1ETCD(1) Nov 2017 ETCD(1)
2
3
4
6 etcd - Distributed reliable key-value store for the most critical data
7 of a distributed system
8
9
11 etcd [flags]
12
13
15 Etcd is a distributed key-value store designed to reliably and quickly
16 preserve and provide access to critical data. It enables reliable dis‐
17 tributed coordination through distributed locking, leader elections,
18 and write barriers. An etcd cluster is intended for high availability
19 and permanent data storage and retrieval.
20
21
23 Member flags
24
25 --data-dir
26 Path to the data directory.
27
28 --wal-dir
29 Path to the dedicated wal directory.
30
31 --listen-peer-urls http://localhost:2380
32 List of URLs to listen on for peer traffic.
33
34 --listen-client-urls http://localhost:2379
35 List of URLs to listen on for client traffic.
36
37 --listen-metrics-urls
38 List of URLs to listen on for the metrics and health end‐
39 points.
40
41 --max-snapshots 5
42 Maximum number of snapshot files to retain (0 is unlim‐
43 ited).
44
45 --max-wals 5
46 Maximum number of wal files to retain (0 is unlimited).
47
48 --name default
49 Human-readable name for this member.
50
51 --snapshot-count 100000
52 Number of committed transactions to trigger a snapshot to
53 disk.
54
55 --heartbeat-interval 100
56 Time (in milliseconds) of a heartbeat interval.
57
58 --election-timeout 1000
59 Time (in milliseconds) for an election to timeout.
60
61 --initial-election-tick-advance true
62 Whether to fast-forward initial election ticks on boot for
63 faster election.
64
65 --quota-backend-bytes 0
66 Raise alarms when backend size exceeds the given quota. 0
67 means use the default quota.
68
69 --backend-batch-interval 0s
70 BackendBatchInterval is the maximum time before commit the
71 backend transaction.
72
73 --backend-batch-limit 0
74 BackendBatchLimit is the maximum operations before commit
75 the backend transaction.
76
77 --max-txn-ops 128
78 Maximum number of operations permitted in a transaction.
79
80 --max-request-bytes 1572864
81 Maximum client request size in bytes the server will
82 accept.
83
84 --grpc-keepalive-min-time 5s
85 Minimum interval duration that a client should wait before
86 pinging server.
87
88 --grpc-keepalive-interval 2h0m0s
89 Frequency duration of server-to-client ping to check if a
90 connection is alive (0 to disable).
91
92 --grpc-keepalive-timeout 20s
93 Additional duration of wait before closing a non-respon‐
94 sive connection (0 to disable).
95
96 Clustering flags
97
98 --initial-advertise-peer-urls http://localhost:2380
99 List of this member's peer URLs to advertise to the rest
100 of the cluster.
101
102 --advertise-client-urls http://localhost:2379
103 List of this member's client URLs to advertise to the pub‐
104 lic.
105
106 --discovery
107 Discovery URL used to bootstrap the cluster.
108
109 --discovery-fallback proxy
110 Valid values include ["exit" "proxy"]
111
112 --discovery-proxy
113 HTTP proxy to use for traffic to discovery service.
114
115 --discovery-srv
116 DNS domain used to bootstrap initial cluster.
117
118 --discovery-srv-name
119 Service name to query when using DNS discovery.
120
121 --initial-cluster default=http://localhost:2380
122 Initial cluster configuration for bootstrapping.
123
124 --initial-cluster-token etcd-cluster
125 Initial cluster token for the etcd cluster during boot‐
126 strap.
127
128 --initial-cluster-state new
129 Initial cluster state ('new' or 'existing').
130
131 --strict-reconfig-check true
132 Reject reconfiguration requests that would cause quorum
133 loss.
134
135 --enable-v2 false
136 Accept etcd V2 client requests.
137
138 --pre-vote false
139 Enable to run an additional Raft election phase.
140
141 Proxy flags
142
143 --proxy off
144 Valid values include ["off" "on" "readonly"]
145
146 --proxy-failure-wait 5000
147 Time (in milliseconds) an endpoint will be held in a
148 failed state.
