1 .TH HBAL 1 2009-03-23 htools "Ganeti H-tools"
3 hbal \- Cluster balancer for Ganeti
14 .BI "[-m " cluster "]"
15 .BI "[-n " nodes-file " ]"
16 .BI "[-i " instances-file "]"
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.
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.
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.
39 The program works in independent steps; at each step, we compute the
40 best instance move that lowers the cluster score.
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:
55 replace primary, a composite move (f, r, f)
58 failover and replace secondary, also composite (f, r)
61 replace secondary and failover, also composite (r, f)
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.
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
78 coefficient of variance of the percent of free memory
81 coefficient of variance of the percent of reserved memory
84 coefficient of variance of the percent of free disk
87 percentage of nodes failing N+1 check
90 percentage of instances living (either as primary or secondary) on
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.
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
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.
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
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 :)
125 .SS OFFLINE INSTANCES
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.
133 .SS OTHER POSSIBLE METRICS
135 It would be desirable to add more metrics to the algorithm, especially
136 dynamically-computed metrics, such as:
140 CPU usage of instances, combined with VCPU versus PCPU count
150 The options that can be passed to the program are as follows:
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.
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.
160 The node list will contain these informations:
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
171 the total node memory
174 the memory used by the node itself
177 the memory used by instances
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
188 the reserved node memory, which is the amount of free memory needed
198 number of primary instances
201 number of secondary instances
204 percent of free memory
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
216 The line will contain four fields:
221 initial cluster score
224 number of steps in the solution
230 improvement in the cluster score
236 This option (which can be given multiple times) will mark nodes as
237 being \fIoffline\fR. This means a couple of things:
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.
247 these nodes will not be included in the score calculation (except for
248 the percentage of instances on offline nodes)
250 Note that hbal will also mark as offline any nodes which are reported
251 by RAPI as such, or that have "?" in file-based input in any numeric
256 .BI "-e" score ", --min-score=" score
257 This parameter denotes the minimum score we are happy with and alters
258 the computation in two ways:
263 if the cluster has the initial score lower than this value, then we
264 don't enter the algorithm at all, and exit with success
267 during the iterative process, if we reach a score lower than this
268 value, we exit the algorithm
270 The default value of the parameter is currently \fI1e-9\fR (chosen
275 .BI "-n" nodefile ", --nodes=" nodefile
276 The name of the file holding node information (if not collecting via
277 RAPI), instead of the default \fInodes\fR file (but see below how to
278 customize the default value via the environment).
281 .BI "-i" instancefile ", --instances=" instancefile
282 The name of the file holding instance information (if not collecting
283 via RAPI), instead of the default \fIinstances\fR file (but see below
284 how to customize the default value via the environment).
288 Collect data not from files but directly from the
290 given as an argument via RAPI. This work for both Ganeti 1.2 and
294 .BI "-l" N ", --max-length=" N
295 Restrict the solution to this length. This can be used for example to
296 automate the execution of the balancing.
300 Increase the output verbosity. Each usage of this option will increase
301 the verbosity (currently more than 2 doesn't make sense) from the
306 Decrease the output verbosity. Each usage of this option will decrease
307 the verbosity (less than zero doesn't make sense) from the default of
312 Just show the program version and exit.
316 The exist status of the command will be zero, unless for some reason
317 the algorithm fatally failed (e.g. wrong node or instance data).
321 If the variables \fBHTOOLS_NODES\fR and \fBHTOOLS_INSTANCES\fR are
322 present in the environment, they will override the default names for
323 the nodes and instances files. These will have of course no effect
328 The program does not check its input data for consistency, and aborts
329 with cryptic errors messages in this case.
331 The algorithm is not perfect.
333 The algorithm doesn't deal with non-\fBdrbd\fR instances, and chokes
334 on input data which has such instances.
336 The output format is not easily scriptable, and the program should
337 feed moves directly into Ganeti (either via RAPI or via a gnt-debug
342 Note that this example are not for the latest version (they don't have
347 With the default options, the program shows each individual step and
348 the improvements it brings in cluster score:
