG3 size = sizeof(port->default_keepalives_idle);
#ifdef TCP_KEEPIDLE
if (getsockopt(port->sock, IPPROTO_TCP, TCP_KEEPIDLE,
(char *) &port->default_keepalives_idle,
&size) < 0)
{
elog(LOG, "getsockopt(TCP_KEEPIDLE) failed: %m");
port->default_keepalives_idle = -1; /* don't know */
}
#else
if (getsockopt(port->sock, IPPROTO_TCP, TCP_KEEPALIVE,
(char *) &port->default_keepalives_idle,
&size) < 0)
{
elog(LOG, "getsockopt(TCP_KEEPALIVE) failed: %m");
port->default_keepalives_idle = -1; /* don't know */
}
#endif /* TCP_KEEPIDLE */
#else /* WIN32 */
/* We can't get the defaults on Windows, so return "don't know" */
port->default_keepalives_idle = -1;
#endif /* WIN32 */
}
return port->default_keepalives_idle;
#else
return 0;
#endif
}
int
pq_setkeepalivescount(int count, Port *port)
{
if (port == NULL || IS_AF_UNIX(port->laddr.addr.ss_family))
return STATUS_OK;
#ifdef TCP_KEEPCNT
if (count == port->keepalives_count)
return STATUS_OK;
if (port->default_keepalives_count <= 0)
{
if (pq_getkeepalivescount(port) < 0)
{
if (count == 0)
return STATUS_OK; /* default is set but unknown */
else
return STATUS_ERROR;
}
}
if (count == 0)
count = port->default_keepalives_count;
if (setsockopt(port->sock, IPPROTO_TCP, TCP_KEEPCNT,
(char *) &count, sizeof(count)) < 0)
{
elog(LOG, "setsockopt(TCP_KEEPCNT) failed: %m");
return STATUS_ERROR;
}
port->keepalives_count = count;
#else
if (count != 0)
{
elog(LOG, "setsockopt(TCP_KEEPCNT) not supported");
return STATUS_ERROR;
}
#endif
return STATUS_OK;
}
int
pq_setkeepalivesidle(int idle, Port *port)
{
if (port == NULL || IS_AF_UNIX(port->laddr.addr.ss_family))
return STATUS_OK;
#ifdef TCP_KEEPIDLE
if (idle == port->keepalives_idle)
return STATUS_OK;
if (port->default_keepalives_idle <= 0)
{
if (pq_getkeepalivesidle(port) < 0)
{
if (idle == 0)
return STATUS_OK; /* default is set but unknown */
else
return STATUS_ERROR;
}
}
if (idle == 0)
idle = port->default_keepalives_idle;
if (setsockopt(port->sock, IPPROTO_TCP, TCP_KEEPIDLE,
(char *) &idle, sizeof(idle)) < 0)
{
elog(LOG, "setsockopt(TCP_KEEPIDLE) failed: %m");
return STATUS_ERROR;
}
port->keepalives_idle = idle;
#else
if (idle != 0)
{
elog(LOG, "setsockopt(TCP_KEEPIDLE) not supported");
return STATUS_ERROR;
}
#endif
return STATUS_OK;
}
/*
* StreamConnection -- create a new connection with client using
* server port. Set port->sock to the FD of the new connection.
*
* ASSUME: that this doesn't need to be non-blocking because
* the Postmaster uses select() to tell when the server master
* socket is ready for accept().
*
* RETURNS: STATUS_OK or STATUS_ERROR
*/
int
StreamConnection(pgsocket server_fd, Port *port)
{
/* accept connection and fill in the client (remote) address */
port->raddr.salen = sizeof(port->raddr.addr);
if ((port->sock = accept(server_fd,
(struct sockaddr *) & port->raddr.addr,
&port->raddr.salen)) < 0)
{
ereport(LOG,
(errcode_for_socket_access(),
errmsg("could not accept new connection: %m")));
/*
* If accept() fails then postmaster.c will still see the server
* socket as read-ready, and will immediately try again. To avoid
* uselessly sucking lots of CPU, delay a bit before trying again.
* (The most likely reason for failure is being out of kernel file
* table slots; we can do little except hope some will get freed up.)
*/
pg_usleep(100000L); /* wait 0.1 sec */
return STATUS_ERROR;
}
#ifdef SCO_ACCEPT_BUG
/*
* UnixWare 7+ and OpenServer 5.0.4 are known to have this bug, but it
* shouldn't hurt to catch it for all versions of those platforms.