149
150 --proxy-refresh-interval 30000
151 Time (in milliseconds) of the endpoints refresh interval.
152
153 --proxy-dial-timeout 1000
154 Time (in milliseconds) for a dial to timeout.
155
156 --proxy-write-timeout 5000
157 Time (in milliseconds) for a write to timeout.
158
159 --proxy-read-timeout 0
160 Time (in milliseconds) for a read to timeout.
161
162 Security flags
163
164 --cert-file
165 Path to the client server TLS cert file.
166
167 --key-file
168 Path to the client server TLS key file.
169
170 --client-cert-auth false
171 Enable client cert authentication.
172
173 --client-crl-file
174 Path to the client certificate revocation list file.
175
176 --client-cert-allowed-hostname
177 Allowed TLS hostname for client cert authentication.
178
179 --trusted-ca-file
180 Path to the client server TLS trusted CA cert file.
181
182 --auto-tls false
183 Client TLS using generated certificates
184
185 --peer-cert-file
186 Path to the peer server TLS cert file.
187
188 --peer-key-file
189 Path to the peer server TLS key file.
190
191 --peer-client-cert-auth false
192 Enable peer client cert authentication.
193
194 --peer-trusted-ca-file
195 Path to the peer server TLS trusted CA file.
196
197 --peer-auto-tls false
198 Peer TLS using generated certificates
199
200 --peer-crl-file
201 Path to the peer certificate revocation list file.
202
203 --peer-cert-allowed-cn
204 Allowed CN for inter peer authentication.
205
206 --peer-cert-allowed-hostname
207 Allowed TLS hostname for inter peer authentication.
208
209 --cipher-suites
210 Comma-separated list of supported TLS cipher suites
211 between client/server and peers (empty will be auto-populated by
212 Go).
213
214 --experimental-peer-skip-client-san-verification false
215 Skip verification of SAN field in client certificate for
216 peer connections.
217
218 --cors *
219 Comma-separated white list of origins for CORS, or cross-
220 origin resource sharing, (empty or * means allow all)
221
222 --host-whitelist *
223 Comma-separated acceptable hostnames from HTTP client
224 requests, if server is not secure (empty means allow all).
225
226 Logging flags
227
228 --logger capnslog
229 Specify 'zap' for structured logging or 'capnslog'. WARN:
230 'capnslog' is being deprecated in v3.5.
231
232 --log-output default
233 [TO BE DEPRECATED IN v3.5] use '--log-outputs'.
234
235 --log-outputs default
236 Specify 'stdout' or 'stderr' to skip journald logging even
237 when running under systemd, or list of comma separated output
238 targets.
239
240 --debug false
241 [TO BE DEPRECATED IN v3.5] Enable debug-level logging for
242 etcd. Use '--log-level=debug' instead.
243
244 --log-level info
245 Configures log level. Only supports debug, info, warn,
246 error, panic, or fatal. Default 'info'.
247
248 --log-package-levels
249 [TO BE DEPRECATED IN v3.5] Specify a particular log level
250 for each etcd package (eg: 'etcdmain=CRITICAL,etcd‐
251 server=DEBUG').
252
253 Version flags
254
255 --version false
256 Print the version and exit.
257
258 --auto-compaction-retention 0
259 Auto compaction retention for mvcc key value store. 0
260 means disable auto compaction.
261
262 --auto-compaction-mode periodic
263 interpret 'auto-compaction-retention' one of: peri‐
264 odic|revision. 'periodic' for duration based retention, default‐
265 ing to hours if no time unit is provided (e.g. '5m'). 'revision'
266 for revision number based retention.
267
268 Profiling flags
269
270 --enable-pprof false
271 Enable runtime profiling data via HTTP server. Address is
272 at client URL + "/debug/pprof/"
273
274 --metrics basic
275 Set level of detail for exported metrics, specify 'exten‐
276 sive' to include histogram metrics
277
278 Auth flags
279
280 --auth-token simple
281 Specify auth token specific options.
282
283 --bcrypt-cost 10
284 Specify bcrypt algorithm cost factor for auth password
285 hashing.
286
287 Gateway flags
288
289 --enable-grpc-gateway true
290 Enable GRPC gateway.
291
292 Experimental flags
293
294 --experimental-initial-corrupt-check false
295 Enable to check data corruption before serving any
296 client/peer traffic.
297
298 --experimental-corrupt-check-time 0s
299 Duration of time between cluster corruption check passes.
300
301 --experimental-enable-v2v3
302 v3 prefix for serving emulated v2 state.
303
304 --experimental-backend-bbolt-freelist-type
305 ExperimentalBackendFreelistType specifies the type of
306 freelist that boltdb backend uses(array and map are supported
307 types)
308
309 --experimental-enable-lease-checkpoint false
310 Enable to persist lease remaining TTL to prevent indefi‐
311 nite auto-renewal of long lived leases.
312
313 --experimental-compaction-batch-limit 0
314 Sets the maximum revisions deleted in each compaction
315 batch.
316
317 Unsafe flags
318
319 --force-new-cluster false
320 Force to create a new one member cluster.
321
322
324 etcdctl(1), etcdctl2(1), etcdctl3(1)
325
326
327
328
329Etcd contributors etcd User Manuals ETCD(1)