backport to buster
[hcoop/debian/openafs.git] / doc / man-pages / pod1 / scout.pod
CommitLineData
805e021f
CE
1=head1 NAME
2
3scout - Monitors the File Server process
4
5=head1 SYNOPSIS
6
7=for html
8<div class="synopsis">
9
10B<scout> [B<initcmd>] S<<< B<-server> <I<servers to monitor>>+ >>>
11 S<<< [B<-basename> <I<base server name>>] >>>
12 S<<< [B<-frequency> <I<poll frequency, in seconds>>] >>> [B<-host>]
13 S<<< [B<-attention> <I<specify attention (highlighting) level>>+] >>>
14 S<<< [B<-columnwidths> <I<number of characters>>+] >>>
15 S<<< [B<-debug> <I<turn debugging output on to the named file>>] >>>
16 [B<-version>] [B<-help>]
17
18B<scout> [B<i>] S<<< B<-s> <I<servers to monitor>>+ >>>
19 S<<< [B<-b> <I<base server name>>] >>> S<<< [B<-f> <I<poll frequency, in seconds>>] >>>
20 [B<-ho>] S<<< [B<-a> <I<specify attention (highlighting) level>>+] >>>
21 S<<< [B<-c> <I<number of characters>+>] >>>
22 S<<< [B<-d> <I<turn debugging output on to the named file>>] >>> [B<-version>] [B<-he>]
23
24=for html
25</div>
26
27=head1 DESCRIPTION
28
29The scout command displays statistics gathered from the File Server
30process running on each machine specified with the B<-server>
31argument. L</OUTPUT> explains the meaning of the statistics and describes
32how they appear in the command shell, which is preferably a window managed
33by a window manager program.
34
35=head1 CAUTIONS
36
37The B<scout> program must be able to access the curses graphics package,
38which it uses to display statistics. Most UNIX distributions include
39curses as a standard utility.
40
41Both dumb terminals and windowing systems that emulate terminals can
42display the B<scout> program's statistics. The display makes use of
43reverse video and cursor addressing, so the display environment must
44support those features for it to look its best (most windowing systems do,
45most dumb terminals do not). Also, set the TERM environment variable to
46the correct terminal type, or one with characteristics similar to the
47actual ones. For machines running the AIX operating system, the
48recommended setting for TERM is C<vt100>, as long as the terminal is
49similar to that. For other operating systems, the wider range of
50acceptable values includes C<xterm>, C<xterms>, C<vt100>, C<vt200>, and
51C<wyse85>.
52
53=head1 OPTIONS
54
55=over 4
56
57=item B<initcmd>
58
59Accommodates the command's use of the AFS command parser, and is optional.
60
61=item B<-server> <I<servers to monitor>>+
62
63Specifies each file server machine running a File Server process to
64monitor. Provide each machine's fully qualified hostname unless the
65B<-basename> argument is used. In that case, specify only the unique
66initial part of each machine name, omitting the domain name suffix (the
67basename) common to all the names. It is also acceptable to use the
68shortest abbreviated form of a host name that distinguishes it from other
69machines, but successful resolution depends on the availability of a name
70resolution service (such as the Domain Name Service or a local host table)
71at the time the command is issued.
72
73=item B<-basename> <I<base server name>>
74
75Specifies the basename (domain name) suffix common to all of the file
76server machine names specified with the B<-server> argument, and is
77automatically appended to them. This argument is normally the name of the
78cell to which the machines belong. Do not include the period that
79separates this suffix from the distinguishing part of each file server
80machine name, but do include any periods that occur within the suffix
81itself. For example, in the Example Corporation cell, the proper value is
82C<example.com> rather than C<.example.com>.
83
84=item B<-frequency> <I<poll frequency>>
85
86Indicates how often to probe the File Server processes. Specify a number
87of seconds greater than C<0> (zero). The default is 60 seconds.
88
89=item B<-host>
90
91Displays the name of the machine that is running the scout program, in the
92banner line of the display screen.