*/
if (port->raddr.addr.ss_family == 0)
port->raddr.addr.ss_family = AF_UNIX;
#endif
/* fill in the server (local) address */
port->laddr.salen = sizeof(port->laddr.addr);
if (getsockname(port->sock,
(struct sockaddr *) & port->laddr.addr,
&port->laddr.salen) < 0)
{
elog(LOG, "getsockname() failed: %m");
return STATUS_ERROR;
}
/* select NODELAY and KEEPALIVE options if it's a TCP connection */
if (!IS_AF_UNIX(port->laddr.addr.ss_family))
{
int on;
#ifdef TCP_NODELAY
on = 1;
if (setsockopt(port->sock, IPPROTO_TCP, TCP_NODELAY,
(char *) &on, sizeof(on)) < 0)
{
elog(LOG, "setsockopt(TCP_NODELAY) failed: %m");
return STATUS_ERROR;
}
#endif
on = 1;
if (setsockopt(port->sock, SOL_SOCKET, SO_KEEPALIVE,
(char *) &on, sizeof(on)) < 0)
{
elog(LOG, "setsockopt(SO_KEEPALIVE) failed: %m");
return STATUS_ERROR;
}
#ifdef WIN32
/*
* This is a Win32 socket optimization. The ideal size is 32k.
* http://support.microsoft.com/kb/823764/EN-US/
*/
on = PQ_BUFFER_SIZE * 4;
if (setsockopt(port->sock, SOL_SOCKET, SO_SNDBUF, (char *) &on,
sizeof(on)) < 0)
{
elog(LOG, "setsockopt(SO_SNDBUF) failed: %m");
return STATUS_ERROR;
}
#endif
/*
* Also apply the current keepalive parameters. If we fail to set a
* parameter, don't error out, because these aren't universally
* supported. (Note: you might think we need to reset the GUC
* variables to 0 in such a case, but it's not necessary because the
* show hooks for these variables report the truth anyway.)
*/
(void) pq_setkeepalivesidle(tcp_keepalives_idle, port);
(void) pq_setkeepalivesinterval(tcp_keepalives_interval, port);
(void) pq_setkeepalivescount(tcp_keepalives_count, port);
}
return STATUS_OK;
}
int
StreamServerPort(int family, char *hostName, unsigned short portNumber,
char *unixSocketName,
pgsocket ListenSocket[], int MaxListen)
{
pgsocket fd;
int err;
int maxconn;
int ret;
char portNumberStr[32];
const char *familyDesc;
char familyDescBuf[64];
char *service;
struct addrinfo *addrs = NULL,
*addr;
struct addrinfo hint;
int listen_index = 0;
int added = 0;
#if !defined(WIN32) || defined(IPV6_V6ONLY)
int one = 1;
#endif
/* Initialize hint structure */
MemSet(&hint, 0, sizeof(hint));
hint.ai_family = family;
hint.ai_flags = AI_PASSIVE;
hint.ai_socktype = SOCK_STREAM;
#ifdef HAVE_UNIX_SOCKETS
if (family == AF_UNIX)
{
/* Lock_AF_UNIX will also fill in sock_path. */
if (Lock_AF_UNIX(portNumber, unixSocketName) != STATUS_OK)
return STATUS_ERROR;
service = sock_path;
}
else
#endif /* HAVE_UNIX_SOCKETS */
{
snprintf(portNumberStr, sizeof(portNumberStr), "%d", portNumber);
service = portNumberStr;
}
ret = pg_getaddrinfo_all(hostName, service, &hint, &addrs);
if (ret || !addrs)
{
if (hostName)
ereport(LOG,
(errmsg("could not translate host name \"%s\", service \"%s\" to address: %s",
hostName, service, gai_strerror(ret))));
else
ereport(LOG,
(errmsg("could not translate service \"%s\" to address: %s",
service, gai_strerror(ret))));
if (addrs)
pg_freeaddrinfo_all(hint.ai_family, addrs);
return STATUS_ERROR;
}
for (addr = addrs; addr; addr = addr->ai_next)
{
if (!IS_AF_UNIX(family) && IS_AF_UNIX(addr->ai_family))
{
/*
* Only set up a unix domain socket when they really asked for it.
* The service/port is different in that case.
