]>
Commit | Line | Data |
---|---|---|
7c673cae FG |
1 | ====================================== |
2 | Pool, PG and CRUSH Config Reference | |
3 | ====================================== | |
4 | ||
5 | .. index:: pools; configuration | |
6 | ||
7 | When you create pools and set the number of placement groups for the pool, Ceph | |
8 | uses default values when you don't specifically override the defaults. **We | |
11fdf7f2 | 9 | recommend** overriding some of the defaults. Specifically, we recommend setting |
7c673cae FG |
10 | a pool's replica size and overriding the default number of placement groups. You |
11 | can specifically set these values when running `pool`_ commands. You can also | |
12 | override the defaults by adding new ones in the ``[global]`` section of your | |
11fdf7f2 | 13 | Ceph configuration file. |
7c673cae FG |
14 | |
15 | ||
16 | .. literalinclude:: pool-pg.conf | |
17 | :language: ini | |
18 | ||
19 | ||
20 | ||
21 | ``mon max pool pg num`` | |
22 | ||
23 | :Description: The maximum number of placement groups per pool. | |
24 | :Type: Integer | |
25 | :Default: ``65536`` | |
26 | ||
27 | ||
11fdf7f2 | 28 | ``mon pg create interval`` |
7c673cae | 29 | |
11fdf7f2 | 30 | :Description: Number of seconds between PG creation in the same |
7c673cae FG |
31 | Ceph OSD Daemon. |
32 | ||
33 | :Type: Float | |
34 | :Default: ``30.0`` | |
35 | ||
36 | ||
11fdf7f2 | 37 | ``mon pg stuck threshold`` |
7c673cae | 38 | |
11fdf7f2 | 39 | :Description: Number of seconds after which PGs can be considered as |
7c673cae FG |
40 | being stuck. |
41 | ||
42 | :Type: 32-bit Integer | |
43 | :Default: ``300`` | |
44 | ||
31f18b77 | 45 | ``mon pg min inactive`` |
7c673cae | 46 | |
31f18b77 FG |
47 | :Description: Issue a ``HEALTH_ERR`` in cluster log if the number of PGs stay |
48 | inactive longer than ``mon_pg_stuck_threshold`` exceeds this | |
49 | setting. A non-positive number means disabled, never go into ERR. | |
50 | :Type: Integer | |
51 | :Default: ``1`` | |
52 | ||
53 | ||
54 | ``mon pg warn min per osd`` | |
55 | ||
56 | :Description: Issue a ``HEALTH_WARN`` in cluster log if the average number | |
57 | of PGs per (in) OSD is under this number. (a non-positive number | |
58 | disables this) | |
59 | :Type: Integer | |
60 | :Default: ``30`` | |
61 | ||
62 | ||
31f18b77 FG |
63 | ``mon pg warn min objects`` |
64 | ||
65 | :Description: Do not warn if the total number of objects in cluster is below | |
66 | this number | |
67 | :Type: Integer | |
68 | :Default: ``1000`` | |
69 | ||
70 | ||
71 | ``mon pg warn min pool objects`` | |
72 | ||
73 | :Description: Do not warn on pools whose object number is below this number | |
74 | :Type: Integer | |
75 | :Default: ``1000`` | |
76 | ||
77 | ||
78 | ``mon pg check down all threshold`` | |
79 | ||
80 | :Description: Threshold of down OSDs percentage after which we check all PGs | |
81 | for stale ones. | |
82 | :Type: Float | |
83 | :Default: ``0.5`` | |
84 | ||
85 | ||
86 | ``mon pg warn max object skew`` | |
87 | ||
88 | :Description: Issue a ``HEALTH_WARN`` in cluster log if the average object number | |
89 | of a certain pool is greater than ``mon pg warn max object skew`` times | |
9f95a23c TL |
90 | the average object number of the whole pool. (zero or a non-positive |
91 | number disables this). Note that this option applies to the managers. | |
31f18b77 FG |
92 | :Type: Float |
93 | :Default: ``10`` | |
94 | ||
95 | ||
96 | ``mon delta reset interval`` | |
97 | ||
