Statistics
| Branch: | Tag: | Revision:

root / hbal.1 @ d09b6ed3

History | View | Annotate | Download (17.4 kB)

1
.TH HBAL 1 2009-03-23 htools "Ganeti H-tools"
2
.SH NAME
3
hbal \- Cluster balancer for Ganeti
4

    
5
.SH SYNOPSIS
6
.B hbal
7
.B "[-C]"
8
.B "[-p]"
9
.B "[-o]"
10
.B "[-v... | -q]"
11
.BI "[-l" limit "]"
12
.BI "[-O" name... "]"
13
.BI "[-e" score "]"
14
.BI "[-m " cluster "]"
15
.BI "[-n " nodes-file " ]"
16
.BI "[-i " instances-file "]"
17

    
18
.B hbal
19
.B --version
20

    
21
.SH DESCRIPTION
22
hbal is a cluster balancer that looks at the current state of the
23
cluster (nodes with their total and free disk, memory, etc.) and
24
instance placement and computes a series of steps designed to bring
25
the cluster into a better state.
26

    
27
The algorithm to do so is designed to be stable (i.e. it will give you
28
the same results when restarting it from the middle of the solution)
29
and reasonably fast. It is not, however, designed to be a perfect
30
algorithm - it is possible to make it go into a corner from which it
31
can find no improvement, because it only look one "step" ahead.
32

    
33
By default, the program will show the solution incrementally as it is
34
computed, in a somewhat cryptic format; for getting the actual Ganeti
35
command list, use the \fB-C\fR option.
36

    
37
.SS ALGORITHM
38

    
39
The program works in independent steps; at each step, we compute the
40
best instance move that lowers the cluster score.
41

    
42
The possible move type for an instance are combinations of
43
failover/migrate and replace-disks such that we change one of the
44
instance nodes, and the other one remains (but possibly with changed
45
role, e.g. from primary it becomes secondary). The list is:
46
.RS 4
47
.TP 3
48
\(em
49
failover (f)
50
.TP
51
\(em
52
replace secondary (r)
53
.TP
54
\(em
55
replace primary, a composite move (f, r, f)
56
.TP
57
\(em
58
failover and replace secondary, also composite (f, r)
59
.TP
60
\(em
61
replace secondary and failover, also composite (r, f)
62
.RE
63

    
64
We don't do the only remaining possibility of replacing both nodes
65
(r,f,r,f or the equivalent f,r,f,r) since these move needs an
66
exhaustive search over both candidate primary and secondary nodes, and
67
is O(n*n) in the number of nodes. Furthermore, it doesn't seems to
68
give better scores but will result in more disk replacements.
69

    
70
.SS CLUSTER SCORING
71

    
72
As said before, the algorithm tries to minimise the cluster score at
73
each step. Currently this score is computed as a sum of the following
74
components:
75
.RS 4
76
.TP 3
77
\(em
78
coefficient of variance of the percent of free memory
79
.TP
80
\(em
81
coefficient of variance of the percent of reserved memory
82
.TP
83
\(em
84
coefficient of variance of the percent of free disk
85
.TP
86
\(em
87
percentage of nodes failing N+1 check
88
.TP
89
\(em
90
percentage of instances living (either as primary or secondary) on
91
offline nodes
92
.RE
93

    
94
The free memory and free disk values help ensure that all nodes are
95
somewhat balanced in their resource usage. The reserved memory helps
96
to ensure that nodes are somewhat balanced in holding secondary
97
instances, and that no node keeps too much memory reserved for
98
N+1. And finally, the N+1 percentage helps guide the algorithm towards
99
eliminating N+1 failures, if possible.
100

    
101
Except for the N+1 failures and offline instances percentage, we use
102
the coefficient of variance since this brings the values into the same
103
unit so to speak, and with a restrict domain of values (between zero
104
and one). The percentage of N+1 failures, while also in this numeric
105
range, doesn't actually has the same meaning, but it has shown to work
106
well.
107

    
108
The other alternative, using for N+1 checks the coefficient of
109
variance of (N+1 fail=1, N+1 pass=0) across nodes could hint the
110
algorithm to make more N+1 failures if most nodes are N+1 fail
111
already. Since this (making N+1 failures) is not allowed by other
112
rules of the algorithm, so the N+1 checks would simply not work
113
anymore in this case.
114

