1CTDB-STATISTICS(7)       CTDB - clustered TDB database      CTDB-STATISTICS(7)
2
3
4

NAME

6       ctdb-statistics - CTDB statistics output
7

OVERALL STATISTICS

9       CTDB maintains information about various messages communicated and some
10       of the important operations per node. See the ctdb(1) commands
11       statistics and statisticsreset for displaying statistics.
12
13   Example: ctdb statistics
14           CTDB version 1
15           Current time of statistics  :                Fri Sep 12 13:32:32 2014
16           Statistics collected since  : (000 01:49:20) Fri Sep 12 11:43:12 2014
17            num_clients                        6
18            frozen                             0
19            recovering                         0
20            num_recoveries                     2
21            client_packets_sent           281293
22            client_packets_recv           296317
23            node_packets_sent             452387
24            node_packets_recv             182394
25            keepalive_packets_sent          3927
26            keepalive_packets_recv          3928
27            node
28                req_call                   48605
29                reply_call                     1
30                req_dmaster                23404
31                reply_dmaster              24917
32                reply_error                    0
33                req_message                  958
34                req_control               197513
35                reply_control             153705
36            client
37                req_call                  130866
38                req_message                  770
39                req_control               168921
40            timeouts
41                call                           0
42                control                        0
43                traverse                       0
44            locks
45                num_calls                    220
46                num_current                    0
47                num_pending                    0
48                num_failed                     0
49            total_calls                   130866
50            pending_calls                      0
51            childwrite_calls                   1
52            pending_childwrite_calls             0
53            memory_used                   334490
54            max_hop_count                     18
55            total_ro_delegations               2
56            total_ro_revokes                   2
57            hop_count_buckets: 42816 5464 26 1 0 0 0 0 0 0 0 0 0 0 0 0
58            lock_buckets: 9 165 14 15 7 2 2 0 0 0 0 0 0 0 0 0
59            locks_latency      MIN/AVG/MAX     0.000685/0.160302/6.369342 sec out of 214
60            reclock_ctdbd      MIN/AVG/MAX     0.004940/0.004969/0.004998 sec out of 2
61            reclock_recd       MIN/AVG/MAX     0.000000/0.000000/0.000000 sec out of 0
62            call_latency       MIN/AVG/MAX     0.000006/0.000719/4.562991 sec out of 126626
63            childwrite_latency MIN/AVG/MAX     0.014527/0.014527/0.014527 sec out of 1
64
65
66   CTDB version
67       Version of the ctdb protocol used by the node.
68
69   Current time of statistics
70       Time when the statistics are generated.
71
72       This is useful when collecting statistics output periodically for
73       post-processing.
74
75   Statistics collected since
76       Time when ctdb was started or the last time statistics was reset. The
77       output shows the duration and the timestamp.
78
79   num_clients
80       Number of processes currently connected to CTDB's unix socket. This
81       includes recovery daemon, ctdb tool and samba processes (smbd,
82       winbindd).
83
84   frozen
85       1 if the databases are currently frozen, 0 otherwise.
86
87   recovering
88       1 if recovery is active, 0 otherwise.
89
90   num_recoveries
91       Number of recoveries since the start of ctdb or since the last
92       statistics reset.
93
94   client_packets_sent
95       Number of packets sent to client processes via unix domain socket.
96
97   client_packets_recv
98       Number of packets received from client processes via unix domain
99       socket.
100
101   node_packets_sent
102       Number of packets sent to the other nodes in the cluster via TCP.
103
104   node_packets_recv
105       Number of packets received from the other nodes in the cluster via TCP.
106
107   keepalive_packets_sent
108       Number of keepalive messages sent to other nodes.
109
110       CTDB periodically sends keepalive messages to other nodes. See
111       KeepaliveInterval tunable in ctdb-tunables(7) for more details.
112
113   keepalive_packets_recv
114       Number of keepalive messages received from other nodes.
115
116   node
117       This section lists various types of messages processed which originated
118       from other nodes via TCP.
119
120       req_call
121           Number of REQ_CALL messages from the other nodes.
122
