4
<!-- Manpage converted by man2html 3.0.1 -->
6
qping - check application status of Grid Engine daemons.
9
qping [-help] [-noalias] [-ssl|-tcp] [ [ [-i <interval>] [-
10
info] [-f] ] | [ [-dump_tag tag [param] ] [-dump] [-
11
nonewline] ] ] <host> <port> <name> <id>
14
<I>Qping</I> is used to validate the runtime status of a Grid
15
Engine service daemon. The current Grid Engine implementa-
16
tion allows one to query the SGE_QMASTER daemon and any run-
17
ning SGE_EXECD daemon. The qping command is used to send a
18
SIM (Status Information Message) to the destination daemon.
19
The communication layer of the specified daemon will respond
20
with a SIRM (Status Information Response Message) which con-
21
tains status information about the consulted daemon.
23
The qping -dump and -dump_tag options allowing an adminis-
24
trator to observe the communication protocol data flow of a
25
Grid Engine service daemon. The qping -dump instruction must
26
be started with root account and on the same host where the
27
observed daemon is running.
31
Show full status information on each ping interval.
33
<I>First</I> <I>output</I> <I>Line</I>: The first output line shows the date and
36
<I>SIRM</I> <I>version</I>: Internal version number of the SIRM (Status
37
Information Response Message)
39
<I>SIRM</I> <I>message</I> <I>id</I>: Current message id for this connection
41
<I>start</I> <I>time</I>: Start time of daemon. The format is as follows:
43
MM/DD/YYYY HH:MM:SS (seconds since 01.01.1970)
45
<I>run</I> <I>time</I> [<I>s</I>]: Run time in seconds since start time
47
<I>messages</I> <I>in</I> <I>read</I> <I>buffer</I>: Nr. of buffered messages in com-
48
munication buffer. The messages are buffered for the appli-
49
cation (daemon). When this number grows too large the daemon
50
is not able to handle all messages sent to it.
52
<I>messages</I> <I>in</I> <I>write</I> <I>buffer</I>: Nr. of buffered messages in the
53
communication write buffer. The messages are sent from the
54
application (daemon) to the connected clients, but the com-
55
munication layer wasn't able to send the messages yet. If
56
this number grows too large, the communication layer isn't
57
able to send them as fast as the application (daemon) wants
58
the messages to be sent.
60
<I>nr</I>. <I>of</I> <I>connected</I> <I>clients</I>: This is the number of actual con-
61
nected clients to this daemon. This also implies the current
64
<I>status</I>: The status value of the daemon. This value depends
65
on the application which reply to the qping request. If the
66
application does not provide any information the status is
67
99999. Here are the possible status information values for
68
the Grid Engine daemons:
72
<I>0</I> There is no unusual timing situation.
74
<I>1</I> One or more threads has reached warning timeout.
75
This may happen when at least one thread does not
76
increment his time stamp for a not usual long time. A
77
possible reason for this is a high workload for this
80
<I>2</I> One or more threads has reached error timeout. This
81
may happen when at least one thread has not incre-
82
mented his time stamp for longer than 10 minutes.
84
<I>3</I> The time measurement is not initialized.
88
<I>0</I> There is no unusual timing situation.
90
<I>1</I> Dispatcher has reached warning timeout. This may
91
happen when the dispatcher does not increment his time
92
stamp for a unusual long time. A possible reason for
93
this is a high workload.
95
<I>2</I> Dispatcher has reached error timeout. This may hap-
96
pen when the dispatcher has not incremented his time
97
stamp for longer than 10 minutes.
99
<I>3</I> The time measurement is not initialized.
102
<I>info</I>: Status message of the daemon. This value depends on
103
the application which reply to the qping request. If the
104
application does not provide any information the info mes-
105
sage is "not available". Here are the possible status
106
information values for the Grid Engine daemons:
111
The info message contains information about the qmas-
112
ter threads followed by a thread state and time infor-
113
mation. Each time when one of the known threads pass
114
through their main loop the time information is
115
updated. Since the qmaster has two message threads
116
every message thread updates the time. This means the
117
timeout for the message thread (MT) can only occur
118
when no message thread is active anymore:
120
THREAD_NAME: THREAD_STATE (THREAD_TIME)
123
EDT: Event Delivery Thread
124
TET: Timed Event Thread
125
MT: Message Thread(s)
128
In addition to these thread names, the name can contain a thread number (for example:
129
<B><A HREF="../htmlman1/MT.html">MT(1)</A></B>), when multiple instances of this thread are running.
137
Time since last timestamp updating.
139
After the dispatcher information follows an additional
140
information string which describes the complete appli-
145
The info message contains information for the execd
147
dispatcher: STATE (TIME)
155
Time since last timestamp updating.
157
After the thread information follows an additional infor-
158
mation string which describes the application status.