93
94=item B<-attention> <I<attention level>>+
95
96Defines a list of entries, each of which pairs a statistic and a threshold
97value. When the value of the statistic exceeds the indicated threshold
98value, it is highlighted (in reverse video) in the display. List the pairs
99in any order. The acceptable values are the following:
100
101=over 4
102
103=item conn <I<connections>>
104
105Indicates the number of open connections to client processes at which to
106highlight the statistic. The statistic returns to regular display when
107the value goes back below the threshold. There is no default threshold.
108
109An example of an acceptable value is conn 300.
110
111=item disk <I<blocks_free>>
112
113Indicates the number of remaining free kilobyte blocks at which to
114highlight the statistic. The statistic returns to regular display when the
115value again exceeds the threshold. There is no default threshold.
116
117An example of an acceptable value is disk 5000.
118
119=item disk <I<percent_full>>%
120
121Indicates the percentage of disk usage at which to highlight the
122statistic. The statistic returns to regular display when the value goes
123back below the threshold. The default threshold is 95%. Acceptable values
124are the integers in the range from C<0> to C<99>, followed by the percent
125sign (C<%>) to distinguish this type of value from the one described just
126previously.
127
128An example is disk 90%.
129
130=item fetch <I<fetch RPCs>>
131
132Indicates the cumulative number of fetch RPCs from client processes at
133which to highlight the statistic. The statistic does not return to regular
134display until the File Server process restarts, at which time the value
135returns to zero. There is no default threshold.
136
137Example of a legal value: fetch 6000000
138
139=item store <I<store RPCs>>
140
141Indicates the cumulative number of store RPCs from client processes at
142which to highlight the statistic. The statistic does not return to regular
143display until the File Server process restarts, at which time the value
144returns to zero. There is no default threshold.
145
146Example of an acceptable value: store 200000
147
148=item ws <I<active client machines>>
149
150Indicates the number of client machines with active open connections at
151which to highlight the statistic. An active connection is defined as one
152over which the File Server and client have communicated in the last 15
153minutes. The statistic returns to regular display when the value goes back
154below the threshold. There is no default threshold.
155
156Example of an acceptable value: ws 65
157
158=back
159
160=item B<-columnwidths> <I<number of characters>>+
161
162Specifies the number of characters to display in each column of the B<scout>
163statistics display region. Specify one to six numbers separated by spaces to
164set the number of characters to be displayed in each column. The values
165specify the widths of the columns in the same order the columns are displayed
166from left to right. Use 0 as a placeholder to specify a default column width.
167
168=item B<-debug> <I<debugging trace file>>
169
170Specifies the pathname of the file into which to write a debugging
171trace. Partial pathnames are interpreted relative to the current working
172directory.
173
174=item B<-help>
175
176Prints the online help for this command. All other valid options are
177ignored.
178
179=item B<-version>
180
181Prints the program version and then exits. All other valid options
182are ignored.
183
184=back
185
186=head1 OUTPUT
187
188The B<scout> program can display statistics either in a dedicated window
189or on a plain screen if a windowing environment is not available. For best
190results, the window or screen needs the ability to print in reverse video.
191
192The B<scout> screen has three main parts: the banner line, the statistics
193display region and the message/probe line.
194
195=head2 The Banner Line
196
197By default, the string C<Scout> appears in the banner line at the top of
198the window or screen. Two optional arguments place additional information
199in the banner line:
200
201=over 4
202
203=item *
204
205The B<-host> flag displays the name of the machine where the B<scout>
206program is running. As mentioned previously, this is useful when running
207the B<scout> program on several machines but displaying the results on a
208single machine.
209
210For example, when the B<-host> flag is included and the B<scout> program
211is running on the machine C<client1.example.com>, the banner line reads as
212follows:
213
214 [client1.example.com] Scout
215
216=item *
217
218The B<-basename> argument displays the indicated basename on the banner
219line. For example, including the argument C<-basename example.com> argument
220results in the following banner line:
221
222 Scout for example.com
223
224=back
225
226=head2 The Statistics Display Region
227
228In this region, which occupies the majority of the window, the B<scout>
229process displays the statistics gathered for each File Server
230process. Each process appears on its own line.
