1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
|
#!/usr/bin/env perl
###############################################################################
# #
# Script to check the status of DQM Harvesting jobs #
# #
# It basically checks whether the root tree is present for all jobs #
# If not, it resubmits the failed jobs on request #
# #
# Usage: check_harvesting.pl [crab_status_output] #
# #
# NOTE: it needs to be run from the directory from which you submitted #
# multicrab jobs #
# #
# Optional argument: #
# crab_status_output: file containing output of "multicrab -status" command #
# if not given, the script will get it itself #
# #
# Author: Vuko Brigljevic, Rudjer Boskovic Institute #
# First Version: January 22, 2010 #
# #
###############################################################################
use strict;
# Gets the crab status from input or produce it
my $nargs = @ARGV;
my $crabstatus = "";
if ($nargs>0 && -e $ARGV[0] ) {
$crabstatus= $ARGV[0];
}
if ($crabstatus eq "") {
$crabstatus = "multicrabstatus.out";
print "Getting multicrab jobs status ... \n";
system "multicrab -status > $crabstatus ";
print "Done \n";
}
# Open the multicrab cfg file to extract the list of job ids
#
my $crabcfg = "multicrab.cfg";
if (! -r $crabcfg ) {
print "No multicrab.cfg in this directory, quitting... \n";
exit;
}
open (CRABCFG,$crabcfg);
#my @lines=<STDIN>;
my @lines=<CRABCFG>;
my $line="";
my $tempfile=".check_harvesting_output";
my $castorbase="/castor/cern.ch";
my $job;
my $nsuccess=0;
my $njobs=0;
my $nfailed=0;
my @failedjobs=();
foreach $line (@lines) {
if ( $line =~ /\[(.*?)\]/ )
{
if ( $1 ne "MULTICRAB") {
$job = $1;
$njobs++;
}
}
if ($line =~ /USER.user_remote_dir/) {
my @words = split (/ = /,$line);
chop($words[1]);
my $castordir=$castorbase.$words[1]; # ADD JOB NAME IN CASTOR DIRECTORY!
# MODIFICATION: to account for the fact that multicrab adds job name in directory tree for output!!!
$castordir=$castordir."/".$job;
# print "CASTOR DIR: $castordir \n";
# this lines needs to be commented out to account for the above modification
# chop($castordir); #remove newline at the end
if ( -e $tempfile ) {
system "rm $tempfile";
}
my $rfcmd="rfdir $castordir > $tempfile";
# Check if root file present in output directory
system $rfcmd ;
open(LIST,$tempfile);
my $tmpline;
my $nrootfiles=0;
while ($tmpline = <LIST> ) {
if ( $tmpline =~ /root/ ) {
$nrootfiles++;
}
}
if ($nrootfiles > 0) {
$nsuccess++;
} else {
$failedjobs[$nfailed]=$job;
$nfailed++;
}
}
}
print "Number of jobs : $njobs \n";
print "Number of successful jobs (root file present : $nsuccess \n";
print "Number of failed or still running jobs : $nfailed \n";
my %jobstatus;
##################################################
# Get job status for jobs without root output
foreach $job (@failedjobs) {
my $jobstatus = get_multicrab_job_status($job,$crabstatus);
$jobstatus{$job} = $jobstatus;
print " $job : $jobstatus \n";
}
if ($nfailed == 0) {
exit;
}
# Resubmit failed jobs if desired
print "Would you like to resubmit the failed jobs? [y/n] \n";
my $reply=<STDIN>;
chop ($reply);
print "reply: $reply \n";
if ( $reply eq "y" || $reply eq "Y" ) {
print "Resubmitting jobs... \n";
foreach $job (@failedjobs) {
my $status = $jobstatus{$job};
my $retrieve = 0;
my $resubmit = 0;
if ( $status eq "Scheduled"
|| $status eq "Pending"
|| $status eq "Running" ) {
print "$job running: leave it in peace... \n";
} elsif ($status eq "Aborted") {
$resubmit = 1;
} elsif ($status eq "Done") {
$retrieve = 1;
$resubmit = 1;
} elsif ($status eq "Retrieved") {
$resubmit = 1;
}
if ($retrieve) {
print "retrieving... \n";
system "crab -c $job -getoutput all";
}
if ($resubmit) {
print "resubmitting... \n";
system "crab -c $job -resubmit all";
}
}
}
##################################################################################
sub get_multicrab_job_status {
#-------------------------------------------------------------------#
# Aim: Get the CRAB job status of a multicrab job #
# by parsing the "multicrab -status" output #
# #
# two input arguments #
# $job : CRAB job name #
# $crabstatus : file with output of "multicrab -status" #
#-------------------------------------------------------------------#
if ( scalar(@_) != 2 ) {
print "get_multicrab_job_status() called with wrong number of arguments: @_ \n";
return "";
}
my ($job, $crabstatus) = @_ ;
open(CRABSTATUS,$crabstatus);
my $jobfound=0;
my $linecounter=0;
my $jobstatus="undefined";
while ( <CRABSTATUS>) {
if ($linecounter > 0) {
if ($linecounter == 2) {
# This is the line with the job status
my @words = split (" ");
$jobstatus=$words[1];
last;
}
$linecounter++;
}
if ($jobfound) {
if (/ID/ && /STATUS/ ) {
$linecounter++
}
next;
}
if (/$job/) {
$jobfound = 1;
}
}
close(CRABSTATUS);
return $jobstatus;
}
|