353 Loaded 20 nodes, 80 instances
354 Cluster is not N+1 happy, continuing but no guarantee that the cluster will end N+1 happy.
355 Initial score: 0.52329131
356 Trying to minimize the CV...
357 1. instance14 node1:node10 => node16:node10 0.42109120 a=f r:node16 f
358 2. instance54 node4:node15 => node16:node15 0.31904594 a=f r:node16 f
359 3. instance4 node5:node2 => node2:node16 0.26611015 a=f r:node16
360 4. instance48 node18:node20 => node2:node18 0.21361717 a=r:node2 f
361 5. instance93 node19:node18 => node16:node19 0.16166425 a=r:node16 f
362 6. instance89 node3:node20 => node2:node3 0.11005629 a=r:node2 f
363 7. instance5 node6:node2 => node16:node6 0.05841589 a=r:node16 f
364 8. instance94 node7:node20 => node20:node16 0.00658759 a=f r:node16
365 9. instance44 node20:node2 => node2:node15 0.00438740 a=f r:node15
366 10. instance62 node14:node18 => node14:node16 0.00390087 a=r:node16
367 11. instance13 node11:node14 => node11:node16 0.00361787 a=r:node16
368 12. instance19 node10:node11 => node10:node7 0.00336636 a=r:node7
369 13. instance43 node12:node13 => node12:node1 0.00305681 a=r:node1
370 14. instance1 node1:node2 => node1:node4 0.00263124 a=r:node4
371 15. instance58 node19:node20 => node19:node17 0.00252594 a=r:node17
372 Cluster score improved from 0.52329131 to 0.00252594
376 In the above output, we can see:
377 - the input data (here from files) shows a cluster with 20 nodes and
379 - the cluster is not initially N+1 compliant
380 - the initial score is 0.52329131
382 The step list follows, showing the instance, its initial
383 primary/secondary nodes, the new primary secondary, the cluster list,
384 and the actions taken in this step (with 'f' denoting failover/migrate
385 and 'r' denoting replace secondary).