    
115
The offline instances percentage (meaning the percentage of instances
116
living on offline nodes) will cause the algorithm to actively move
117
instances away from offline nodes. This, coupled with the restriction
118
on placement given by offline nodes, will cause evacuation of such
119
nodes.
120

    
121
On a perfectly balanced cluster (all nodes the same size, all
122
instances the same size and spread across the nodes equally), all
123
values would be zero. This doesn't happen too often in practice :)
124

    
125
.SS OFFLINE INSTANCES
126

    
127
Since current Ganeti versions do not report the memory used by offline
128
(down) instances, ignoring the run status of instances will cause
129
wrong calculations. For this reason, the algorithm subtracts the
130
memory size of down instances from the free node memory of their
131
primary node, in effect simulating the startup of such instances.
132

    
133
.SS OTHER POSSIBLE METRICS
134

    
135
It would be desirable to add more metrics to the algorithm, especially
136
dynamically-computed metrics, such as:
137
.RS 4
138
.TP 3
139
\(em
140
CPU usage of instances, combined with VCPU versus PCPU count
141
.TP
142
\(em
143
Disk IO usage
144
.TP
145
\(em
146
Network IO
147
.RE
148

    
149
.SH OPTIONS
150
The options that can be passed to the program are as follows:
151
.TP
152
.B -C, --print-commands
153
Print the command list at the end of the run. Without this, the
154
program will only show a shorter, but cryptic output.
155
.TP
156
.B -p, --print-nodes
157
Prints the before and after node status, in a format designed to allow
158
the user to understand the node's most important parameters.
159

    
160
The node list will contain these informations:
161
.RS
162
.TP
163
.B F
164
a character denoting the status of the node, with '-' meaning an
165
offline node, '*' meaning N+1 failure and blank meaning a good node
166
.TP
167
.B Name
168
the node name
169
.TP
170
.B t_mem
171
the total node memory
172
.TP
173
.B n_mem
174
the memory used by the node itself
175
.TP
176
.B i_mem
177
the memory used by instances
178
.TP
179
.B x_mem
180
amount memory which seems to be in use but cannot be determined why or
181
by which instance; usually this means that the hypervisor has some
182
overhead or that there are other reporting errors
183
.TP
184
.B f_mem
185
the free node memory
186
.TP
187
.B r_mem
188
the reserved node memory, which is the amount of free memory needed
189
for N+1 compliance
190
.TP
191
.B t_dsk
192
total disk
193
.TP
194
.B f_dsk
195
free disk
196
.TP
197
.B pri
198
number of primary instances
199
.TP
200
.B sec
201
number of secondary instances
202
.TP
203
.B p_fmem
204
percent of free memory
205
.TP
206
.B p_fdsk
207
percent of free disk
208
.RE
209

    
210
.TP
211
.B -o, --oneline
212
Only shows a one-line output from the program, designed for the case
213
when one wants to look at multiple clusters at once and check their
214
status.
215

    
216
The line will contain four fields:
217
.RS
218
.RS 4
219
.TP 3
220
\(em
221
initial cluster score
222
.TP
223
\(em
224
number of steps in the solution
225
.TP
226
\(em
227
final cluster score
228
.TP
229
\(em
230
improvement in the cluster score
231
.RE
232
.RE
233

    
234
.TP
235
.BI "-O " name
236
This option (which can be given multiple times) will mark nodes as
237
being \fIoffline\fR. This means a couple of things:
238
.RS
239
.RS 4
240
.TP 3
241
\(em
242
instances won't be placed on these nodes, not even temporarily;
243
e.g. the \fIreplace primary\fR move is not available if the secondary
244
node is offline, since this move requires a failover.
245
.TP
246
\(em
247
these nodes will not be included in the score calculation (except for
248
the percentage of instances on offline nodes)
249
.RE
250
.RE
251

    
252
.TP
253
.BI "-e" score ", --min-score=" score
254
This parameter denotes the minimum score we are happy with and alters
255
the computation in two ways:
256
.RS
257
.RS 4
258
.TP 3
259
\(em
260
if the cluster has the initial score lower than this value, then we
261
don't enter the algorithm at all, and exit with success
262
.TP
263
\(em
264
during the iterative process, if we reach a score lower than this
265
value, we exit the algorithm
266
.RE
267
The default value of the parameter is currently \fI1e-9\fR (chosen
268
empirically).
269
.RE
270