*/
continue;
}
/* See if there is still room to add 1 more socket. */
for (; listen_index < MaxListen; listen_index++)
{
if (ListenSocket[listen_index] == PGINVALID_SOCKET)
break;
}
if (listen_index >= MaxListen)
{
ereport(LOG,
(errmsg("could not bind to all requested addresses: MAXLISTEN (%d) exceeded",
MaxListen)));
break;
}
/* set up family name for possible error messages */
switch (addr->ai_family)
{
case AF_INET:
familyDesc = _("IPv4");
break;
#ifdef HAVE_IPV6
case AF_INET6:
familyDesc = _("IPv6");
break;
#endif
#ifdef HAVE_UNIX_SOCKETS
case AF_UNIX:
familyDesc = _("Unix");
break;
#endif
default:
snprintf(familyDescBuf, sizeof(familyDescBuf),
_("unrecognized address family %d"),
addr->ai_family);
familyDesc = familyDescBuf;
break;
}
if ((fd = socket(addr->ai_family, SOCK_STREAM, 0)) < 0)
{
ereport(LOG,
(errcode_for_socket_access(),
/* translator: %s is IPv4, IPv6, or Unix */
errmsg("could not create %s socket: %m",
familyDesc)));
continue;
}
#ifndef WIN32
/*
* Without the SO_REUSEADDR flag, a new postmaster can't be started
* right away after a stop or crash, giving "address already in use"
* error on TCP ports.
*
* On win32, however, this behavior only happens if the
* SO_EXLUSIVEADDRUSE is set. With SO_REUSEADDR, win32 allows multiple
* servers to listen on the same address, resulting in unpredictable
* behavior. With no flags at all, win32 behaves as Unix with
* SO_REUSEADDR.
*/
if (!IS_AF_UNIX(addr->ai_family))
{
if ((setsockopt(fd, SOL_SOCKET, SO_REUSEADDR,
(char *) &one, sizeof(one))) == -1)
{
ereport(LOG,
(errcode_for_socket_access(),
errmsg("setsockopt(SO_REUSEADDR) failed: %m")));
closesocket(fd);
continue;
}
}
#endif
#ifdef IPV6_V6ONLY
if (addr->ai_family == AF_INET6)
{
if (setsockopt(fd, IPPROTO_IPV6, IPV6_V6ONLY,
(char *) &one, sizeof(one)) == -1)
{
ereport(LOG,
(errcode_for_socket_access(),
errmsg("setsockopt(IPV6_V6ONLY) failed: %m")));
closesocket(fd);
continue;
}
}
#endif
/*
* Note: This might fail on some OS's, like Linux older than
* 2.4.21-pre3, that don't have the IPV6_V6ONLY socket option, and map
* ipv4 addresses to ipv6. It will show ::ffff:ipv4 for all ipv4
* connections.
*/
err = bind(fd, addr->ai_addr, addr->ai_addrlen);
if (err < 0)
{
ereport(LOG,
(errcode_for_socket_access(),
/* translator: %s is IPv4, IPv6, or Unix */
errmsg("could not bind %s socket: %m",
familyDesc),
(IS_AF_UNIX(addr->ai_family)) ?
errhint("Is another postmaster already running on port %d?"
" If not, remove socket file \"%s\" and retry.",
(int) portNumber, sock_path) :
errhint("Is another postmaster already running on port %d?"
" If not, wait a few seconds and retry.",
(int) portNumber)));
closesocket(fd);
continue;
}
#ifdef HAVE_UNIX_SOCKETS
if (addr->ai_family == AF_UNIX)
{
if (Setup_AF_UNIX() != STATUS_OK)
{
closesocket(fd);
break;
}
}
#endif
/*
* Select appropriate accept-queue length limit. PG_SOMAXCONN is only
* intended to provide a clamp on the request on platforms where an
* overly large request provokes a kernel error (are there any?).
*/
maxconn = MaxBackends * 2;
if (maxconn > PG_SOMAXCONN)
maxconn = PG_SOMAXCONN;
err = listen(fd, maxconn);
if (err < 0)
{
ereport(LOG,
(errcode_for_socket_access(),
/* translator: %s is IPv4, IPv6, or Unix */
errmsg("could not listen on %s socket: %m",
familyDesc)));
closesocket(fd);
continue;
}
ListenSocket[listen_index] = fd;
added++;
}
pg_freeaddrinfo_all(hint.ai_family, addrs);
if (!added)
return STATUS_ERROR;
return STATUS_OK;
}
/* ----------------
* GTMPQconnectPoll
*
* Poll an asynchronous connection.
*
* Returns a GTMClientPollingStatusType.
* Before calling this function, use select(2) to determine when data
* has arrived..