387 Finally, the program shows the improvement in cluster score.
389 A more detailed output is obtained via the \fB-C\fR and \fB-p\fR options:
394 Loaded 20 nodes, 80 instances
395 Cluster is not N+1 happy, continuing but no guarantee that the cluster will end N+1 happy.
396 Initial cluster status:
397 N1 Name t_mem f_mem r_mem t_dsk f_dsk pri sec p_fmem p_fdsk
398 * node1 32762 1280 6000 1861 1026 5 3 0.03907 0.55179
399 node2 32762 31280 12000 1861 1026 0 8 0.95476 0.55179
400 * node3 32762 1280 6000 1861 1026 5 3 0.03907 0.55179
401 * node4 32762 1280 6000 1861 1026 5 3 0.03907 0.55179
402 * node5 32762 1280 6000 1861 978 5 5 0.03907 0.52573
403 * node6 32762 1280 6000 1861 1026 5 3 0.03907 0.55179
404 * node7 32762 1280 6000 1861 1026 5 3 0.03907 0.55179
405 node8 32762 7280 6000 1861 1026 4 4 0.22221 0.55179
406 node9 32762 7280 6000 1861 1026 4 4 0.22221 0.55179
407 * node10 32762 7280 12000 1861 1026 4 4 0.22221 0.55179
408 node11 32762 7280 6000 1861 922 4 5 0.22221 0.49577
409 node12 32762 7280 6000 1861 1026 4 4 0.22221 0.55179
410 node13 32762 7280 6000 1861 922 4 5 0.22221 0.49577
411 node14 32762 7280 6000 1861 922 4 5 0.22221 0.49577
412 * node15 32762 7280 12000 1861 1131 4 3 0.22221 0.60782
413 node16 32762 31280 0 1861 1860 0 0 0.95476 1.00000
414 node17 32762 7280 6000 1861 1106 5 3 0.22221 0.59479
415 * node18 32762 1280 6000 1396 561 5 3 0.03907 0.40239
416 * node19 32762 1280 6000 1861 1026 5 3 0.03907 0.55179
417 node20 32762 13280 12000 1861 689 3 9 0.40535 0.37068
419 Initial score: 0.52329131
420 Trying to minimize the CV...
421 1. instance14 node1:node10 => node16:node10 0.42109120 a=f r:node16 f
422 2. instance54 node4:node15 => node16:node15 0.31904594 a=f r:node16 f
423 3. instance4 node5:node2 => node2:node16 0.26611015 a=f r:node16
424 4. instance48 node18:node20 => node2:node18 0.21361717 a=r:node2 f
425 5. instance93 node19:node18 => node16:node19 0.16166425 a=r:node16 f
426 6. instance89 node3:node20 => node2:node3 0.11005629 a=r:node2 f
427 7. instance5 node6:node2 => node16:node6 0.05841589 a=r:node16 f
428 8. instance94 node7:node20 => node20:node16 0.00658759 a=f r:node16
429 9. instance44 node20:node2 => node2:node15 0.00438740 a=f r:node15
430 10. instance62 node14:node18 => node14:node16 0.00390087 a=r:node16
431 11. instance13 node11:node14 => node11:node16 0.00361787 a=r:node16
432 12. instance19 node10:node11 => node10:node7 0.00336636 a=r:node7
433 13. instance43 node12:node13 => node12:node1 0.00305681 a=r:node1
434 14. instance1 node1:node2 => node1:node4 0.00263124 a=r:node4
435 15. instance58 node19:node20 => node19:node17 0.00252594 a=r:node17
436 Cluster score improved from 0.52329131 to 0.00252594
438 Commands to run to reach the above solution:
440 echo gnt-instance migrate instance14
441 echo gnt-instance replace-disks -n node16 instance14
442 echo gnt-instance migrate instance14
444 echo gnt-instance migrate instance54
445 echo gnt-instance replace-disks -n node16 instance54
446 echo gnt-instance migrate instance54
448 echo gnt-instance migrate instance4
449 echo gnt-instance replace-disks -n node16 instance4
451 echo gnt-instance replace-disks -n node2 instance48
452 echo gnt-instance migrate instance48
454 echo gnt-instance replace-disks -n node16 instance93
455 echo gnt-instance migrate instance93
457 echo gnt-instance replace-disks -n node2 instance89
458 echo gnt-instance migrate instance89
460 echo gnt-instance replace-disks -n node16 instance5
461 echo gnt-instance migrate instance5
463 echo gnt-instance migrate instance94
464 echo gnt-instance replace-disks -n node16 instance94
466 echo gnt-instance migrate instance44
467 echo gnt-instance replace-disks -n node15 instance44
469 echo gnt-instance replace-disks -n node16 instance62
471 echo gnt-instance replace-disks -n node16 instance13
473 echo gnt-instance replace-disks -n node7 instance19
475 echo gnt-instance replace-disks -n node1 instance43
477 echo gnt-instance replace-disks -n node4 instance1
479 echo gnt-instance replace-disks -n node17 instance58
481 Final cluster status:
482 N1 Name t_mem f_mem r_mem t_dsk f_dsk pri sec p_fmem p_fdsk
483 node1 32762 7280 6000 1861 1026 4 4 0.22221 0.55179
484 node2 32762 7280 6000 1861 1026 4 4 0.22221 0.55179
485 node3 32762 7280 6000 1861 1026 4 4 0.22221 0.55179
486 node4 32762 7280 6000 1861 1026 4 4 0.22221 0.55179
487 node5 32762 7280 6000 1861 1078 4 5 0.22221 0.57947
488 node6 32762 7280 6000 1861 1026 4 4 0.22221 0.55179
489 node7 32762 7280 6000 1861 1026 4 4 0.22221 0.55179
490 node8 32762 7280 6000 1861 1026 4 4 0.22221 0.55179
491 node9 32762 7280 6000 1861 1026 4 4 0.22221 0.55179
492 node10 32762 7280 6000 1861 1026 4 4 0.22221 0.55179
493 node11 32762 7280 6000 1861 1022 4 4 0.22221 0.54951
494 node12 32762 7280 6000 1861 1026 4 4 0.22221 0.55179
495 node13 32762 7280 6000 1861 1022 4 4 0.22221 0.54951
496 node14 32762 7280 6000 1861 1022 4 4 0.22221 0.54951
497 node15 32762 7280 6000 1861 1031 4 4 0.22221 0.55408
498 node16 32762 7280 6000 1861 1060 4 4 0.22221 0.57007
499 node17 32762 7280 6000 1861 1006 5 4 0.22221 0.54105
500 node18 32762 7280 6000 1396 761 4 2 0.22221 0.54570
501 node19 32762 7280 6000 1861 1026 4 4 0.22221 0.55179
502 node20 32762 13280 6000 1861 1089 3 5 0.40535 0.58565
507 Here we see, beside the step list, the initial and final cluster
508 status, with the final one showing all nodes being N+1 compliant, and
509 the command list to reach the final solution. In the initial listing,
510 we see which nodes are not N+1 compliant.
512 The algorithm is stable as long as each step above is fully completed,
513 e.g. in step 8, both the migrate and the replace-disks are
514 done. Otherwise, if only the migrate is done, the input data is
515 changed in a way that the program will output a different solution
516 list (but hopefully will end in the same state).
519 .BR hn1 "(1), " hscan "(1), " ganeti "(7), " gnt-instance "(8), "