    
271
.TP
272
.BI "-n" nodefile ", --nodes=" nodefile
273
The name of the file holding node information (if not collecting via
274
RAPI), instead of the default \fInodes\fR file (but see below how to
275
customize the default value via the environment).
276

    
277
.TP
278
.BI "-i" instancefile ", --instances=" instancefile
279
The name of the file holding instance information (if not collecting
280
via RAPI), instead of the default \fIinstances\fR file (but see below
281
how to customize the default value via the environment).
282

    
283
.TP
284
.BI "-m" cluster
285
Collect data not from files but directly from the
286
.I cluster
287
given as an argument via RAPI. This work for both Ganeti 1.2 and
288
Ganeti 2.0.
289

    
290
.TP
291
.BI "-l" N ", --max-length=" N
292
Restrict the solution to this length. This can be used for example to
293
automate the execution of the balancing.
294

    
295
.TP
296
.B -v, --verbose
297
Increase the output verbosity. Each usage of this option will increase
298
the verbosity (currently more than 2 doesn't make sense) from the
299
default of one.
300

    
301
.TP
302
.B -q, --quiet
303
Decrease the output verbosity. Each usage of this option will decrease
304
the verbosity (less than zero doesn't make sense) from the default of
305
one.
306

    
307
.TP
308
.B -V, --version
309
Just show the program version and exit.
310

    
311
.SH EXIT STATUS
312

    
313
The exist status of the command will be zero, unless for some reason
314
the algorithm fatally failed (e.g. wrong node or instance data).
315

    
316
.SH ENVIRONMENT
317

    
318
If the variables \fBHTOOLS_NODES\fR and \fBHTOOLS_INSTANCES\fR are
319
present in the environment, they will override the default names for
320
the nodes and instances files. These will have of course no effect
321
when RAPI is used.
322

    
323
.SH BUGS
324

    
325
The program does not check its input data for consistency, and aborts
326
with cryptic errors messages in this case.
327

    
328
The algorithm is not perfect.
329

    
330
The algorithm doesn't deal with non-\fBdrbd\fR instances, and chokes
331
on input data which has such instances.
332

    
333
The output format is not easily scriptable, and the program should
334
feed moves directly into Ganeti (either via RAPI or via a gnt-debug
335
input file).
336

    
337
.SH EXAMPLE
338

    
339
Note that this example are not for the latest version (they don't have
340
full node data).
341

    
342
.SS Default output
343

    
344
With the default options, the program shows each individual step and
345
the improvements it brings in cluster score:
346

    
347
.in +4n
348
.nf
349
.RB "$" " hbal"
350
Loaded 20 nodes, 80 instances
351
Cluster is not N+1 happy, continuing but no guarantee that the cluster will end N+1 happy.
352
Initial score: 0.52329131
353
Trying to minimize the CV...
354
    1. instance14  node1:node10  => node16:node10 0.42109120 a=f r:node16 f
355
    2. instance54  node4:node15  => node16:node15 0.31904594 a=f r:node16 f
356
    3. instance4   node5:node2   => node2:node16  0.26611015 a=f r:node16
357
    4. instance48  node18:node20 => node2:node18  0.21361717 a=r:node2 f
358
    5. instance93  node19:node18 => node16:node19 0.16166425 a=r:node16 f
359
    6. instance89  node3:node20  => node2:node3   0.11005629 a=r:node2 f
360
    7. instance5   node6:node2   => node16:node6  0.05841589 a=r:node16 f
361
    8. instance94  node7:node20  => node20:node16 0.00658759 a=f r:node16
362
    9. instance44  node20:node2  => node2:node15  0.00438740 a=f r:node15
363
   10. instance62  node14:node18 => node14:node16 0.00390087 a=r:node16
364
   11. instance13  node11:node14 => node11:node16 0.00361787 a=r:node16
365
   12. instance19  node10:node11 => node10:node7  0.00336636 a=r:node7
366
   13. instance43  node12:node13 => node12:node1  0.00305681 a=r:node1
367
   14. instance1   node1:node2   => node1:node4   0.00263124 a=r:node4
368
   15. instance58  node19:node20 => node19:node17 0.00252594 a=r:node17
369
Cluster score improved from 0.52329131 to 0.00252594
370
.fi
371
.in
372