98 | :Description: Seconds of inactivity before we reset the pg delta to 0. We keep | |
99 | track of the delta of the used space of each pool, so, for | |
100 | example, it would be easier for us to understand the progress of | |
101 | recovery or the performance of cache tier. But if there's no | |
102 | activity reported for a certain pool, we just reset the history of | |
103 | deltas of that pool. | |
104 | :Type: Integer | |
105 | :Default: ``10`` | |
106 | ||
107 | ||
108 | ``mon osd max op age`` | |
109 | ||
110 | :Description: Maximum op age before we get concerned (make it a power of 2). | |
111 | A ``HEALTH_WARN`` will be issued if a request has been blocked longer | |
112 | than this limit. | |
113 | :Type: Float | |
114 | :Default: ``32.0`` | |
115 | ||
116 | ||
117 | ``osd pg bits`` | |
7c673cae FG |
118 | |
119 | :Description: Placement group bits per Ceph OSD Daemon. | |
120 | :Type: 32-bit Integer | |
11fdf7f2 | 121 | :Default: ``6`` |
7c673cae FG |
122 | |
123 | ||
31f18b77 | 124 | ``osd pgp bits`` |
7c673cae FG |
125 | |
126 | :Description: The number of bits per Ceph OSD Daemon for PGPs. | |
127 | :Type: 32-bit Integer | |
128 | :Default: ``6`` | |
129 | ||
130 | ||
131 | ``osd crush chooseleaf type`` | |
132 | ||
11fdf7f2 | 133 | :Description: The bucket type to use for ``chooseleaf`` in a CRUSH rule. Uses |
7c673cae FG |
134 | ordinal rank rather than name. |
135 | ||
136 | :Type: 32-bit Integer | |
137 | :Default: ``1``. Typically a host containing one or more Ceph OSD Daemons. | |
138 | ||
139 | ||
140 | ``osd crush initial weight`` | |
141 | ||
142 | :Description: The initial crush weight for newly added osds into crushmap. | |
143 | ||
144 | :Type: Double | |
145 | :Default: ``the size of newly added osd in TB``. By default, the initial crush | |
146 | weight for the newly added osd is set to its volume size in TB. | |
147 | See `Weighting Bucket Items`_ for details. | |
148 | ||
149 | ||
b32b8144 | 150 | ``osd pool default crush rule`` |
7c673cae | 151 | |
b32b8144 | 152 | :Description: The default CRUSH rule to use when creating a replicated pool. |
7c673cae | 153 | :Type: 8-bit Integer |
b32b8144 FG |
154 | :Default: ``-1``, which means "pick the rule with the lowest numerical ID and |
155 | use that". This is to make pool creation work in the absence of rule 0. | |
7c673cae FG |
156 | |
157 | ||
158 | ``osd pool erasure code stripe unit`` | |
159 | ||
160 | :Description: Sets the default size, in bytes, of a chunk of an object | |
161 | stripe for erasure coded pools. Every object of size S | |
162 | will be stored as N stripes, with each data chunk | |
163 | receiving ``stripe unit`` bytes. Each stripe of ``N * | |
164 | stripe unit`` bytes will be encoded/decoded | |
165 | individually. This option can is overridden by the | |
166 | ``stripe_unit`` setting in an erasure code profile. | |
167 | ||
168 | :Type: Unsigned 32-bit Integer | |
169 | :Default: ``4096`` | |
170 | ||
171 | ||
172 | ``osd pool default size`` | |
173 | ||
174 | :Description: Sets the number of replicas for objects in the pool. The default | |
175 | value is the same as | |
176 | ``ceph osd pool set {pool-name} size {size}``. | |
177 | ||
178 | :Type: 32-bit Integer | |
179 | :Default: ``3`` | |
180 | ||
181 | ||
182 | ``osd pool default min size`` | |
183 | ||
11fdf7f2 TL |
184 | :Description: Sets the minimum number of written replicas for objects in the |
185 | pool in order to acknowledge a write operation to the client. If | |
186 | minimum is not met, Ceph will not acknowledge the write to the | |