123       reply_call
124           Number of REPLY_CALL messages from the other nodes.
125
126       req_dmaster
127           Number of REQ_DMASTER messages from the other nodes.
128
129       reply_dmaster
130           Number of REPLY_DMASTER messages from the other nodes.
131
132       reply_error
133           Number of REPLY_ERROR messages from the other nodes.
134
135       req_message
136           Number of REQ_MESSAGE messages from the other nodes.
137
138       req_control
139           Number of REQ_CONTROL messages from the other nodes.
140
141       reply_control
142           Number of REPLY_CONTROL messages from the other nodes.
143
144       req_tunnel
145           Number of REQ_TUNNEL messages from the other nodes.
146
147   client
148       This section lists various types of messages processed which originated
149       from clients via unix domain socket.
150
151       req_call
152           Number of REQ_CALL messages from the clients.
153
154       req_message
155           Number of REQ_MESSAGE messages from the clients.
156
157       req_control
158           Number of REQ_CONTROL messages from the clients.
159
160       req_tunnel
161           Number of REQ_TUNNEL messages from the clients.
162
163   timeouts
164       This section lists timeouts occurred when sending various messages.
165
166       call
167           Number of timeouts for REQ_CALL messages.
168
169       control
170           Number of timeouts for REQ_CONTROL messages.
171
172       traverse
173           Number of timeouts for database traverse operations.
174
175   locks
176       This section lists locking statistics.
177
178       num_calls
179           Number of completed lock calls. This includes database locks and
180           record locks.
181
182       num_current
183           Number of scheduled lock calls. This includes database locks and
184           record locks.
185
186       num_pending
187           Number of queued lock calls. This includes database locks and
188           record locks.
189
190       num_failed
191           Number of failed lock calls. This includes database locks and
192           record locks.
193
194   total_calls
195       Number of req_call messages processed from clients. This number should
196       be same as client --> req_call.
197
198   pending_calls
199       Number of req_call messages which are currently being processed. This
200       number indicates the number of record migrations in flight.
201
202   childwrite_calls
203       Number of record update calls. Record update calls are used to update a
204       record under a transaction.
205
206   pending_childwrite_calls
207       Number of record update calls currently active.
208
209   memory_used
210       The amount of memory in bytes currently used by CTDB using talloc. This
211       includes all the memory used for CTDB's internal data structures. This
212       does not include the memory mapped TDB databases.
213
214   max_hop_count
215       The maximum number of hops required for a record migration request to
216       obtain the record. High numbers indicate record contention.
217
218   total_ro_delegations
219       Number of readonly delegations created.
220
221   total_ro_revokes
222       Number of readonly delegations that were revoked. The difference
223       between total_ro_revokes and total_ro_delegations gives the number of
224       currently active readonly delegations.
225
226   hop_count_buckets
227       Distribution of migration requests based on hop counts values. Buckets
228       are 0, < 2, < 4, < 8, < 16, < 32, < 64, < 128, < 256, < 512, < 1024,
229       < 2048, < 4096, < 8192, < 16384, ≥ 16384.
230
231   lock_buckets
232       Distribution of record lock requests based on time required to obtain
233       locks. Buckets are < 1ms, < 10ms, < 100ms, < 1s, < 2s, < 4s, < 8s,
234       < 16s, < 32s, < 64s, ≥ 64s.
235
236   locks_latency
237       The minimum, the average and the maximum time (in seconds) required to
238       obtain record locks.
239
240   reclock_ctdbd
241       The minimum, the average and the maximum time (in seconds) required to
242       check if recovery lock is still held by recovery daemon when recovery
243       mode is changed. This check is done in ctdb daemon.
244
245   reclock_recd
246       The minimum, the average and the maximum time (in seconds) required to
247       check if recovery lock is still held by recovery daemon during
248       recovery. This check is done in recovery daemon.
249
250   call_latency
251       The minimum, the average and the maximum time (in seconds) required to
252       process a REQ_CALL message from client. This includes the time required
253       to migrate a record from remote node, if the record is not available on
254       the local node.
255
256   childwrite_latency
257       Default: 0
258
259       The minimum, the average and the maximum time (in seconds) required to
260       update records under a transaction.
261

