On Unix-ish platforms, EWOULDBLOCK may be the same as EAGAIN, which is
*not* a success return, at least not on Linux. We need to treat it as a
failure to avoid giving a misleading error message. Per the Single Unix
Spec, only EINPROGRESS and EINTR returns indicate that the connection
attempt is in progress.
On Windows, on the other hand, EWOULDBLOCK (WSAEWOULDBLOCK) is the expected
case. We must accept EINPROGRESS as well because Cygwin will return that,
and it doesn't seem worth distinguishing Cygwin from native Windows here.
It's not very clear whether EINTR can occur on Windows, but let's leave
that part of the logic alone in the absence of concrete trouble reports.
Also, remove the test for errno == 0, effectively reverting commit
da9501bddb, which AFAICS was just a thinko;
or at best it might have been a workaround for a platform-specific bug,
which we can hope is gone now thirteen years later. In any case, since
libpq makes no effort to reset errno to zero before calling connect(),
it seems unlikely that that test has ever reliably done anything useful.
Andres Freund and Tom Lane
The length of a socket path name is constrained by the size of struct
sockaddr_un, and there's not a lot we can do about it since that is a
kernel API. However, it would be a good thing if we produced an
intelligible error message when the user specifies a socket path that's too
long --- and getaddrinfo's standard API is too impoverished to do this in
the natural way. So insert explicit tests at the places where we construct
a socket path name. Now you'll get an error that makes sense and even
tells you what the limit is, rather than something generic like
"Non-recoverable failure in name resolution".
Per trouble report from Jeremy Drake and a fix idea from Andrew Dunstan.
These days, such a response is far more likely to signify a server-side
problem, such as fork failure. Reporting "server does not support SSL"
(in sslmode=require) could be quite misleading. But the results could
be even worse in sslmode=prefer: if the problem was transient and the
next connection attempt succeeds, we'll have silently fallen back to
protocol version 2.0, possibly disabling features the user needs.
Hence, it seems best to just eliminate the assumption that backing off
to non-SSL/2.0 protocol is the way to recover from an "E" response, and
instead treat the server error the same as we would in non-SSL cases.
I tested this change against a pre-7.0 server, and found that there
was a second logic bug in the "prefer" path: the test to decide whether
to make a fallback connection attempt assumed that we must have opened
conn->ssl, which in fact does not happen given an "E" response. After
fixing that, the code does indeed connect successfully to pre-7.0,
as long as you didn't set sslmode=require. (If you did, you get
"Unsupported frontend protocol", which isn't completely off base
given the server certainly doesn't support SSL.)
Since there seems no reason to believe that pre-7.0 servers exist anymore
in the wild, back-patch to all supported branches.
There are assorted situations wherein PQconnectPoll() will abandon a
connection attempt and try again with different parameters (eg, SSL versus
not SSL). However, the code forgot to discard any pending data in libpq's
I/O buffers when doing this. In at least one case (server returns E
message during SSL negotiation), there is unread input data which bollixes
the next connection attempt. I have not checked to see whether this is
possible in the other cases where we close the socket and retry, but it
seems like a matter of good defensive programming to add explicit
buffer-flushing code to all of them.
This is one of several issues exposed by Daniel Farina's report of
misbehavior after a server-side fork failure.
This has been wrong since forever, so back-patch to all supported branches.
The code to assemble ldap_get_values_len's output into a single string
wrote the terminating null one byte past where it should. Fix that,
and make some other cosmetic adjustments to make the code a trifle more
readable and more in line with usual Postgres coding style.
Also, free the "result" string when done with it, to avoid a permanent
memory leak.
Bug report and patch by Albe Laurenz, cosmetic adjustments by me.
This would lead to leaking the PGconn structure after an error detected by
conninfo_array_parse(), as well as failing to return a useful error message
in such cases. Backpatch to 9.0 where the error was introduced.
Joseph Adams
This adds four additional connection parameters to libpq: keepalives,
keepalives_idle, keepalives_count, and keepalives_interval.
keepalives default to on, per discussion, but can be turned off by
specifying keepalives=0. The remaining parameters, where supported,
can be used to adjust how often keepalives are sent and how many
can be lost before the connection is broken.