*
* You must call GTMPQfinish whether or not this fails.
*/
GTMClientPollingStatusType
GTMPQconnectPoll(GTM_Conn *conn)
{
if (conn == NULL)
return PGRES_POLLING_FAILED;
/* Get the new data */
switch (conn->status)
{
/*
* We really shouldn't have been polled in these two cases, but we
* can handle it.
*/
case CONNECTION_BAD:
return PGRES_POLLING_FAILED;
case CONNECTION_OK:
return PGRES_POLLING_OK;
/* These are reading states */
case CONNECTION_AWAITING_RESPONSE:
case CONNECTION_AUTH_OK:
{
/* Load waiting data */
int n = gtmpqReadData(conn);
if (n < 0)
goto error_return;
if (n == 0)
return PGRES_POLLING_READING;
break;
}
/* These are writing states, so we just proceed. */
case CONNECTION_STARTED:
case CONNECTION_MADE:
break;
case CONNECTION_NEEDED:
break;
default:
appendGTMPQExpBuffer(&conn->errorMessage,
"invalid connection state, "
"probably indicative of memory corruption\n"
);
goto error_return;
}
keep_going: /* We will come back to here until there is
* nothing left to do. */
switch (conn->status)
{
case CONNECTION_NEEDED:
{
/*
* Try to initiate a connection to one of the addresses
* returned by gtm_getaddrinfo_all(). conn->addr_cur is the
* next one to try. We fail when we run out of addresses
* (reporting the error returned for the *last* alternative,
* which may not be what users expect :-().
*/
while (conn->addr_cur != NULL)
{
struct addrinfo *addr_cur = conn->addr_cur;
/* Remember current address for possible error msg */
memcpy(&conn->raddr.addr, addr_cur->ai_addr,
addr_cur->ai_addrlen);
conn->raddr.salen = addr_cur->ai_addrlen;
/* Open a socket */
conn->sock = socket(addr_cur->ai_family, SOCK_STREAM, 0);
if (conn->sock < 0)
{
/*
* ignore socket() failure if we have more addresses
* to try
*/
if (addr_cur->ai_next != NULL)
{
conn->addr_cur = addr_cur->ai_next;
continue;
}
appendGTMPQExpBuffer(&conn->errorMessage,
"could not create socket: \n");
break;
}
/*
* Select socket options: no delay of outgoing data for
* TCP sockets, nonblock mode, close-on-exec. Fail if any
* of this fails.
*/
if (!IS_AF_UNIX(addr_cur->ai_family))
{
if (!connectNoDelay(conn))
{
close(conn->sock);
conn->sock = -1;
conn->addr_cur = addr_cur->ai_next;
continue;
}
}
/*
* Start/make connection. This should not block, since we
* are in nonblock mode. If it does, well, too bad.
*/
if (connect(conn->sock, addr_cur->ai_addr,
addr_cur->ai_addrlen) < 0)
{
if (SOCK_ERRNO == EINPROGRESS ||
SOCK_ERRNO == EWOULDBLOCK ||
SOCK_ERRNO == EINTR ||
SOCK_ERRNO == 0)
{
/*
* This is fine - we're in non-blocking mode, and
* the connection is in progress. Tell caller to
* wait for write-ready on socket.
*/
conn->status = CONNECTION_STARTED;
return PGRES_POLLING_WRITING;
}
/* otherwise, trouble */
}
else
{
/*
* Hm, we're connected already --- seems the "nonblock
* connection" wasn't. Advance the state machine and
* go do the next stuff.
*/
conn->status = CONNECTION_STARTED;
goto keep_going;
}
/*
* This connection failed --- set up error report, then
* close socket (do it this way in case close() affects
* the value of errno...). We will ignore the connect()
* failure and keep going if there are more addresses.
*/
connectFailureMessage(conn, SOCK_ERRNO);
if (conn->sock >= 0)
{
close(conn->sock);
conn->sock = -1;
}
/*
* Try the next address, if any.
*/
conn->addr_cur = addr_cur->ai_next;
} /* loop over addresses */
/*
* Ooops, no more addresses. An appropriate error message is
* already set up, so just set the right status.
*/
goto error_return;
}
case CONNECTION_STARTED:
{
int optval;
size_t optlen = sizeof(optval);
/*
* Write ready, since we've made it here, so the connection
* has been made ... or has failed.
*/
/*
* Now check (using getsockopt) that there is not an error
* state waiting for us on the socket.
