Confusing MFAKTc output

swl551 · 2013-11-30, 00:31

Hey,

My worktoDo.txt show I am working on 68969669,71,73

But MFAKTc suddenly is working on a totally different set of exponents as seen in the captured output below.

Code:

no factor for M89589361 from 2^67 to 2^68 [mfaktc 0.20 barrett76_mul32_gs]
tf(): total time spent:  2m 15.126s

got assignment: exp=89589707 bit_min=67 bit_max=68 (0.33 GHz-days)
Starting trial factoring M89589707 from 2^67 to 2^68 (0.33 GHz-days)

These exponents are not even assigned to me (and not actually in my workToDo.txt file.

The last checkpoint file in my directory is M68969669.ckp, but is hours old and Results.txt is not being written to. It as if the program in in some diagnostic loop???

I killed MFAKTc and restarted and it resumed working on M68969669.

ANYONE have a clue on what is going on here?

chalsall · 2013-11-30, 01:03

Quote:

Originally Posted by swl551

ANYONE have a clue on what is going on here?

I can only speak to what I can observe.

68969669 was assigned to you (and only to you) by GPU72 to TF from 71 to 73.

What, exactly, is the trouble?

swl551 · 2013-11-30, 01:08

Quote:

Originally Posted by chalsall

I can only speak to what I can observe.

68969669 was assigned to you (and only to you) by GPU72 to TF from 71 to 73.

What, exactly, is the trouble?

Hey,
The exact trouble is mfaktc began processing exponents not in my workTodo
M89589361
M89589707

failed to write anything into results.txt for hours as if it were in some diagnostic mode.

I am clueless why this happened.

chalsall · 2013-11-30, 01:18

Quote:

Originally Posted by swl551

I am clueless why this happened.

As am I.

Neither M89589361 nor M89589707 ever passed through the GPU72 system according to my records.

To the best of my understanding, neither of these were ever assigned by GPU72.

There's a bug somewhere. Happy to accept it might be on my end, but I don't think so.

LaurV · 2013-11-30, 04:19

Looking for those exponents in the 0.20 sources, turns out nothing. There is no (easy to spot) hidden trick that would cause crunching on them. Cosmic rays? Maybe god is talking to you... you should LL those exponents, one of them might turn out a mersenne prime

(assuming the expos are prime, I did not check).

swl551 · 2013-11-30, 14:19

Like I said. mfaktc ran these components for hours without updating results.txt. I was notified of the situation by MISFIT's stall detection. No ckp files written and no results updated....

2013-11-30, 00:31	#1
swl551 Aug 2012 New Hampshire 2³·101 Posts	Confusing MFAKTc output Hey, My worktoDo.txt show I am working on 68969669,71,73 But MFAKTc suddenly is working on a totally different set of exponents as seen in the captured output below. Code: no factor for M89589361 from 2^67 to 2^68 [mfaktc 0.20 barrett76_mul32_gs] tf(): total time spent: 2m 15.126s got assignment: exp=89589707 bit_min=67 bit_max=68 (0.33 GHz-days) Starting trial factoring M89589707 from 2^67 to 2^68 (0.33 GHz-days) These exponents are not even assigned to me (and not actually in my workToDo.txt file. The last checkpoint file in my directory is M68969669.ckp, but is hours old and Results.txt is not being written to. It as if the program in in some diagnostic loop??? I killed MFAKTc and restarted and it resumed working on M68969669. ANYONE have a clue on what is going on here? Last fiddled with by swl551 on 2013-11-30 at 01:00

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Intel Core 8000 series is a confusing mess	VictordeHolland	Hardware	1	2018-03-05 17:37
Reducing logging output of mfaktc	daxmick	GPU Computing	3	2018-01-18 01:16
is confusing	zakary	Miscellaneous Math	29	2014-11-17 08:31
confusing server error messages	ramgeis	PrimeNet	2	2013-06-09 23:53
Confusing terms/acronyms?	Uncwilly	Lounge	0	2010-12-29 01:27

2013-11-30, 04:19	#5
LaurV Romulan Interpreter Jun 2011 Thailand 9161₁₀ Posts	Looking for those exponents in the 0.20 sources, turns out nothing. There is no (easy to spot) hidden trick that would cause crunching on them. Cosmic rays? Maybe god is talking to you... you should LL those exponents, one of them might turn out a mersenne prime (assuming the expos are prime, I did not check).

2013-11-30, 14:19	#6
swl551 Aug 2012 New Hampshire 2³·101 Posts	Like I said. mfaktc ran these components for hours without updating results.txt. I was notified of the situation by MISFIT's stall detection. No ckp files written and no results updated....