The immediate motivation for this patch is to make sure that
walreceiver will eventually notice if the master reboots without
closing the connection cleanly, but it should be helpful in other
cases as well.
Tollef Fog Heen, Fujii Masao, and me.
additional cases correctly. The original coding failed to load additional
(chain) certificates from the client cert file, meaning that indirectly signed
client certificates didn't work unless one hacked the server's root.crt file
to include intermediate CAs (not the desired approach). Another problem was
that everything got loaded into the shared SSL_context object, which meant
that concurrent connections trying to use different sslcert settings could
well fail due to conflicting over the single available slot for a keyed
certificate.
To fix, get rid of the use of SSL_CTX_set_client_cert_cb(), which is
deprecated anyway in the OpenSSL documentation, and instead just
unconditionally load the client cert and private key during connection
initialization. This lets us use SSL_CTX_use_certificate_chain_file(),
which does the right thing with additional certs, and is lots simpler than
the previous hacking about with BIO-level access. A small disadvantage is
that we have to load the primary client cert a second time with
SSL_use_certificate_file, so that that one ends up in the correct slot
within the connection's SSL object where it can get paired with the key.
Given the other overhead of making an SSL connection, that doesn't seem
worth worrying about.
Per discussion ensuing from bug #5468.
set ferror() but never set feof(). This is known to be the case for recent
glibc when trying to read a directory as a file, and might be true for other
platforms/cases too. Per report from Ed L. (There is more that we ought to
do about his report, but this is one easily identifiable issue.)
If expand_dbname is non-zero and dbname contains an = sign, it is taken as
a conninfo string in exactly the same way as if it had been passed to
PQconnectdb. This is equivalent to the way PQsetdbLogin() works, allowing
PQconnectdbParams() to be a complete alternative.
Also improve the way the new function is called from psql and replace a
previously missed call to PQsetdbLogin() in psql. Additionally use
PQconnectdbParams() for pg_dump and friends, and the bin/scripts
command line utilities such as vacuumdb, createdb, etc.
Finally, update the documentation for the new parameter, as well as the
nuances of precedence in cases where key words are repeated or duplicated
in the conninfo string.
PQconnectStartParams. These are analogous to PQconnectdb and PQconnectStart
respectively. They differ from the legacy functions in that they accept
two NULL-terminated arrays, keywords and values, rather than conninfo
strings. This avoids the need to build the conninfo string in cases
where it might be inconvenient to do so. Includes documentation.
Also modify psql to utilize PQconnectdbParams rather than PQsetdbLogin.
This allows the new config parameter application_name to be set, which
in turn is displayed in the pg_stat_activity view and included in CSV
log entries. This will also ensure both new functions get regularly
exercised.
Patch by Guillaume Lelarge with review and minor adjustments by
Joe Conway.
This includes two new kinds of postmaster processes, walsenders and
walreceiver. Walreceiver is responsible for connecting to the primary server
and streaming WAL to disk, while walsender runs in the primary server and
streams WAL from disk to the client.
Documentation still needs work, but the basics are there. We will probably
pull the replication section to a new chapter later on, as well as the
sections describing file-based replication. But let's do that as a separate
patch, so that it's easier to see what has been added/changed. This patch
also adds a new section to the chapter about FE/BE protocol, documenting the
protocol used by walsender/walreceivxer.
Bump catalog version because of two new functions,
pg_last_xlog_receive_location() and pg_last_xlog_replay_location(), for
monitoring the progress of replication.
Fujii Masao, with additional hacking by me
is made, include it in the startup-packet options. This makes it work more
like every other libpq connection option, in particular it now has the same
response to RESET ALL as the rest. This also saves one network round trip
for new applications using application_name. The cost is that if the server
is pre-8.5, it'll reject the startup packet altogether, forcing us to retry
the entire connection cycle. But on balance we shouldn't be optimizing that
case in preference to the behavior with a new server, especially when doing
so creates visible behavioral oddities. Per discussion.
sockopt(SO_NOSIGPIPE) or the MSG_NOSIGNAL flag to send().