187 | client, **which may result in data loss**. This setting ensures | |
188 | a minimum number of replicas when operating in ``degraded`` mode. | |
7c673cae FG |
189 | |
190 | :Type: 32-bit Integer | |
11fdf7f2 | 191 | :Default: ``0``, which means no particular minimum. If ``0``, |
7c673cae FG |
192 | minimum is ``size - (size / 2)``. |
193 | ||
194 | ||
11fdf7f2 | 195 | ``osd pool default pg num`` |
7c673cae | 196 | |
11fdf7f2 | 197 | :Description: The default number of placement groups for a pool. The default |
7c673cae FG |
198 | value is the same as ``pg_num`` with ``mkpool``. |
199 | ||
200 | :Type: 32-bit Integer | |
92f5a8d4 | 201 | :Default: ``32`` |
7c673cae FG |
202 | |
203 | ||
11fdf7f2 | 204 | ``osd pool default pgp num`` |
7c673cae | 205 | |
11fdf7f2 TL |
206 | :Description: The default number of placement groups for placement for a pool. |
207 | The default value is the same as ``pgp_num`` with ``mkpool``. | |
7c673cae FG |
208 | PG and PGP should be equal (for now). |
209 | ||
210 | :Type: 32-bit Integer | |
211 | :Default: ``8`` | |
212 | ||
213 | ||
214 | ``osd pool default flags`` | |
215 | ||
11fdf7f2 | 216 | :Description: The default flags for new pools. |
7c673cae FG |
217 | :Type: 32-bit Integer |
218 | :Default: ``0`` | |
219 | ||
220 | ||
221 | ``osd max pgls`` | |
222 | ||
11fdf7f2 | 223 | :Description: The maximum number of placement groups to list. A client |
7c673cae FG |
224 | requesting a large number can tie up the Ceph OSD Daemon. |
225 | ||
226 | :Type: Unsigned 64-bit Integer | |
227 | :Default: ``1024`` | |
228 | :Note: Default should be fine. | |
229 | ||
230 | ||
11fdf7f2 | 231 | ``osd min pg log entries`` |
7c673cae | 232 | |
11fdf7f2 | 233 | :Description: The minimum number of placement group logs to maintain |
7c673cae FG |
234 | when trimming log files. |
235 | ||
236 | :Type: 32-bit Int Unsigned | |
9f95a23c TL |
237 | :Default: ``250`` |
238 | ||
239 | ||
240 | ``osd max pg log entries`` | |
241 | ||
242 | :Description: The maximum number of placement group logs to maintain | |
243 | when trimming log files. | |
244 | ||
245 | :Type: 32-bit Int Unsigned | |
246 | :Default: ``10000`` | |
7c673cae FG |
247 | |
248 | ||
249 | ``osd default data pool replay window`` | |
250 | ||
251 | :Description: The time (in seconds) for an OSD to wait for a client to replay | |
252 | a request. | |
253 | ||
254 | :Type: 32-bit Integer | |
255 | :Default: ``45`` | |
256 | ||
3efd9988 FG |
257 | ``osd max pg per osd hard ratio`` |
258 | ||
259 | :Description: The ratio of number of PGs per OSD allowed by the cluster before | |
260 | OSD refuses to create new PGs. OSD stops creating new PGs if the number | |
261 | of PGs it serves exceeds | |
262 | ``osd max pg per osd hard ratio`` \* ``mon max pg per osd``. | |
263 | ||
264 | :Type: Float | |
265 | :Default: ``2`` | |
7c673cae | 266 | |
11fdf7f2 TL |
267 | ``osd recovery priority`` |
268 | ||
269 | :Description: Priority of recovery in the work queue. | |
270 | ||
271 | :Type: Integer | |
272 | :Default: ``5`` | |
273 | ||
274 | ``osd recovery op priority`` | |
275 | ||
276 | :Description: Default priority used for recovery operations if pool doesn't override. | |
277 | ||
278 | :Type: Integer | |
279 | :Default: ``3`` | |
280 | ||
7c673cae FG |
281 | .. _pool: ../../operations/pools |
282 | .. _Monitoring OSDs and PGs: ../../operations/monitoring-osd-pg#peering | |
283 | .. _Weighting Bucket Items: ../../operations/crush-map#weightingbucketitems |