160
<I>Monitor</I>: If available, displays statistics on a thread. The
161
data for each thread is displayed in one line. The format of
162
this line can be changed at any time. Only the master imple-
163
ments the monitoring.
167
Prints a list of all options.
171
Set qping interval time.
173
The default interval time is one second. Qping will send a
174
SIM (Status Information Message) on each interval time.
178
Show full status information (see -f for more information)
179
and exit. The exit value 0 indicates no error. On errors
180
qping returns with 1.
184
Ignore host_aliases file, which is located at
185
<<I>sge</I>_<I>root</I>>/<<I>cell</I>>/<I>common</I>/<I>host</I>_<I>aliases</I>. If this option is
186
used it is not necessary to set any Grid Engine environment
191
This option can be used to specify an SSL (Secure Socket
192
Layer) configuration. The qping will use the configuration
193
to connect to services running SSL. If the SGE settings file
194
is not sourced, you have to use the -noalias option to
195
bypass the need for the SGE_ROOT environment variable. The
196
following environment variables are used to specifiy your
198
SSL_CA_CERT_FILE - CA certificate file
199
SSL_CERT_FILE - certificates file
200
SSL_KEY_FILE - key file
201
SSL_RAND_FILE - rand file
205
This option is used to select TCP/IP as the protocol used to
206
connect to other services.
210
Dump output will not have a linebreak within a message and
211
binary messages are not unpacked.
214
This option allows an administrator to observe the communi-
215
cation protocol data flow of a Grid Engine service daemon.
216
The qping -dump instruction must be started as root and on
217
the same host where the observed daemon is running.
219
The output is written to stdout. The enviroment variable
220
"SGE_QPING_OUTPUT_FORMAT" can be set to hide columns, set a
221
default column width or to set a hostname output format. The
222
value of the environment variable can be set to any combina-
223
tion of the following specifiers separated by a space char-
225
"h:X" -> hide column X
226
"s:X" -> show column X
227
"w:X:Y" -> set width of column X to Y
228
"hn:X" -> set hostname output parameter X.
229
X values are "long" or "short"
231
Start qping -help to see which columns are available.
235
-dump_tag tag [param]
236
This option has the same the same meaning as -dump, but can
237
provide more information by specifying the debug level and
238
message types qping should print:
239
-dump_tag ALL <debug level>
240
This option shows all possible debug messages
241
(APP+MSG) for the debug levels, ERROR, WARNING, INFO,
242
DEBUG and DPRINTF. The contacted service must support
243
this kind of debugging. This option is not currently
245
-dump_tag APP <debug level>
246
This option shows only application debug messages for
247
the debug levels, ERROR, WARNING, INFO, DEBUG and
248
DPRINTF. The contacted service must support this kind
249
of debugging. This option is not currently imple-
252
This option has the same behaviour as the -dump
257
Host where daemon is running.
261
Port which daemon has bound (used sge_qmaster/sge_execd port
266
Name of communication endpoint ("qmaster" or "execd"). A
267
communication endpoint is a triplet of hostname/endpoint
268
name/endpoint id (e.g. hostA/qmaster/1 or subhost/qstat/4).
272
Id of communication endpoint ("1" for daemons)
277
>qping master_host 31116 qmaster
278
08/24/2004 16:41:15 endpoint master_host/qmaster/1 at port 31116 is up since 365761 seconds
279
08/24/2004 16:41:16 endpoint master_host/qmaster/1 at port 31116 is up since 365762 seconds
280
08/24/2004 16:41:17 endpoint master_host/qmaster/1 at port 31116 is up since 365763 seconds
282
> qping -info master_host 31116 qmaster 1
286
start time: 08/20/2004 11:05:14 (1092992714)
288
messages in read buffer: 0
289
messages in write buffer: 0
290
nr. of connected clients: 4
294
> qping -info execd_host 31117 execd 1
298
start time: 08/20/2004 11:06:13 (1092992773)
300
messages in read buffer: 0
301
messages in write buffer: 0
302
nr. of connected clients: 2
308
ENVIRONMENTAL VARIABLES
309
SGE_ROOT Specifies the location of the Grid Engine
310
standard configuration files.
312
SGE_CELL If set, specifies the default Grid Engine
316
<B><A HREF="../htmlman1/sge_intro.html">sge_intro(1)</A></B>, <B><A HREF="../htmlman5/SGE_H_ALIASES.html">SGE_H_ALIASES(5)</A></B>, <B><A HREF="../htmlman8/sge_qmaster.html">sge_qmaster(8)</A></B>,
317
<B><A HREF="../htmlman8/sge_execd.html">sge_execd(8)</A></B>.
320
See <B><A HREF="../htmlman1/sge_intro.html">sge_intro(1)</A></B> for a full statement of rights and permis-
372
Man(1) output converted with
373
<a href="http://www.oac.uci.edu/indiv/ehood/man2html.html">man2html</a>