We assume these features are available if (1) the symbol is defined at
compile time and (2) the kernel doesn't reject the call at runtime.
It might turn out that there are some platforms where (1) and (2) are
true and yet the signal isn't really blocked, in which case applications
would die on server crash. If that sort of thing gets reported, then
we'll have to add additional defenses of some kind.
Jeremy Kerr
"verify-ca" and "verify-full".
Since "prefer" remains the default, this will make certificate validation
off by default, which should lead to less upgrade issues.
and certificate revokation list by using connection parameters or environment
variables.
Original patch by Mark Woodward, heavily reworked by Alvaro Herrera and
Magnus Hagander.
results (ie, an empty "broken" buffer) if memory overrun occurs anywhere
along the way to filling the buffer. The previous coding would just silently
discard portions of the intended buffer contents, as exhibited in trouble
report from Sam Mason. Also, tweak psql's main loop to correctly detect
and report such overruns. There's probably much more that should be done
in this line, but this is a start.
libpq. As noted by Peter, adding this variable created a risk of unexpected
connection failures when talking to older server versions, and since it
doesn't do anything you can't do with PGOPTIONS, it doesn't seem really
necessary. Removing it does occasion a few extra lines in pg_regress.c,
but saving a getenv() call per libpq connection attempt is perhaps worth
that anyway.
from DateStyle, and create a new interval style that produces output matching
the SQL standard (at least for interval values that fall within the standard's
restrictions). IntervalStyle is also used to resolve the conflict between the
standard and traditional Postgres rules for interpreting negative interval
input.
Ron Mayer
after each other (since we already add a newline on each, this makes them
multiline).
Previously a new error would just overwrite the old one, so for example any
error caused when trying to connect with SSL enabled would be overwritten
by the error message form the non-SSL connection when using sslmode=prefer.
that presence of the password in the conninfo string must be checked *before*
risking a connection attempt, there is no point in checking it afterwards.
This makes the specification of PQconnectionUsedPassword() a bit simpler
and perhaps more generally useful, too.
conninfo string *before* trying to connect to the remote server, not after.
As pointed out by Marko Kreen, in certain not-very-plausible situations
this could result in sending a password from the postgres user's .pgpass file,
or other places that non-superusers shouldn't have access to, to an
untrustworthy remote server. The cleanest fix seems to be to expose libpq's
conninfo-string-parsing code so that dblink can check for a password option
without duplicating the parsing logic.
Joe Conway, with a little cleanup by Tom Lane
we are on a 64-bit machine (ie, size_t is wider than int) and someone passes
in a query string that approaches or exceeds INT_MAX bytes. Also, just for
paranoia's sake, guard against similar overflows in sizing the input buffer.
The backend will not in the foreseeable future be prepared to send or receive
strings exceeding 1GB, so I didn't take the more invasive step of switching
all the buffer index variables from int to size_t; though someday we might
want to do that.
I have a suspicion that this is not the only such bug in libpq, but this
fix is enough to take care of the crash reported by Francisco Reyes.
should always succeed, but in the likely event of a failure we would
previously fall through *without locking* - the new code will exit(1).
Printing the error message on stderr will not work for all applications, but
it's better than nothing at all - and our API doesn't provide a way to return
the error to the caller.
key files that are similar to the one for the postmaster's data directory
permissions check. (I chose to standardize on that one since it's the most
heavily used and presumably best-wordsmithed by now.) Also eliminate explicit
tests on file ownership in these places, since the ensuing read attempt must
fail anyway if it's wrong, and there seems no value in issuing the same error
message for distinct problems. (But I left in the explicit ownership test in
postmaster.c, since it had its own error message anyway.) Also be more
specific in the documentation's descriptions of these checks. Per a gripe
from Kevin Hunter.
the patch for those features put its cleanup code into freePGconn() which is
really the wrong place. Remove redundant code from freePGconn() and add
comments in hopes of preventing similar mistakes in future.
Noticed while trying (futilely) to reproduce bug #3902.
