Revision d2ac5526 hbal.1
| b/hbal.1 | ||
|---|---|---|
| 1 |
.TH HBAL 1 2009-03-14 htools "Ganeti H-tools"
|
|
| 1 |
.TH HBAL 1 2009-03-22 htools "Ganeti H-tools"
|
|
| 2 | 2 |
.SH NAME |
| 3 | 3 |
hbal \- Cluster balancer for Ganeti |
| 4 | 4 |
|
| ... | ... | |
| 7 | 7 |
.B "[-C]" |
| 8 | 8 |
.B "[-p]" |
| 9 | 9 |
.B "[-o]" |
| 10 |
.B "-l" |
|
| 11 |
.BI "[ -m " cluster "]" |
|
| 10 |
.BI "[-l" limit "]" |
|
| 11 |
.BI "[-O" name... "]" |
|
| 12 |
.BI "[-m " cluster "]" |
|
| 12 | 13 |
.BI "[-n " nodes-file " ]" |
| 13 |
.BI "[ -i " instances-file "]"
|
|
| 14 |
.BI "[-i " instances-file "]" |
|
| 14 | 15 |
|
| 15 | 16 |
.B hbal |
| 16 | 17 |
.B --version |
| ... | ... | |
| 61 | 62 |
- coefficient of variance of the percent of reserved memory |
| 62 | 63 |
- coefficient of variance of the percent of free disk |
| 63 | 64 |
- percentage of nodes failing N+1 check |
| 65 |
- percentage of instances living (either as primary or secondary) on |
|
| 66 |
offline nodes |
|
| 64 | 67 |
|
| 65 | 68 |
The free memory and free disk values help ensure that all nodes are |
| 66 | 69 |
somewhat balanced in their resource usage. The reserved memory helps |
| ... | ... | |
| 69 | 72 |
N+1. And finally, the N+1 percentage helps guide the algorithm towards |
| 70 | 73 |
eliminating N+1 failures, if possible. |
| 71 | 74 |
|
| 72 |
Except for the N+1 failures, we use the coefficient of variance since |
|
| 73 |
this brings the values into the same unit so to speak, and with a |
|
| 74 |
restrict domain of values (between zero and one). The percentage of |
|
| 75 |
N+1 failures, while also in this numeric range, doesn't actually has |
|
| 76 |
the same meaning, but it has shown to work well. |
|
| 75 |
Except for the N+1 failures and offline instances percentage, we use |
|
| 76 |
the coefficient of variance since this brings the values into the same |
|
| 77 |
unit so to speak, and with a restrict domain of values (between zero |
|
| 78 |
and one). The percentage of N+1 failures, while also in this numeric |
|
| 79 |
range, doesn't actually has the same meaning, but it has shown to work |
|
| 80 |
well. |
|
| 77 | 81 |
|
| 78 | 82 |
The other alternative, using for N+1 checks the coefficient of |
| 79 | 83 |
variance of (N+1 fail=1, N+1 pass=0) across nodes could hint the |
| ... | ... | |
| 82 | 86 |
rules of the algorithm, so the N+1 checks would simply not work |
| 83 | 87 |
anymore in this case. |
| 84 | 88 |
|
| 89 |
The offline instances percentage (meaning the percentage of instances |
|
| 90 |
living on offline nodes) will cause the algorithm to actively move |
|
| 91 |
instances away from offline nodes. This, coupled with the restriction |
|
| 92 |
on placement given by offline nodes, will cause evacuation of such |
|
| 93 |
nodes. |
|
| 94 |
|
|
| 85 | 95 |
On a perfectly balanced cluster (all nodes the same size, all |
| 86 | 96 |
instances the same size and spread across the nodes equally), all |
| 87 | 97 |
values would be zero. This doesn't happen too often in practice :) |
| ... | ... | |
| 106 | 116 |
the user to understand the node's most important parameters. |
| 107 | 117 |
|
| 108 | 118 |
The node list will contain these informations: |
| 109 |
- a character denoting the status of the node, with '-' meaning an |
|
| 110 |
offline node, '*' meaning N+1 failure and blank meaning a good |
|
| 111 |
node |
|
| 112 |
- the node name |
|
| 113 |
- the total node memory |
|
| 114 |