*/
if (getsockopt(conn->sock, SOL_SOCKET, SO_ERROR,
(char *) &optval, (socklen_t *)&optlen) == -1)
{
appendGTMPQExpBuffer(&conn->errorMessage,
libpq_gettext("could not get socket error status: \n"));
goto error_return;
}
else if (optval != 0)
{
/*
* When using a nonblocking connect, we will typically see
* connect failures at this point, so provide a friendly
* error message.
*/
connectFailureMessage(conn, optval);
/*
* If more addresses remain, keep trying, just as in the
* case where connect() returned failure immediately.
*/
if (conn->addr_cur->ai_next != NULL)
{
if (conn->sock >= 0)
{
close(conn->sock);
conn->sock = -1;
}
conn->addr_cur = conn->addr_cur->ai_next;
conn->status = CONNECTION_NEEDED;
goto keep_going;
}
goto error_return;
}
/* Fill in the client address */
conn->laddr.salen = sizeof(conn->laddr.addr);
if (getsockname(conn->sock,
(struct sockaddr *) & conn->laddr.addr,
(socklen_t *)&conn->laddr.salen) < 0)
{
appendGTMPQExpBuffer(&conn->errorMessage,
"could not get client address from socket:\n");
goto error_return;
}
/*
* Make sure we can write before advancing to next step.
*/
conn->status = CONNECTION_MADE;
return PGRES_POLLING_WRITING;
}
case CONNECTION_MADE:
{
GTM_StartupPacket sp;
/*
* Build a startup packet. We tell the GTM server/proxy our
* PGXC Node name and whether we are a proxy or not.
*
* When the connection is made from the proxy, we let the GTM
* server know about it so that some special headers are
* handled correctly by the server.
*/
strcpy(sp.sp_node_name, conn->gc_node_name);
sp.sp_remotetype = conn->remote_type;
sp.sp_ispostmaster = conn->is_postmaster;
/*
* Send the startup packet.
*
* Theoretically, this could block, but it really shouldn't
* since we only got here if the socket is write-ready.
*/
if (pqPacketSend(conn, 'A', &sp, sizeof (GTM_StartupPacket)) != STATUS_OK)
{
appendGTMPQExpBuffer(&conn->errorMessage,
"could not send startup packet: \n");
goto error_return;
}
conn->status = CONNECTION_AWAITING_RESPONSE;
return PGRES_POLLING_READING;
}
/*
* Handle authentication exchange: wait for postmaster messages
* and respond as necessary.
*/
case CONNECTION_AWAITING_RESPONSE:
{
char beresp;
/*
* Scan the message from current point (note that if we find
* the message is incomplete, we will return without advancing
* inStart, and resume here next time).
*/
conn->inCursor = conn->inStart;
/* Read type byte */
if (gtmpqGetc(&beresp, conn))
{
/* We'll come back when there is more data */
return PGRES_POLLING_READING;
}
/*
* Validate message type: we expect only an authentication
* request or an error here. Anything else probably means
* it's not GTM on the other end at all.
*/
if (!(beresp == 'R' || beresp == 'E'))
{
appendGTMPQExpBuffer(&conn->errorMessage,
"expected authentication request from "
"server, but received %c\n",
beresp);
goto error_return;
}
/* Handle errors. */
if (beresp == 'E')
{
if (gtmpqGets_append(&conn->errorMessage, conn))
{
/* We'll come back when there is more data */
return PGRES_POLLING_READING;
}
/* OK, we read the message; mark data consumed */
conn->inStart = conn->inCursor;
goto error_return;
}
{
/*
* Server sends a dummy message body of size 4 bytes
*/
int tmp_int;
gtmpqGetInt(&tmp_int, 4, conn);
}
/*
* OK, we successfully read the message; mark data consumed
*/
conn->inStart = conn->inCursor;
/* We are done with authentication exchange */
conn->status = CONNECTION_AUTH_OK;
/* Look to see if we have more data yet. */
goto keep_going;
}
case CONNECTION_AUTH_OK:
{
/* We can release the address list now. */
gtm_freeaddrinfo_all(conn->addrlist_family, conn->addrlist);
conn->addrlist = NULL;
conn->addr_cur = NULL;
/* Otherwise, we are open for business! */
conn->status = CONNECTION_OK;
return PGRES_POLLING_OK;
}
default:
appendGTMPQExpBuffer(&conn->errorMessage,
"invalid connection state %c, "
"probably indicative of memory corruption\n"
,
conn->status);
goto error_return;
}
/* Unreachable */
error_return:
/*
* We used to close the socket at this point, but that makes it awkward
* for those above us if they wish to remove this socket from their own
* records (an fd_set for example). We'll just have this socket closed
* when GTMPQfinish is called (which is compulsory even after an error, since
* the connection structure must be freed).