    
373
In the above output, we can see:
374
  - the input data (here from files) shows a cluster with 20 nodes and
375
    80 instances
376
  - the cluster is not initially N+1 compliant
377
  - the initial score is 0.52329131
378

    
379
The step list follows, showing the instance, its initial
380
primary/secondary nodes, the new primary secondary, the cluster list,
381
and the actions taken in this step (with 'f' denoting failover/migrate
382
and 'r' denoting replace secondary).
383

    
384
Finally, the program shows the improvement in cluster score.
385

    
386
A more detailed output is obtained via the \fB-C\fR and \fB-p\fR options:
387

    
388
.in +4n
389
.nf
390
.RB "$" " hbal"
391
Loaded 20 nodes, 80 instances
392
Cluster is not N+1 happy, continuing but no guarantee that the cluster will end N+1 happy.
393
Initial cluster status:
394
N1 Name   t_mem f_mem r_mem t_dsk f_dsk pri sec  p_fmem  p_fdsk
395
 * node1  32762  1280  6000  1861  1026   5   3 0.03907 0.55179
396
   node2  32762 31280 12000  1861  1026   0   8 0.95476 0.55179
397
 * node3  32762  1280  6000  1861  1026   5   3 0.03907 0.55179
398
 * node4  32762  1280  6000  1861  1026   5   3 0.03907 0.55179
399
 * node5  32762  1280  6000  1861   978   5   5 0.03907 0.52573
400
 * node6  32762  1280  6000  1861  1026   5   3 0.03907 0.55179
401
 * node7  32762  1280  6000  1861  1026   5   3 0.03907 0.55179
402
   node8  32762  7280  6000  1861  1026   4   4 0.22221 0.55179
403
   node9  32762  7280  6000  1861  1026   4   4 0.22221 0.55179
404
 * node10 32762  7280 12000  1861  1026   4   4 0.22221 0.55179
405
   node11 32762  7280  6000  1861   922   4   5 0.22221 0.49577
406
   node12 32762  7280  6000  1861  1026   4   4 0.22221 0.55179
407
   node13 32762  7280  6000  1861   922   4   5 0.22221 0.49577
408
   node14 32762  7280  6000  1861   922   4   5 0.22221 0.49577
409
 * node15 32762  7280 12000  1861  1131   4   3 0.22221 0.60782
410
   node16 32762 31280     0  1861  1860   0   0 0.95476 1.00000
411
   node17 32762  7280  6000  1861  1106   5   3 0.22221 0.59479
412
 * node18 32762  1280  6000  1396   561   5   3 0.03907 0.40239
413
 * node19 32762  1280  6000  1861  1026   5   3 0.03907 0.55179
414
   node20 32762 13280 12000  1861   689   3   9 0.40535 0.37068
415

    
416
Initial score: 0.52329131
417
Trying to minimize the CV...
418
    1. instance14  node1:node10  => node16:node10 0.42109120 a=f r:node16 f
419
    2. instance54  node4:node15  => node16:node15 0.31904594 a=f r:node16 f
420
    3. instance4   node5:node2   => node2:node16  0.26611015 a=f r:node16
421
    4. instance48  node18:node20 => node2:node18  0.21361717 a=r:node2 f
422
    5. instance93  node19:node18 => node16:node19 0.16166425 a=r:node16 f
423
    6. instance89  node3:node20  => node2:node3   0.11005629 a=r:node2 f
424
    7. instance5   node6:node2   => node16:node6  0.05841589 a=r:node16 f
425
    8. instance94  node7:node20  => node20:node16 0.00658759 a=f r:node16
426
    9. instance44  node20:node2  => node2:node15  0.00438740 a=f r:node15
427
   10. instance62  node14:node18 => node14:node16 0.00390087 a=r:node16
428
   11. instance13  node11:node14 => node11:node16 0.00361787 a=r:node16
429
   12. instance19  node10:node11 => node10:node7  0.00336636 a=r:node7
430
   13. instance43  node12:node13 => node12:node1  0.00305681 a=r:node1
431
   14. instance1   node1:node2   => node1:node4   0.00263124 a=r:node4
432
   15. instance58  node19:node20 => node19:node17 0.00252594 a=r:node17
433
Cluster score improved from 0.52329131 to 0.00252594
434