- the memory used by the node itself |
|
| 115 |
- the free node memory |
|
| 116 |
- the reserved node memory, which is the amount of free memory |
|
| 117 |
needed for N+1 compliance |
|
| 118 |
- total disk |
|
| 119 |
- free disk |
|
| 120 |
- number of primary instances |
|
| 121 |
- number of secondary instances |
|
| 122 |
- percent of free memory |
|
| 123 |
- percent of free disk |
|
| 119 |
.RS |
|
| 120 |
.TP |
|
| 121 |
.B F |
|
| 122 |
a character denoting the status of the node, with '-' meaning an |
|
| 123 |
offline node, '*' meaning N+1 failure and blank meaning a good node |
|
| 124 |
.TP |
|
| 125 |
.B Name |
|
| 126 |
the node name |
|
| 127 |
.TP |
|
| 128 |
.B t_mem |
|
| 129 |
the total node memory |
|
| 130 |
.TP |
|
| 131 |
.B n_mem |
|
| 132 |
the memory used by the node itself |
|
| 133 |
.TP |
|
| 134 |
.B i_mem |
|
| 135 |
the memory used by instances |
|
| 136 |
.TP |
|
| 137 |
.B x_mem |
|
| 138 |
amount memory which seems to be in use but cannot be determined why or |
|
| 139 |
by which instance; usually this means that the hypervisor has some |
|
| 140 |
overhead or that there are other reporting errors |
|
| 141 |
.TP |
|
| 142 |
.B f_mem |
|
| 143 |
the free node memory |
|
| 144 |
.TP |
|
| 145 |
.B r_mem |
|
| 146 |
the reserved node memory, which is the amount of free memory needed |
|
| 147 |
for N+1 compliance |
|
| 148 |
.TP |
|
| 149 |
.B t_dsk |
|
| 150 |
total disk |
|
| 151 |
.TP |
|
| 152 |
.B f_dsk |
|
| 153 |
free disk |
|
| 154 |
.TP |
|
| 155 |
.B pri |
|
| 156 |
number of primary instances |
|
| 157 |
.TP |
|
| 158 |
.B sec |
|
| 159 |
number of secondary instances |
|
| 160 |
.TP |
|
| 161 |
.B p_fmem |
|
| 162 |
percent of free memory |
|
| 163 |
.TP |
|
| 164 |
.B p_fdsk |
|
| 165 |
percent of free disk |
|
| 166 |
.RE |
|
| 124 | 167 |
|
| 125 | 168 |
.TP |
| 126 | 169 |
.B -o, --oneline |
| ... | ... | |
| 135 | 178 |
- improvement in the cluster score |
| 136 | 179 |
|
| 137 | 180 |
.TP |
| 181 |
.BI "-O " name |
|
| 182 |
This option (which can be given multiple times) will mark nodes as |
|
| 183 |
being \fIoffline\fR. This means a couple of things: |
|
| 184 |
.RS |
|
| 185 |
.TP |
|
| 186 |
- |
|
| 187 |
instances won't be placed on these nodes, not even temporarily; |
|
| 188 |
e.g. the \fIreplace primary\fR move is not available if the secondary |
|
| 189 |
node is offline, since this move requires a failover. |
|
| 190 |
.TP |
|
| 191 |
- |
|
| 192 |
these nodes will not be included in the score calculation (except for |
|
| 193 |
the percentage of instances on offline nodes) |
|
| 194 |
.RE |
|
| 195 |
|
|
| 196 |
.TP |
|
| 138 | 197 |
.BI "-n" nodefile ", --nodes=" nodefile |
| 139 | 198 |
The name of the file holding node information (if not collecting via |
| 140 | 199 |
RAPI), instead of the default |
| ... | ... | |
| 188 | 247 |
|
| 189 | 248 |
.SH EXAMPLE |
| 190 | 249 |
|
| 250 |
Note that this example are not for the latest version (they don't have |
|
| 251 |
full node data). |
|
| 252 |
|
|
| 191 | 253 |
.SS Default output |
| 192 | 254 |
|
| 193 | 255 |
With the default options, the program shows each individual step and |
| ... | ... | |
| 362 | 424 |
list (but hopefully will end in the same state). |
| 363 | 425 |
|
| 364 | 426 |
.SH SEE ALSO |
| 365 |
hn1(1), ganeti(7), gnt-instance(8), gnt-node(8) |
|
| 427 |
.BR hn1 "(1), " hscan "(1), " ganeti "(7), " gnt-instance "(8), " |
|
| 428 |
.BR gnt-node "(8)" |
|
Also available in: Unified diff