DATABASE STATISTICS

263       CTDB maintains per database statistics about important operations. See
264       the ctdb(1) command dbstatistics for displaying database statistics.
265
266   Example: ctdb dbstatistics notify_index.tdb
267           DB Statistics: notify_index.tdb
268            ro_delegations                     0
269            ro_revokes                         0
270            locks
271                total                        131
272                failed                         0
273                current                        0
274                pending                        0
275            hop_count_buckets: 9890 5454 26 1 0 0 0 0 0 0 0 0 0 0 0 0
276            lock_buckets: 4 117 10 0 0 0 0 0 0 0 0 0 0 0 0 0
277            locks_latency      MIN/AVG/MAX     0.000683/0.004198/0.014730 sec out of 131
278            Num Hot Keys:     3
279                Count:7 Key:2f636c75737465726673
280                Count:18 Key:2f636c757374657266732f64617461
281                Count:7 Key:2f636c757374657266732f646174612f636c69656e7473
282
283
284   DB Statistics
285       Name of the database.
286
287   ro_delegations
288       Number of readonly delegations created in the database.
289
290   ro_revokes
291       Number of readonly delegations revoked. The difference in
292       ro_delegations and ro_revokes indicates the currently active readonly
293       delegations.
294
295   locks
296       This section lists locking statistics.
297
298       total
299           Number of completed lock calls. This includes database locks and
300           record locks.
301
302       failed
303           Number of failed lock calls. This includes database locks and
304           record locks.
305
306       current
307           Number of scheduled lock calls. This includes database locks and
308           record locks.
309
310       pending
311           Number of queued lock calls. This includes database locks and
312           record locks.
313
314   hop_count_buckets
315       Distribution of migration requests based on hop counts values. Buckets
316       are 0, < 2, < 4, < 8, < 16, < 32, < 64, < 128, < 256, < 512, < 1024,
317       < 2048, < 4096, < 8192, < 16384, ≥ 16384.
318
319   lock_buckets
320       Distribution of record lock requests based on time required to obtain
321       locks. Buckets are < 1ms, < 10ms, < 100ms, < 1s, < 2s, < 4s, < 8s,
322       < 16s, < 32s, < 64s, ≥ 64s.
323
324   locks_latency
325       The minimum, the average and the maximum time (in seconds) required to
326       obtain record locks.
327
328   Num Hot Keys
329       Number of contended records determined by hop count. CTDB keeps track
330       of top 10 hot records and the output shows hex encoded keys for the hot
331       records.
332

SEE ALSO

334       ctdb(1), ctdbd(1), ctdb-tunables(7), http://ctdb.samba.org/
335

AUTHOR

337       This documentation was written by Amitay Isaacs, Martin Schwenke
338
340       Copyright © 2007 Andrew Tridgell, Ronnie Sahlberg
341
342       This program is free software; you can redistribute it and/or modify it
343       under the terms of the GNU General Public License as published by the
344       Free Software Foundation; either version 3 of the License, or (at your
345       option) any later version.
346
347       This program is distributed in the hope that it will be useful, but
348       WITHOUT ANY WARRANTY; without even the implied warranty of
349       MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
350       General Public License for more details.
351
352       You should have received a copy of the GNU General Public License along
353       with this program; if not, see http://www.gnu.org/licenses.
354
355
356
357
358ctdb                              06/01/2021                CTDB-STATISTICS(7)
Impressum