[Scip] Error <-1> and <-8>

Timo Berthold berthold at zib.de
Mon Feb 17 08:28:46 CET 2014 

Dear Tanya,

all different types of error codes are listed here:
http://scip.zib.de/doc/html/type__retcode_8h.shtml

In the described case, the memory error occured in a sub-SCIP of a
heuristic, made it terminate in an invalid stage; therefore the heuristic
returned an invalid call.

An out-of-memory in general means that the machine cannot allocate any
more memory. Bigger RAM sticks would help as a fix ;-)
Seriously, options are
(a) use a machine with considerably more RAM
(b) check whether there is a hard memory limit set outside of SCIP (dunno
how neos handles this)
(c) you could try to use DFS as node selection rule. It saves memory, but
typically has a bad convegrence behavior w.r.t. primal and dual bounds
(d) find out from the SCIP statistics (SCIPprintStatistics()) whether
there are any features which work particularly well and make them run more
frequently in a subsequent call
(e) use a different model ;-)

>From the lgos that you send us, I would hold a bet that the model does not
solve  with SCIP's default settings within a day. But you never know, of
course. So or so, if (e) is an option, consider it. ;-)

Cheers
Timo
> There is a log file generated that is available at the bottom of the link
> in the OP.
>
> The last lines before the errors start from the log file are:
>
>  time | node  | left  |LP iter|LP it/n| mem |mdpt |frac |extbr|vars |cons
> |rows |cuts |confs|strbr|  dualbound   | primalbound  |  gap
>  1651s|614100 |260527 |  9457k|  15.4 |2445M| 115 |   0 |  10 | 344 |3395
> | 257 |2243k| 101k|  15k| 1.485300e+04 | 1.770800e+04 |  19.22%
>  1652s|614200 |260573 |  9459k|  15.4 |2445M| 115 |   0 |   5 | 344 |3390
> | 260 |2243k| 101k|  15k| 1.485300e+04 | 1.770800e+04 |  19.22%
>  1652s|614300 |260617 |  9461k|  15.4 |2446M| 115 |   0 |   0 | 344 |3385
> | 264 |2243k| 101k|  15k| 1.485306e+04 | 1.770800e+04 |  19.22%
>  1652s|614400 |260659 |  9462k|  15.4 |2446M| 115 |   0 |   0 | 344 |3380
> | 265 |2244k| 101k|  15k| 1.485330e+04 | 1.770800e+04 |  19.22%
>  1653s|614500 |260707 |  9463k|  15.4 |2447M| 115 |   8 |   0 | 344 |3382
> | 291 |2244k| 101k|  15k| 1.485330e+04 | 1.770800e+04 |  19.22%
>  1653s|614600 |260757 |  9465k|  15.4 |2447M| 115 |   - |   0 | 344 |3371
> | 291 |2245k| 101k|  15k| 1.485330e+04 | 1.770800e+04 |  19.22%
>  1653s|614700 |260801 |  9466k|  15.4 |2447M| 115 |   - |   0 | 344 |3366
> |   0 |2245k| 101k|  15k| 1.485342e+04 | 1.770800e+04 |  19.22%
>  1653s|614800 |260843 |  9468k|  15.4 |2448M| 115 |   - |   0 | 344 |3358
> | 345 |2245k| 101k|  15k| 1.485361e+04 | 1.770800e+04 |  19.22%
>  1654s|614900 |260883 |  9470k|  15.4 |2448M| 115 |   - |   0 | 344 |3360
> | 282 |2246k| 101k|  15k| 1.485400e+04 | 1.770800e+04 |  19.21%
>  1654s|615000 |260922 |  9471k|  15.4 |2448M| 115 |   - |   0 | 344 |3386
