php.net |  support |  documentation |  report a bug |  advanced search |  search howto |  statistics |  random bug |  login
Bug #36469 pg_get_result does not block
Submitted: 2006-02-21 08:18 UTC Modified: 2008-11-10 01:00 UTC
Votes:5
Avg. Score:3.6 ± 0.8
Reproduced:3 of 3 (100.0%)
Same Version:1 (33.3%)
Same OS:1 (33.3%)
From: levi at alliancesoftware dot com dot au Assigned: helly (profile)
Status: No Feedback Package: PostgreSQL related
PHP Version: 5.1.2 OS: FC3
Private report: No CVE-ID: None
Welcome back! If you're the original bug submitter, here's where you can edit the bug or add additional notes.
If this is not your bug, you can add a comment by following this link.
If this is your bug, but you forgot your password, you can retrieve your password here.
Password:
Status:
Package:
Bug Type:
Summary:
From: levi at alliancesoftware dot com dot au
New email:
PHP Version: OS:

 

 [2006-02-21 08:18 UTC] levi at alliancesoftware dot com dot au
Description:
------------
According to the pg_connection_busy documentation, if pg_get_result is called on a pg_send_query that is currently executing, it should block until results become available.

Occasionally this does not happen if the query had an error in it.

Reproduce code:
---------------
#!/usr/local/bin/php51
<?php
  $connStr = sprintf("host=%s dbname=%s user=%s password=%s", 'localhost', 'testdb', 'liam', 'liam');
  $dbconn = pg_connect($connStr) or die("Could not connect");

        $i = 0;

        while (!pg_connection_busy($dbconn)) {
                pg_send_query($dbconn, 'i am a syntax error');
                pg_get_result($dbconn);
                $i++;
        }

        echo "got a busy signal after $i commands";

?>


Expected result:
----------------
(Loops forever)

Actual result:
--------------
got a busy signal after 221 commands

(actual count varies, but always returns within 1 second. Approximately 1/3 of the time it gives a "happy" run and will loop forever)


Note that if the SQL statement was a valid statement, I cannot get the erroneous behaviour to occur.

Patches

Add a Patch

Pull Requests

Add a Pull Request

History

AllCommentsChangesGit/SVN commitsRelated reports
 [2006-02-22 10:57 UTC] tony2001@php.net
What PG version are you using and did you try to upgrade it (and rebuild PHP with the upgraded libs) ?
 [2006-02-23 02:35 UTC] levi at alliancesoftware dot com dot au
I have tried and can repeat the error on the following:

Fedora Core 2 (x86):
 PHP 5.0.4* PG 8.1*
 PHP 5.1.2* PG 8.1*

Fedora Core 3 (x86-64):
 PHP 5.0.4** PG 8.1*
 PHP 5.1.2*  PG 8.1*

* = compiled from source
** = system RPM

 I also doublechecked that the connection only says its busy if the previous query had an error (ie change the query to a valid SELECT and it doesn't happen)
 [2008-11-02 12:31 UTC] jani@php.net
Please try using this CVS snapshot:

  http://snaps.php.net/php5.2-latest.tar.gz
 
For Windows:

  http://windows.php.net/snapshots/


 [2008-11-10 01:00 UTC] php-bugs at lists dot php dot net
No feedback was provided for this bug for over a week, so it is
being suspended automatically. If you are able to provide the
information that was originally requested, please do so and change
the status of the bug back to "Open".
 [2009-11-09 00:43 UTC] levi at alliancesoftware dot com dot au
Edit: We gave up on Postgres for that project, but had to revisit this recently; the issue still exists but it seems the cause is actually that pg_connection_busy() stays true for some time after a query that causes any sort of postgres error.

Tested additionally with
x86_64/FC10 + PHP [5.2.9 | 5.2.11 | 5.3.0] + Postgres 8.3


#!/usr/bin/php
<?php
    $connStr = sprintf("host=%s dbname=%s user=%s password=%s", 'localhost', 'testlevi', 'levi', 'levi');
    ($dbConn = pg_connect($connStr)) or die("Could not connect");

    @pg_query($dbConn, "drop table thisTableExists");
    @pg_query($dbConn, "drop table thisTableDoesNotExist");
    pg_query($dbConn, "create table thisTableExists (x varchar(255))") || die("Can't create table");
    pg_query($dbConn, "insert into thisTableExists values ('foo')") || die("Can't insert first row");
    pg_query($dbConn, "insert into thisTableExists values ('foo2')") || die("Can't insert second row");

    $loopCount = 0;
    while (true) {

            // First 20 queries are completely valid; after that we try invalid queries

            $sql = ($loopCount < 20)
                ? "select * from thisTableExists; select * from thisTableExists;"
                : "select * from thisTableExists; select * from thisTableDoesNotExist;";

            // Issue the queries
            pg_send_query($dbConn, $sql);

            // Check first result
            $result1 = pg_get_result($dbConn);
            $rowCount1 = pg_num_rows($result1);
            if ($rowCount1 != 2) die("Bad result set returned (1)");

            // Check the second result
            $result2 = pg_get_result($dbConn);
            $rowCount2 = pg_num_rows($result2);
            $expectedRowCount = ($loopCount >= 20) ? 0 : 2;
            if ($rowCount2 != $expectedRowCount) die("Bad result set returned (2)");

            if (pg_connection_busy($dbConn)) {
                echo (pg_connection_status($dbConn) == PGSQL_CONNECTION_OK) ? '+' : '-';
                while (pg_connection_busy($dbConn)) {
                    echo '!';
                    usleep(1);
                }
            }
            echo ".";
            $loopCount++;
    }
?>

Expected output:
........................... etc
Actual output:
....................+!.+!.+...............+.+!!!!!!!!!!!!!!!!!!!!!!!!!!!!!.....+!.+!.+!.+



Comments for anyone else experiencing this:

It seems pg_get_result() correctly waits until results are ready, but even after pg_get_result() returns, if there was an error in the query then pg_connection_busy() will stay true for some time.
 
PHP Copyright © 2001-2024 The PHP Group
All rights reserved.
Last updated: Tue Apr 16 13:01:30 2024 UTC