*/
conn->status = CONNECTION_BAD;
return PGRES_POLLING_FAILED;
}
/*
* StreamConnection -- create a new connection with client using
* server port. Set port->sock to the FD of the new connection.
*
* ASSUME: that this doesn't need to be non-blocking because
* the Postmaster uses select() to tell when the server master
* socket is ready for accept().
*
* RETURNS: STATUS_OK or STATUS_ERROR
*/
int
StreamConnection(pgsocket server_fd, Port *port)
{
/* accept connection and fill in the client (remote) address */
port->raddr.salen = sizeof(port->raddr.addr);
if ((port->sock = accept(server_fd,
(struct sockaddr *) & port->raddr.addr,
&port->raddr.salen)) == PGINVALID_SOCKET)
{
ereport(LOG,
(errcode_for_socket_access(),
errmsg("could not accept new connection: %m")));
/*
* If accept() fails then postmaster.c will still see the server
* socket as read-ready, and will immediately try again. To avoid
* uselessly sucking lots of CPU, delay a bit before trying again.
* (The most likely reason for failure is being out of kernel file
* table slots; we can do little except hope some will get freed up.)
*/
pg_usleep(100000L); /* wait 0.1 sec */
return STATUS_ERROR;
}
/* fill in the server (local) address */
port->laddr.salen = sizeof(port->laddr.addr);
if (getsockname(port->sock,
(struct sockaddr *) & port->laddr.addr,
&port->laddr.salen) < 0)
{
elog(LOG, "getsockname() failed: %m");
return STATUS_ERROR;
}
/* select NODELAY and KEEPALIVE options if it's a TCP connection */
if (!IS_AF_UNIX(port->laddr.addr.ss_family))
{
int on;
#ifdef WIN32
int oldopt;
int optlen;
int newopt;
#endif
#ifdef TCP_NODELAY
on = 1;
if (setsockopt(port->sock, IPPROTO_TCP, TCP_NODELAY,
(char *) &on, sizeof(on)) < 0)
{
elog(LOG, "setsockopt(TCP_NODELAY) failed: %m");
return STATUS_ERROR;
}
#endif
on = 1;
if (setsockopt(port->sock, SOL_SOCKET, SO_KEEPALIVE,
(char *) &on, sizeof(on)) < 0)
{
elog(LOG, "setsockopt(SO_KEEPALIVE) failed: %m");
return STATUS_ERROR;
}
#ifdef WIN32
/*
* This is a Win32 socket optimization. The OS send buffer should be
* large enough to send the whole Postgres send buffer in one go, or
* performance suffers. The Postgres send buffer can be enlarged if a
* very large message needs to be sent, but we won't attempt to
* enlarge the OS buffer if that happens, so somewhat arbitrarily
* ensure that the OS buffer is at least PQ_SEND_BUFFER_SIZE * 4.
* (That's 32kB with the current default).
*
* The default OS buffer size used to be 8kB in earlier Windows
* versions, but was raised to 64kB in Windows 2012. So it shouldn't
* be necessary to change it in later versions anymore. Changing it
* unnecessarily can even reduce performance, because setting
* SO_SNDBUF in the application disables the "dynamic send buffering"
* feature that was introduced in Windows 7. So before fiddling with
* SO_SNDBUF, check if the current buffer size is already large enough
* and only increase it if necessary.
*
* See https://support.microsoft.com/kb/823764/EN-US/ and
* https://msdn.microsoft.com/en-us/library/bb736549%28v=vs.85%29.aspx
*/
optlen = sizeof(oldopt);
if (getsockopt(port->sock, SOL_SOCKET, SO_SNDBUF, (char *) &oldopt,
&optlen) < 0)
{
elog(LOG, "getsockopt(SO_SNDBUF) failed: %m");
return STATUS_ERROR;
}
newopt = PQ_SEND_BUFFER_SIZE * 4;
if (oldopt < newopt)
{
if (setsockopt(port->sock, SOL_SOCKET, SO_SNDBUF, (char *) &newopt,
sizeof(newopt)) < 0)
{
elog(LOG, "setsockopt(SO_SNDBUF) failed: %m");
return STATUS_ERROR;
}
}
#endif
/*
* Also apply the current keepalive parameters. If we fail to set a
* parameter, don't error out, because these aren't universally
* supported. (Note: you might think we need to reset the GUC
* variables to 0 in such a case, but it's not necessary because the
* show hooks for these variables report the truth anyway.)