    
435
Commands to run to reach the above solution:
436
  echo step 1
437
  echo gnt-instance migrate instance14
438
  echo gnt-instance replace-disks -n node16 instance14
439
  echo gnt-instance migrate instance14
440
  echo step 2
441
  echo gnt-instance migrate instance54
442
  echo gnt-instance replace-disks -n node16 instance54
443
  echo gnt-instance migrate instance54
444
  echo step 3
445
  echo gnt-instance migrate instance4
446
  echo gnt-instance replace-disks -n node16 instance4
447
  echo step 4
448
  echo gnt-instance replace-disks -n node2 instance48
449
  echo gnt-instance migrate instance48
450
  echo step 5
451
  echo gnt-instance replace-disks -n node16 instance93
452
  echo gnt-instance migrate instance93
453
  echo step 6
454
  echo gnt-instance replace-disks -n node2 instance89
455
  echo gnt-instance migrate instance89
456
  echo step 7
457
  echo gnt-instance replace-disks -n node16 instance5
458
  echo gnt-instance migrate instance5
459
  echo step 8
460
  echo gnt-instance migrate instance94
461
  echo gnt-instance replace-disks -n node16 instance94
462
  echo step 9
463
  echo gnt-instance migrate instance44
464
  echo gnt-instance replace-disks -n node15 instance44
465
  echo step 10
466
  echo gnt-instance replace-disks -n node16 instance62
467
  echo step 11
468
  echo gnt-instance replace-disks -n node16 instance13
469
  echo step 12
470
  echo gnt-instance replace-disks -n node7 instance19
471
  echo step 13
472
  echo gnt-instance replace-disks -n node1 instance43
473
  echo step 14
474
  echo gnt-instance replace-disks -n node4 instance1
475
  echo step 15
476
  echo gnt-instance replace-disks -n node17 instance58
477

    
478
Final cluster status:
479
N1 Name   t_mem f_mem r_mem t_dsk f_dsk pri sec  p_fmem  p_fdsk
480
   node1  32762  7280  6000  1861  1026   4   4 0.22221 0.55179
481
   node2  32762  7280  6000  1861  1026   4   4 0.22221 0.55179
482
   node3  32762  7280  6000  1861  1026   4   4 0.22221 0.55179
483
   node4  32762  7280  6000  1861  1026   4   4 0.22221 0.55179
484
   node5  32762  7280  6000  1861  1078   4   5 0.22221 0.57947
485
   node6  32762  7280  6000  1861  1026   4   4 0.22221 0.55179
486
   node7  32762  7280  6000  1861  1026   4   4 0.22221 0.55179
487
   node8  32762  7280  6000  1861  1026   4   4 0.22221 0.55179
488
   node9  32762  7280  6000  1861  1026   4   4 0.22221 0.55179
489
   node10 32762  7280  6000  1861  1026   4   4 0.22221 0.55179
490
   node11 32762  7280  6000  1861  1022   4   4 0.22221 0.54951
491
   node12 32762  7280  6000  1861  1026   4   4 0.22221 0.55179
492
   node13 32762  7280  6000  1861  1022   4   4 0.22221 0.54951
493
   node14 32762  7280  6000  1861  1022   4   4 0.22221 0.54951
494
   node15 32762  7280  6000  1861  1031   4   4 0.22221 0.55408
495
   node16 32762  7280  6000  1861  1060   4   4 0.22221 0.57007
496
   node17 32762  7280  6000  1861  1006   5   4 0.22221 0.54105
497
   node18 32762  7280  6000  1396   761   4   2 0.22221 0.54570
498
   node19 32762  7280  6000  1861  1026   4   4 0.22221 0.55179
499
   node20 32762 13280  6000  1861  1089   3   5 0.40535 0.58565
500

    
501
.fi
502
.in
503

    
504
Here we see, beside the step list, the initial and final cluster
505
status, with the final one showing all nodes being N+1 compliant, and
506
the command list to reach the final solution. In the initial listing,
507
we see which nodes are not N+1 compliant.
508

    
509
The algorithm is stable as long as each step above is fully completed,
510
e.g. in step 8, both the migrate and the replace-disks are
511
done. Otherwise, if only the migrate is done, the input data is
512
changed in a way that the program will output a different solution
513
list (but hopefully will end in the same state).
514

    
515
.SH SEE ALSO
516
.BR hn1 "(1), " hscan "(1), " ganeti "(7), " gnt-instance "(8), "
517
.BR gnt-node "(8)"