Line 0
Link Here
|
0 |
- |
1 |
#!/usr/bin/perl |
|
|
2 |
|
3 |
# This file is part of Koha. |
4 |
# |
5 |
# Koha is free software; you can redistribute it and/or modify it |
6 |
# under the terms of the GNU General Public License as published by |
7 |
# the Free Software Foundation; either version 3 of the License, or |
8 |
# (at your option) any later version. |
9 |
# |
10 |
# Koha is distributed in the hope that it will be useful, but |
11 |
# WITHOUT ANY WARRANTY; without even the implied warranty of |
12 |
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
13 |
# GNU General Public License for more details. |
14 |
# |
15 |
# You should have received a copy of the GNU General Public License |
16 |
# along with Koha; if not, see <http://www.gnu.org/licenses>. |
17 |
|
18 |
=head1 NAME |
19 |
|
20 |
background_jobs_worker_es.pl - Worker script that will process background Elasticsearch jobs |
21 |
|
22 |
=head1 SYNOPSIS |
23 |
|
24 |
./background_jobs_worker_es.pl |
25 |
|
26 |
=head1 DESCRIPTION |
27 |
|
28 |
This script will connect to the Stomp server (RabbitMQ) and subscribe to the Elasticsearch queue, processing batches every second. |
29 |
If a Stomp server is not active it will poll the database every 10s for new jobs in the Elasticsearch queue |
30 |
and process them in batches every second. |
31 |
|
32 |
=cut |
33 |
|
34 |
use Modern::Perl; |
35 |
use JSON qw( decode_json ); |
36 |
use Try::Tiny qw( catch try ); |
37 |
use Pod::Usage; |
38 |
use Getopt::Long; |
39 |
|
40 |
use C4::Context; |
41 |
use Koha::BackgroundJobs; |
42 |
use Koha::SearchEngine; |
43 |
use Koha::SearchEngine::Indexer; |
44 |
|
45 |
my ( $help ); |
46 |
GetOptions( |
47 |
'h|help' => \$help, |
48 |
) || pod2usage(1); |
49 |
|
50 |
pod2usage(0) if $help; |
51 |
|
52 |
die "Not using Elasticsearch" unless C4::Context->preference('SearchEngine') eq 'Elasticsearch'; |
53 |
|
54 |
my $conn; |
55 |
try { |
56 |
$conn = Koha::BackgroundJob->connect; |
57 |
} catch { |
58 |
warn sprintf "Cannot connect to the message broker, the jobs will be processed anyway (%s)", $_; |
59 |
}; |
60 |
|
61 |
if ( $conn ) { |
62 |
# FIXME cf note in Koha::BackgroundJob about $namespace |
63 |
my $namespace = C4::Context->config('memcached_namespace'); |
64 |
$conn->subscribe({ destination => sprintf("/queue/%s-%s", $namespace, 'elastic_index'), ack => 'client' }); |
65 |
} |
66 |
my $indexer = Koha::SearchEngine::Indexer->new({ index => $Koha::SearchEngine::BIBLIOS_INDEX }); |
67 |
my @jobs = (); |
68 |
my @records = (); |
69 |
my $frame; |
70 |
|
71 |
while (1) { |
72 |
local $SIG{ALRM} = sub { commit_records() if scalar @jobs }; # NB: \n required |
73 |
alarm 1; |
74 |
if ( $conn ) { |
75 |
$frame = $conn->receive_frame; |
76 |
if ( !defined $frame ) { |
77 |
# maybe log connection problems |
78 |
next; # will reconnect automatically |
79 |
} |
80 |
|
81 |
my $body = $frame->body; |
82 |
my $args = decode_json($body); # TODO Should this be from_json? Check utf8 flag.<F9> |
83 |
|
84 |
# FIXME This means we need to have create the DB entry before |
85 |
# It could work in a first step, but then we will want to handle job that will be created from the message received |
86 |
my $job = Koha::BackgroundJobs->find($args->{job_id}); |
87 |
next unless defined $job; |
88 |
|
89 |
push @records, @{ $args->{record_ids} }; |
90 |
push @jobs, $job; |
91 |
|
92 |
|
93 |
} else { |
94 |
my $jobs = Koha::BackgroundJobs->search({ status => 'new', queue => 'elastic_index' }); |
95 |
while ( my $job = $jobs->next ) { |
96 |
my $args = $job->json->decode($job->data); |
97 |
push @records, @{ $args->{record_ids} }; |
98 |
push @jobs, $job; |
99 |
} |
100 |
sleep 10; |
101 |
} |
102 |
} |
103 |
$conn->disconnect; |
104 |
|
105 |
sub commit_records { |
106 |
eval { |
107 |
$indexer->update_index(\@records); |
108 |
}; |
109 |
if ( $@ ){ |
110 |
warn $@; |
111 |
} |
112 |
foreach my $job (@jobs){ |
113 |
$job->status('finished')->store; |
114 |
} |
115 |
@jobs = (); |
116 |
@records = (); |
117 |
$conn->ack( { frame => $frame } ) if $frame; # FIXME depending on success? |
118 |
# Acknowledging the current frame also acknowledges all previously batched frames |
119 |
} |
120 |
|
121 |
|
122 |
sub process_job { |
123 |
my ( $job, $args ) = @_; |
124 |
|
125 |
my $pid; |
126 |
if ( $pid = fork ) { |
127 |
wait; |
128 |
return; |
129 |
} |
130 |
|
131 |
die "fork failed!" unless defined $pid; |
132 |
|
133 |
$job->process( $args ); |
134 |
|
135 |
exit; |
136 |
} |