> |   0 |2246k| 101k|  15k| 1.485400e+04 | 1.770800e+04 |  19.21%
>  1654s|615100 |260952 |  9472k|  15.4 |2449M| 115 |   0 |  20 | 344 |3394
> | 267 |2246k| 101k|  15k| 1.485400e+04 | 1.770800e+04 |  19.21%
>  1655s|615200 |261002 |  9474k|  15.4 |2449M| 115 |   - |   0 | 344 |3400
> |   0 |2247k| 101k|  15k| 1.485400e+04 | 1.770800e+04 |  19.21%
>  1655s|615300 |261052 |  9475k|  15.4 |2450M| 115 |   0 |   8 | 344 |3410
> | 271 |2247k| 101k|  15k| 1.485400e+04 | 1.770800e+04 |  19.21%
>  1655s|615400 |261088 |  9477k|  15.4 |2450M| 115 |   - |   0 | 344 |3411
> | 299 |2248k| 101k|  15k| 1.485400e+04 | 1.770800e+04 |  19.21%
>  1655s|615500 |261126 |  9478k|  15.4 |2452M| 115 |   3 |   0 | 344 |3408
> | 308 |2248k| 101k|  15k| 1.485400e+04 | 1.770800e+04 |  19.21%
>  time | node  | left  |LP iter|LP it/n| mem |mdpt |frac |extbr|vars |cons
> |rows |cuts |confs|strbr|  dualbound   | primalbound  |  gap
>  1656s|615600 |261172 |  9479k|  15.4 |2452M| 115 |   3 |   0 | 344 |3411
> | 307 |2248k| 101k|  15k| 1.485400e+04 | 1.770800e+04 |  19.21%
>  1656s|615700 |261220 |  9481k|  15.4 |2452M| 115 |   9 |   0 | 344 |3414
> | 292 |2249k| 101k|  15k| 1.485400e+04 | 1.770800e+04 |  19.21%
>  1656s|615800 |261259 |  9482k|  15.4 |2453M| 115 |   - |   0 | 344 |3424
> | 301 |2249k| 101k|  15k| 1.485400e+04 | 1.770800e+04 |  19.21%
>  1657s|615900 |261294 |  9484k|  15.4 |2453M| 115 |  14 |   0 | 344 |3420
> | 323 |2250k| 101k|  15k| 1.485403e+04 | 1.770800e+04 |  19.21%
>  1657s|616000 |261325 |  9486k|  15.4 |2453M| 115 |   - |   0 | 344 |3417
> |   0 |2250k| 101k|  15k| 1.485465e+04 | 1.770800e+04 |  19.21%
>
> Is there an option to increase the memory allotted or is the problem not
> solvable?
>
> Thanks for your help!
>
> Tanya
>
> On Feb 16, 2014, at 9:10 PM, Stefan Vigerske <stefan at math.hu-berlin.de>
> wrote:
>
>> Hi,
>>
>> does NEOS also makes you the logfile of the run available?
>>
>> cons_quadratic.c:1189 is a memory allocation, so probably the machine
>> run out of memory or a memory limit was hit.
>>
>> Stefan
>>
>> On 02/17/2014 02:32 AM, Tanya Kannon wrote:
>>> Hello,
>>>
>>> I am solving an MINLP through the NEOS server with a gams file uploaded
>>> for SCIP.  There are 4 files that are returning the following error
>>> messages:
>>> [../../../ThirdParty/SCIP/scip/src/scip/cons_quadratic.c:1189] ERROR:
>>> Error <-1> in function call
>>> [../../../ThirdParty/SCIP/scip/src/scip/cons_quadratic.c:9499] ERROR:
>>> Error <-1> in function call
>>> [../../../ThirdParty/SCIP/scip/src/scip/cons.c:5647] ERROR: Error <-1>
>>> in function call
>>> [../../../ThirdParty/SCIP/scip/src/scip/prob.c:546] ERROR: Error <-1>
>>> in function call
>>> [../../../ThirdParty/SCIP/scip/src/scip/scip.c:10971] ERROR: Error <-1>
>>> in function call
>>> [../../../ThirdParty/SCIP/scip/src/scip/heur_subnlp.c:968] ERROR: Error
>>> <-1> infunction call
>>> [../../../ThirdParty/SCIP/scip/src/scip/heur_subnlp.c:1631] ERROR:
