[Cbc] What is trap 6 error (i.e. cbc-mip: line 13: 6790 Abort trap: 6 ) in CBC 2.9.3 and 2.9.4?

Mon Apr 6 22:19:44 EDT 2015

Hi,

Thanks for looking after this issue.

I will you send copy of the glpk logs and 750 stands with adjacency
problem that also has a problem. See the logs below.

Noli

~~~~~~
Welcome to the CBC MILP Solver
Version: 2.9.4
Build Date: Apr  5 2015
Revision Number: 2179

command line - /usr/local/bin/cbc
MIP_750_stands_data_750_case1_ver1.mod% -randomCbcSeed 0 -randomSeed 0
-Rens on -printi csv -solve -solu solution_new_cbc_latest_rev6.csv
(default strategy 1)
GMPL model file ./MIP_750_stands_data_750_case1_ver1.mod and data file ./
Reading model section from ./MIP_750_stands_data_750_case1_ver1.mod...
Reading data section from ./MIP_750_stands_data_750_case1_ver1.mod...
6677 lines were read
Generating Totalprofit...
Generating Discounted_Net_Value...
Generating Stand_Cut...
Generating Harvest_Volume1...
Generating Area...
Generating Evenflow_Harvest_Volume_HV_Alpha...
Generating Evenflow_Harvest_Volume_HV_Beta...
Generating Adjacent_Constraint...
Model has been successfully generated

.
.
Cbc0038I Pass  29: (27.70 seconds) suminf.    0.35838 (2) obj.
-1.69562e+07 iterations 3
Cbc0038I Pass  30: (27.93 seconds) suminf.    1.27016 (6) obj.
-1.67653e+07 iterations 358
Cbc0038I No solution found this major pass
Cbc0038I Before mini branch and bound, 6432 integers at bound fixed
and 0 continuous
Cbc0038I Full problem 12000 rows 7070 columns, reduced to 555 rows 636 columns
Cbc0038I Mini branch and bound did not improve solution (30.74 seconds)
Cbc0038I Full problem 12001 rows 7070 columns, reduced to 12001 rows
7070 columns - too large
Cbc0038I After 38.60 seconds - Feasibility pump exiting - took 15.09 seconds
Cbc0038I Full problem 12000 rows 7070 columns, reduced to 330 rows 802 columns
Cbc0038I Full problem 12000 rows 7070 columns, reduced to 328 rows 235 columns
Assertion failed: (nErrors), function resetFakeBounds, file
ClpSimplexDual.cpp, line 7684.
Cbc0031I 28 added rows had average density of 6.2142857
Cbc0013I At root node, 28 cuts changed objective from -19247302 to
-19245737 in 3 passes
Cbc0014I Cut generator 0 (Probing) - 0 row cuts average 0.0 elements,
0 column cuts (0 active)  in 0.382 seconds - new frequency is -100
Cbc0014I Cut generator 1 (Gomory) - 82 row cuts average 12.3 elements,
0 column cuts (0 active)  in 0.354 seconds - new frequency is 1
Cbc0014I Cut generator 2 (Knapsack) - 0 row cuts average 0.0 elements,
0 column cuts (0 active)  in 0.202 seconds - new frequency is -100
Cbc0014I Cut generator 3 (Clique) - 34 row cuts average 3.0 elements,
0 column cuts (0 active)  in 0.021 seconds - new frequency is 1
Cbc0014I Cut generator 4 (MixedIntegerRounding2) - 0 row cuts average
0.0 elements, 0 column cuts (0 active)  in 0.118 seconds - new
frequency is -100
Cbc0014I Cut generator 5 (FlowCover) - 0 row cuts average 0.0
elements, 0 column cuts (0 active)  in 0.021 seconds - new frequency
is -100
Cbc0014I Cut generator 6 (TwoMirCuts) - 74 row cuts average 13.1
elements, 0 column cuts (0 active)  in 0.451 seconds - new frequency
is -100
Cbc0010I After 0 nodes, 1 on tree, 1e+50 best solution, best possible
-19245737 (44.36 seconds)
Cbc0010I After 100 nodes, 60 on tree, 1e+50 best solution, best
possible -19245737 (103.30 seconds)
Cbc0010I After 200 nodes, 107 on tree, 1e+50 best solution, best
possible -19245737 (140.84 seconds)
Cbc0010I After 300 nodes, 158 on tree, 1e+50 best solution, best
possible -19245737 (170.18 seconds)
Cbc0010I After 400 nodes, 206 on tree, 1e+50 best solution, best
possible -19245737 (195.64 seconds)
Cbc0010I After 500 nodes, 259 on tree, 1e+50 best solution, best
possible -19245737 (215.17 seconds)
Cbc0010I After 600 nodes, 305 on tree, 1e+50 best solution, best
possible -19245737 (234.03 seconds)
Cbc0010I After 700 nodes, 354 on tree, 1e+50 best solution, best
possible -19245737 (252.83 seconds)
Cbc0010I After 800 nodes, 405 on tree, 1e+50 best solution, best
possible -19245737 (279.62 seconds)
Cbc0010I After 900 nodes, 457 on tree, 1e+50 best solution, best
possible -19245737 (299.45 seconds)
Cbc0010I After 1000 nodes, 505 on tree, 1e+50 best solution, best
possible -19245737 (317.50 seconds)
Cbc0010I After 1100 nodes, 604 on tree, 1e+50 best solution, best
possible -19245733 (352.79 seconds)
Cbc0010I After 1200 nodes, 704 on tree, 1e+50 best solution, best
possible -19245733 (375.12 seconds)
Cbc0010I After 1300 nodes, 804 on tree, 1e+50 best solution, best
possible -19245733 (395.19 seconds)
Cbc0012I Integer solution of -19111622 found by rounding after 25312
iterations and 1385 nodes (413.16 seconds)
Cbc0038I Full problem 12000 rows 7070 columns, reduced to 69 rows 51 columns
/Users/nsicad/.config/gedit/tools/cbc-mip: line 13: 95633 Abort trap:
6           /usr/local/bin/cbc $GEDIT_CURRENT_DOCUMENT_NAME%
-randomCbcSeed 0 -randomSeed 0 -Rens on -printi csv -solve -solu
solution_new_cbc_latest_rev6.csv

real	22m48.782s
user	6m55.120s
sys	0m8.372s

Exited: 34304

On 4/7/15, John Forrest <john.forrest at fastercoin.com> wrote:
> Noli,
>
> Took some time to get fault to occur - but eventually got it to happen.
> In the end not very important (and is OK with NDEBUG set), but I have
> modified code in svn.
>
> With -naive on, code gets solution of 0.0 immediately and RINS heuristic
> gets reasonable solution after a few minutes.  My test is still running
> with best solution of -78735.4.
>
> What were glpk settings that solved it completely?
>
> John Forrest
> On 06/04/15 03:31, Noli Sicad wrote:
>> Hi John,
>>
>> Here's (below) is the log of CBC.
>>
>> The model has 1000 polygons / forest stands to be harvested. This is
>> the base case, no adjacency problem.
>>
>> GLPK 4.55 solved this problem using proximity search.
>>
>> However, CBC solver 2.9.4 could not find a solution (see error below).
>>
>> No threads this time.
>>
>> I can send you the mps format of this model, if you like
>>
>> Thanks, Noli