231
232The region is divided into six columns, labeled as indicated and
233displaying the following information:
234
235=over 4
236
237=item Conn
238
239The first column displays the number of RPC connections open between the
240File Server process and client machines. This number equals or exceeds
241the number in the C<Ws> column (see the fourth entry below), because each
242user on the machine can have several separate connections open at once,
243and one client machine can handle several users.
244
245=item Fetch
246
247The second column displays the number of fetch-type RPCs (fetch data,
248fetch access list, and fetch status) that client machines have made to the
249File Server process since the latter started. This number is reset to
250zero each time the File Server process restarts.
251
252=item Store
253
254The third column displays the number of store-type RPCs (store data, store
255access list, and store status) that client machines have made to the File
256Server process since the latter started. This number is reset to zero each
257time the File Server process restarts.
258
259=item Ws
260
261The fourth column displays the number of client machines (C<Ws> stands for
262workstations) that have communicated with the File Server process within
263the last 15 minutes. Such machines are termed I<active>). This number is
264likely to be smaller than the number in the first (C<Conn>) column because
265a single client machine can have several connections open to one File
266Server.
267
268=item server name
269
270The fifth, unlabeled, column displays the name of the file server machine
271on which the File Server process is running. Names of 12 characters or
272less are displayed in full; longer names are truncated and an asterisk
273(C<*>) appears as the last character in the name. Using the B<-basename>
274argument is a good way to avoid truncation, but only if all machine names
275end in a common string.
276
277=item Disk attn
278
279The sixth column displays the number of available kilobyte blocks on each
280AFS disk partition on the file server machine.
281
282The display for each partition has the following form:
283
284 x:<free_blocks>
285
286where C<x> indicates the partition name. For example, C<a:8949> specifies
287that the F</vicepa> partition has 8,949 1-KB blocks free. Available space
288can be displayed for up to 26 partitions. If the window is not wide enough
289for all partition entries to appear on a single line, the B<scout> process
290automatically creates multiple lines, stacking the partition entries into
291sub-columns within the sixth column.
292
293The label on the C<Disk attn> column indicates the threshold value at
294which entries in the column become highlighted. By default, the label is
295
296 Disk attn: > 95% used
297
298because by default the scout program highlights the entry for any
299partition that is over 95% full.
300
301=back
302
303For all columns except the fifth (file server machine name), the optional
304B<-attention> argument sets the value at which entries in the column are
305highlighted to indicate that a certain value has been exceeded. Only
306values in the fifth and C<Disk attn> columns ever become highlighted by
307default.
308
309If the scout program is unable to access or otherwise obtain information
310about a partition, it generates a message similar to the following
311example:
312
313 Could not get information on server fs1.example.com partition /vicepa
314
315=head2 The Message/Probe Line
316
317The bottom line of the scout screen indicates how many times the B<scout>
318program has probed the File Server processes for statistics. The
319statistics gathered in the latest probe appear in the statistics display
320region. The B<-frequency> argument overrides the default probe frequency
321of 60 seconds.
322
323=head1 EXAMPLES
324
325See the chapter on monitoring tools in the I<OpenAFS Administration
326Guide>, which illustrates the displays that result from different
327combinations of options.
328
329=head1 PRIVILEGE REQUIRED
330
331None
332
333=head1 SEE ALSO
334
335L<afsmonitor(1)>,
336L<fstrace(8)>
337
338=head1 COPYRIGHT
339
340IBM Corporation 2000. <http://www.ibm.com/> All Rights Reserved.
341
342This documentation is covered by the IBM Public License Version 1.0. It was
343converted from HTML to POD by software written by Chas Williams and Russ
344Allbery, based on work by Alf Wachsmann and Elizabeth Cassell.