*/
(void) pq_setkeepalivesidle(tcp_keepalives_idle, port);
(void) pq_setkeepalivesinterval(tcp_keepalives_interval, port);
(void) pq_setkeepalivescount(tcp_keepalives_count, port);
}
return STATUS_OK;
}
void body()
{
/* Initialize hint structure */
#ifdef HAVE_UNIX_SOCKETS
if (family == AF_UNIX)
{
/* Lock_AF_UNIX will also fill in sock_path. */
/* if (Lock_AF_UNIX(portNumber, unixSocketName) != STATUS_OK) */
/* return STATUS_ERROR; */
service = sock_path;
}
else
#endif /* HAVE_UNIX_SOCKETS */
{
snprintf(1, sizeof(1), "%d", portNumber);
service = 1;
}
ret = getaddrinfo_all(hostName, service, &hint, &addrs);
if (ret || !addrs)
{
if (hostName) {
/* ereport(LOG, */
/* (errmsg("could not translate host name \"%s\", service \"%s\" to address: %s", */
/* hostName, service, gai_strerror(ret)))); */
} else {
/* ereport(LOG, */
/* (errmsg("could not translate service \"%s\" to address: %s", */
/* service, gai_strerror(ret)))); */
/* if (addrs) */
/* freeaddrinfo_all(hint.ai_family, addrs); */
}
//ret = STATUS_ERROR;
//while(1) { int qqq; qqq=qqq;}
}
for (addr = addrs; addr < MAXADDR; addr++) // = addr->ai_next)
{
if (!IS_AF_UNIX(family) && IS_AF_UNIX(addr_ai_family))
{
/*
* Only set up a unix domain socket when they really asked for
* it. The service/port is different in that case.
*/
goto loc_continue;
}
/* See if there is still room to add 1 more socket. */
for (; listen_index < MaxListen; listen_index++)
{
if (ListenSocket_OF_listen_index == -1)
break;
}
if (listen_index >= MaxListen)
{
/* ereport(LOG, */
/* (errmsg("could not bind to all requested addresses: MAXLISTEN (%d) exceeded", */
/* MaxListen))); */
break;
}
/* set up family name for possible error messages */
if(addr_ai_family==AF_INET) {
gettext("IPv4");
} else if(addr_ai_family==AF_INET6) {
gettext("IPv6");
} else if(addr_ai_family==AF_UNIX) {
gettext("Unix");
} else {
snprintf(1,/*familyDescBuf*/ 1, //sizeof(familyDescBuf),
gettext("unrecognized address family %d"),
addr_ai_family);
//familyDesc = familyDescBuf;
}
if ((fd = nondet())) // socket(addr_ai_family, SOCK_STREAM, 0)) < 0)
{
/* ereport(LOG, */
/* (errcode_for_socket_access(), */
/* /\* translator: %s is IPv4, IPv6, or Unix *\/ */
/* errmsg("could not create %s socket: %m", */
/* familyDesc))); */
goto loc_continue;
}
if (nondet()) // !IS_AF_UNIX(addr_ai_family))
{
if (nondet()) //(setsockopt(fd, SOL_SOCKET, SO_REUSEADDR,
// (char *) &one, sizeof(one))) == -1)
{
/* ereport(LOG, */
/* (errcode_for_socket_access(), */
/* errmsg("setsockopt(SO_REUSEADDR) failed: %m"))); */
/* closesocket(fd); */
goto loc_continue;
}
}
#ifdef IPV6_V6ONLY
if (nondet()) // addr_ai_family == AF_INET6)
{
if (nondet()) // setsockopt(fd, IPPROTO_IPV6, IPV6_V6ONLY,
// (char *) &one, sizeof(one)) == -1)
{
/* ereport(LOG, */
/* (errcode_for_socket_access(), */
/* errmsg("setsockopt(IPV6_V6ONLY) failed: %m"))); */
closesocket(fd);
goto loc_continue;
}
}
#endif
/*
* Note: This might fail on some OS's, like Linux older than
* 2.4.21-pre3, that don't have the IPV6_V6ONLY socket option, and
* map ipv4 addresses to ipv6. It will show ::ffff:ipv4 for all
* ipv4 connections.