>>> Error <-1> in function call
>>> [../../../ThirdParty/SCIP/scip/src/scip/heur_subnlp.c:2085] ERROR:
>>> Error <-1> in function call
>>> [../../../ThirdParty/SCIP/scip/src/scip/heur.c:453] ERROR: Error <-1>
>>> in function call
>>> [../../../ThirdParty/SCIP/scip/src/scip/solve.c:265] ERROR: Error <-1>
>>> in function call
>>> [../../../ThirdParty/SCIP/scip/src/scip/solve.c:4188] ERROR: Error <-1>
>>> in function call
>>> [../../../ThirdParty/SCIP/scip/src/scip/scip.c:12439] ERROR: Error <-1>
>>> in function call
>>> [../../../ThirdParty/SCIP/scip/src/scip/dialog_default.c:1407] ERROR:
>>> Error <-1> in function call
>>> [../../../ThirdParty/SCIP/scip/src/scip/dialog.c:828] ERROR: Error <-1>
>>> in function call
>>> [../../../ThirdParty/SCIP/scip/src/scip/dialog.c:370] ERROR: Error <-1>
>>> in function call
>>> [../../../ThirdParty/SCIP/scip/src/scip/scip.c:7583] ERROR: Error <-1>
>>> in function call
>>> [../../../ThirdParty/SCIP/scip/src/scip/scip.c:12616] ERROR: invalid
>>> SCIP stage<2>
>>> [../../../ThirdParty/SCIP/scip/src/scip/scip.c:8143] ERROR: Error <-8>
>>> in function call
>>> [../../../ThirdParty/SCIP/scip/src/scip/scip.c:721] ERROR: Error <-8>
>>> in function call
>>> [../../../ThirdParty/SCIP/scip/src/scip/heur_subnlp.c:370] ERROR: Error
>>> <-8> in function call
>>> [../../../ThirdParty/SCIP/scip/src/scip/heur_subnlp.c:1939] ERROR:
>>> Error <-8> in function call
>>> [../../../ThirdParty/SCIP/scip/src/scip/heur.c:330] ERROR: Error <-8>
>>> in function call
>>> [../../../ThirdParty/SCIP/scip/src/scip/set.c:3942] ERROR: Error <-8>
>>> in function call
>>> [../../../ThirdParty/SCIP/scip/src/scip/scip.c:11986] ERROR: Error <-8>
>>> in function call
>>> [../../../ThirdParty/SCIP/scip/src/scip/scip.c:12543] ERROR: Error <-8>
>>> in function call
>>> [../../../ThirdParty/SCIP/scip/src/scip/scip.c:12604] ERROR: Error <-8>
>>> in function call
>>> [../../../ThirdParty/SCIP/scip/src/scip/scip.c:8143] ERROR: Error <-8>
>>> in function call
>>> [../../../ThirdParty/SCIP/scip/src/scip/scip.c:721] ERROR: Error <-8>
>>> in function call
>>> The same network is being used to solve a second formulation that
>>> results in  more decision variables and constraints.
>>> I am trying to determine what is causing the error and how to resolve
>>> it. I have looked through the archives of the list serve and the
>>> Doxygen documentation and didn’t find anything about Error <-1> or
>>> Error <-8>.
>>> The entire results can be seen at
>>> http://www.neos-server.org/neos/jobs/2790000/2796145.html
>>> Thanks,
>>> Tanya
>>>
>>>
>>>
>>>
>>> _______________________________________________
>>> Scip mailing list
>>> Scip at zib.de
>>> http://listserv.zib.de/mailman/listinfo/scip
>>>
>>
>
>
> _______________________________________________
> Scip mailing list
> Scip at zib.de
> http://listserv.zib.de/mailman/listinfo/scip
>

More information about the Scip mailing list