view image_filter_finalize.t @ 1752:ba6e24e38f03

Tests: improved stop_daemons() to send signal again. As was observed, it's possible that a signal to complete a uwsgi daemon can be ignored while it is starting up, which results in tests hang due to eternal waiting on child processes termination. Notably, it is seen when running tests with a high number of prove jobs on a low-profile VM against nginx with broken modules and/or configuration. To reproduce: $ TEST_NGINX_GLOBALS=ERROR prove -j16 uwsgi*.t Inspecting uwsgi under ktrace on FreeBSD confirms that a SIGTERM signal is ignored at the very beginning of uwsgi startup. It is then replaced with a default action after listen(), thus waiting until uwsgi is ready to accept new TCP connections doesn't completely solve the hang window. The fix is to retry sending a signal some time after waitpid(WNOHANG) continuously demonstrated no progress with reaping a signaled process. It is modelled after f13ead27f89c that improved stop() for nginx.
author Sergey Kandaurov <pluknet@nginx.com>
date Wed, 29 Dec 2021 22:29:23 +0300
parents 196d33c2bb45
children
line wrap: on
line source

#!/usr/bin/perl

# (C) Maxim Dounin

# Tests for http filter finalize code.

###############################################################################

use warnings;
use strict;

use Test::More;

BEGIN { use FindBin; chdir($FindBin::Bin); }

use lib 'lib';
use Test::Nginx;

###############################################################################

select STDERR; $| = 1;
select STDOUT; $| = 1;

my $t = Test::Nginx->new()->has(qw/http proxy cache image_filter limit_req/)
	->has(qw/rewrite/)->plan(3)
	->write_file_expand('nginx.conf', <<'EOF');

%%TEST_GLOBALS%%

daemon off;

events {
}

http {
    %%TEST_GLOBALS_HTTP%%

    proxy_cache_path %%TESTDIR%%/cache keys_zone=cache:1m;

    limit_req_zone $binary_remote_addr zone=limit:1m rate=25r/m;

    log_format time "$request_uri:$status:$upstream_response_time";
    access_log time.log time;

    upstream u {
        server 127.0.0.1:8081;
        server 127.0.0.1:8081;
        server 127.0.0.1:8081;
        server 127.0.0.1:8081;
        server 127.0.0.1:8080;
    }

    server {
        listen       127.0.0.1:8080;
        server_name  localhost;

        # this used to cause a segmentation fault before 07f028df3879 (1.3.1)
        # http://nginx.org/pipermail/nginx/2011-January/024703.html

        location /t1 {
            proxy_pass http://127.0.0.1:8080/bad;
            proxy_cache cache;
            proxy_cache_valid any 1h;

            image_filter   resize  150 100;
            error_page     415   = /empty;
        }

        location /empty {
            return 204;
        }

        location /bad {
            return 404;
        }

        # another segfault, introduced in 204b780a89de (1.3.0),
        # fixed in 07f028df3879 (1.3.1)

        location /t2 {
            proxy_pass http://127.0.0.1:8080/big;
            proxy_store on;

            image_filter_buffer 10m;
            image_filter   resize  150 100;
            error_page     415   = /empty;
        }

        location /big {
            # big enough static file
        }

        # filter finalization may cause duplicate upstream finalization,
        # resulting in wrong $upstream_response_time,
        # http://nginx.org/pipermail/nginx-devel/2015-February/006539.html

        # note that we'll need upstream response time to be at least 1 second,
        # and at least 4 failed requests to make sure r->upstream_states will
        # not be reallocated

        location /t3 {
            proxy_pass http://u/slow;
            proxy_buffering off;

            image_filter   resize  150 100;
            error_page     415   = /upstream;
        }

        location /slow {
            limit_req zone=limit burst=5;
        }

        location /upstream {
            proxy_pass http://127.0.0.1:8080/empty;
        }

        location /time.log {
            # access to log
        }
    }

    server {
        listen       127.0.0.1:8081;
        server_name  localhost;
        return 444;
    }
}

EOF

$t->write_file('big', "x" x 10240000);
$t->write_file('slow', "x");

$t->run();

###############################################################################

like(http_get('/t1'), qr/HTTP/, 'image filter and cache');
like(http_get('/t2'), qr/HTTP/, 'image filter and store');

http_get('/slow');
http_get('/t3');
like(http_get('/time.log'), qr!/t3:.*, [1-9]\.!, 'upstream response time');

# "aio_write" is used to produce the following alert on some platforms:
# "readv() failed (9: Bad file descriptor) while reading upstream"

$t->todo_alerts() if $t->read_file('nginx.conf') =~ /aio_write on/
	and $t->read_file('nginx.conf') =~ /aio threads/;

###############################################################################