*/
err = nondet(); // bind(fd, addr->ai_addr, addr->ai_addrlen);
if (err < 0)
{
/* ereport(LOG, */
/* (errcode_for_socket_access(), */
/* /\* translator: %s is IPv4, IPv6, or Unix *\/ */
/* errmsg("could not bind %s socket: %m", */
/* familyDesc), */
/* (IS_AF_UNIX(addr_ai_family)) ? */
/* errhint("Is another postmaster already running on port %d?" */
/* " If not, remove socket file \"%s\" and retry.", */
/* (int) portNumber, sock_path) : */
/* errhint("Is another postmaster already running on port %d?" */
/* " If not, wait a few seconds and retry.", */
/* (int) portNumber))); */
closesocket(fd);
goto loc_continue;
}
#ifdef HAVE_UNIX_SOCKETS
if (addr_ai_family == AF_UNIX)
{
if (nondet() != STATUS_OK)
{
closesocket(fd);
break;
}
}
#endif
/*
* Select appropriate accept-queue length limit. PG_SOMAXCONN is
* only intended to provide a clamp on the request on platforms
* where an overly large request provokes a kernel error (are
* there any?).
*/
maxconn = MaxBackends * 2;
if (maxconn > PG_SOMAXCONN)
maxconn = PG_SOMAXCONN;
err = listen(fd, maxconn);
if (err < 0)
{
closesocket(fd);
goto loc_continue;
}
ListenSocket_OF_listen_index = fd;
added++;
if(1) { loc_continue: 0; }
}
//freeaddrinfo_all(hint.ai_family, addrs);
if (!added) {
ret = STATUS_ERROR;
while(1) { int rrr; rrr=rrr; }
}
ret = STATUS_OK;
while(1) { int ddd; ddd=ddd; }
}
/*
* StreamConnection -- create a new connection with client using
* server port.
*
* ASSUME: that this doesn't need to be non-blocking because
* the Postmaster uses select() to tell when the server master
* socket is ready for accept().
*
* RETURNS: STATUS_OK or STATUS_ERROR
*/
int
StreamConnection(int server_fd, Port *port)
{
/* accept connection and fill in the client (remote) address */
port->raddr.salen = sizeof(port->raddr.addr);
if ((port->sock = accept(server_fd,
(struct sockaddr *) & port->raddr.addr,
&port->raddr.salen)) < 0)
{
ereport(LOG,
(errcode_for_socket_access(),
errmsg("could not accept new connection: %m")));
return STATUS_ERROR;
}
#ifdef SCO_ACCEPT_BUG
/*
* UnixWare 7+ and OpenServer 5.0.4 are known to have this bug, but it
* shouldn't hurt to catch it for all versions of those platforms.
*/
if (port->raddr.addr.ss_family == 0)
port->raddr.addr.ss_family = AF_UNIX;
#endif
/* fill in the server (local) address */
port->laddr.salen = sizeof(port->laddr.addr);
if (getsockname(port->sock,
(struct sockaddr *) & port->laddr.addr,
&port->laddr.salen) < 0)
{
elog(LOG, "getsockname() failed: %m");
return STATUS_ERROR;
}
/* select NODELAY and KEEPALIVE options if it's a TCP connection */
if (!IS_AF_UNIX(port->laddr.addr.ss_family))
{
int on;
#ifdef TCP_NODELAY
on = 1;
if (setsockopt(port->sock, IPPROTO_TCP, TCP_NODELAY,
(char *) &on, sizeof(on)) < 0)
{
elog(LOG, "setsockopt(TCP_NODELAY) failed: %m");
return STATUS_ERROR;
}
#endif
on = 1;
if (setsockopt(port->sock, SOL_SOCKET, SO_KEEPALIVE,
(char *) &on, sizeof(on)) < 0)
{
elog(LOG, "setsockopt(SO_KEEPALIVE) failed: %m");
return STATUS_ERROR;
}
/*
* Also apply the current keepalive parameters. If we fail to set a
* parameter, don't error out, because these aren't universally
* supported. (Note: you might think we need to reset the GUC
* variables to 0 in such a case, but it's not necessary because the
* show hooks for these variables report the truth anyway.)
*/
(void) pq_setkeepalivesidle(tcp_keepalives_idle, port);
(void) pq_setkeepalivesinterval(tcp_keepalives_interval, port);
(void) pq_setkeepalivescount(tcp_keepalives_count, port);
}
return STATUS_OK;
}
