• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

Load balancing /Kraken

Q

Quisarious

Limp Gawd
Joined
Apr 20, 2012
Messages
147
Not sure if this is a problem, or just the way it is...

I'm running Ubuntu 12.04, FAH 6.34 with the Kraken installed.

Whenever I stop and restart a WU load balancing turns on. If the client finishes a WU and starts a new one it does not. This results in a significant increase in TPF.

Is there a way to force dynamic load balancing?

Thanks!
 
Hmmmm.... I have not seen that one before. What command line options did you use when you ran the thekraken to install the wrapper?
 
if i recollect correctly..DLB is not gurranteed to start everytime a new WU kicks off
 
There is no way to automatically force DLB via Stanford's software. However, the latest version(s?) of theKraken takes advantage of the behavior you observed by automatically restarting a new WU shortly after a checkpoint (IF you enable the feature). You should download the latest version (0.6) and there are instructions in the README on how to enable it.
 
Last edited:
I have 0.6 with autorestart set.

I usually stop a WU and restart it anyway (need to reboot to windows briefly), but there have been 2 where I have not (both 8101's) and the kraken has not restarted the WU after the first checkpoint.

Not a big deal. I'm going to be switching this computer to SMP soon anyway so I have more flexibility with it.
 
Many of us have reported that our Thekraken reliably engages with with work units, but that DLB (Dynamic Load Balancing) sometimes fails to start.
 
The Kraken, when properly configured, should restart after first checkpoint.

Please provide thekraken.log from run you didn't observe autorestart so we can tell
more...

If The Kraken determines there is more than one incomplete units in work/
directory it will refuse to autorestart.
 
tear, as I reported previously (not disparaging you - you keep up with far more than I do!) a few weeks back:
- Top always, without exception, shows Thekraken as running.
- When a new work unit downloads and starts, Kraken always shows as running, but
- when a new work unit downloads and starts, Kraken shows in top, but in the workdata_0X.log file, "NOTE: Turning on dynamic load balancing" sometimes does not NOT appear. When it does not appear, indeed, "load imbalance is X.Y%" is high, around 7 to 9%. (About one out of every four new work unit starts, DLB does not show as turned on)
- when a work starts with "NOTE: Turning on dynamic load balancing," then DLB stays on throughout the entire execution of the work unit, including after checkpoints.
- If DLB does not engage when a work unit starts, it may or may not engage during the the rest of the work unit execution.

Maybe I still don't understand: aren't Thekraken and DLB separate things? Are they intertwined? (sorry, I don't know the correct terminology to phrase my questions)

The current work unit is folding with both Kraken and DLB engaged properly, so I don't think the current Kraken log would be of use, as there should be nothing out of order. I also looked in the "thekraken-prev.log," but I don't remember if the work unit(s) had DLB engaged or not. For what it's worth, I've enclosed log in code tags:

Code: 
thekraken: The Kraken 0.6 (compiled Fri Jan 20 19:38:48 AKST 2012 by all@greyhound5)
thekraken: Processor affinity wrapper for Folding@Home
thekraken: The Kraken comes with ABSOLUTELY NO WARRANTY; licensed under GPLv2
thekraken: PID: 2952
thekraken: launch binary: ./thekraken-FahCore_a5.exe
thekraken: config file: ./thekraken.cfg
thekraken: config: autorestart=1
thekraken: autorestart: examining work directory...
thekraken: autorestart is on
thekraken: autorestart slot: 7
thekraken: Forked 2953.
thekraken: autorestart: filesystem time offset: 0 seconds
thekraken: child: ptrace(PTRACE_TRACEME) returns 0
thekraken: child: Executing...
thekraken: waitpid() returns 2953 with status 0x0000057f
thekraken: 2953: stopped with signal 0x00000005
thekraken: 2953: initial attach
thekraken: 2953: Continuing.
thekraken: waitpid() returns 2953 with status 0x0003057f
thekraken: 2953: stopped with signal 0x00000005
thekraken: 2953: cloned 2954
thekraken: 2954: binding to cpu 0
thekraken: 2953: Continuing.
thekraken: waitpid() returns 2954 with status 0x0000137f
thekraken: 2954: stopped with signal 0x00000013
thekraken: 2954: Continuing.
thekraken: waitpid() returns 2955 with status 0x0000137f
thekraken: 2955: stopped with signal 0x00000013
thekraken: 2955: Continuing.
thekraken: waitpid() returns 2953 with status 0x0003057f
thekraken: 2953: stopped with signal 0x00000005
thekraken: 2953: cloned 2955
thekraken: 2953: Continuing.
thekraken: waitpid() returns 2956 with status 0x0000137f
thekraken: 2956: stopped with signal 0x00000013
thekraken: 2956: Continuing.
thekraken: waitpid() returns 2953 with status 0x0003057f
thekraken: 2953: stopped with signal 0x00000005
thekraken: 2953: cloned 2956
thekraken: 2953: Continuing.
thekraken: waitpid() returns 2957 with status 0x0000137f
thekraken: 2957: stopped with signal 0x00000013
thekraken: 2957: Continuing.
thekraken: waitpid() returns 2954 with status 0x0003057f
thekraken: 2954: stopped with signal 0x00000005
thekraken: 2954: cloned 2957
thekraken: 2957: binding to cpu 1
thekraken: 2954: Continuing.
thekraken: waitpid() returns 2958 with status 0x0000137f
thekraken: 2958: stopped with signal 0x00000013
thekraken: 2958: Continuing.
thekraken: waitpid() returns 2954 with status 0x0003057f
thekraken: 2954: stopped with signal 0x00000005
thekraken: 2954: cloned 2958
thekraken: 2958: binding to cpu 2
thekraken: 2954: Continuing.
thekraken: waitpid() returns 2959 with status 0x0000137f
thekraken: 2959: stopped with signal 0x00000013
thekraken: 2959: Continuing.
thekraken: waitpid() returns 2954 with status 0x0003057f
thekraken: 2954: stopped with signal 0x00000005
thekraken: 2954: cloned 2959
thekraken: 2959: binding to cpu 3
thekraken: 2954: Continuing.
thekraken: waitpid() returns 2960 with status 0x0000137f
thekraken: 2960: stopped with signal 0x00000013
thekraken: 2960: Continuing.
thekraken: waitpid() returns 2954 with status 0x0003057f
thekraken: 2954: stopped with signal 0x00000005
thekraken: 2954: cloned 2960
thekraken: 2960: binding to cpu 4
thekraken: 2954: Continuing.
thekraken: waitpid() returns 2961 with status 0x0000137f
thekraken: 2961: stopped with signal 0x00000013
thekraken: 2961: Continuing.
thekraken: waitpid() returns 2954 with status 0x0003057f
thekraken: 2954: stopped with signal 0x00000005
thekraken: 2954: cloned 2961
thekraken: 2961: binding to cpu 5
thekraken: 2954: Continuing.
thekraken: waitpid() returns 2962 with status 0x0000137f
thekraken: 2962: stopped with signal 0x00000013
thekraken: 2962: Continuing.
thekraken: waitpid() returns 2954 with status 0x0003057f
thekraken: 2954: stopped with signal 0x00000005
thekraken: 2954: cloned 2962
thekraken: 2962: binding to cpu 6
thekraken: 2954: Continuing.
thekraken: waitpid() returns 2963 with status 0x0000137f
thekraken: 2963: stopped with signal 0x00000013
thekraken: 2963: Continuing.
thekraken: waitpid() returns 2954 with status 0x0003057f
thekraken: 2954: stopped with signal 0x00000005
thekraken: 2954: cloned 2963
thekraken: 2963: binding to cpu 7
thekraken: 2954: Continuing.
thekraken: waitpid() returns 2964 with status 0x0000137f
thekraken: 2964: stopped with signal 0x00000013
thekraken: 2964: Continuing.
thekraken: waitpid() returns 2954 with status 0x0003057f
thekraken: 2954: stopped with signal 0x00000005
thekraken: 2954: cloned 2964
thekraken: 2964: binding to cpu 8
thekraken: 2954: Continuing.
thekraken: waitpid() returns 2965 with status 0x0000137f
thekraken: 2965: stopped with signal 0x00000013
thekraken: 2965: Continuing.
thekraken: waitpid() returns 2954 with status 0x0003057f
thekraken: 2954: stopped with signal 0x00000005
thekraken: 2954: cloned 2965
thekraken: 2965: binding to cpu 9
thekraken: 2954: Continuing.
thekraken: waitpid() returns 2966 with status 0x0000137f
thekraken: 2966: stopped with signal 0x00000013
thekraken: 2966: Continuing.
thekraken: waitpid() returns 2954 with status 0x0003057f
thekraken: 2954: stopped with signal 0x00000005
thekraken: 2954: cloned 2966
thekraken: 2966: binding to cpu 10
thekraken: 2954: Continuing.
thekraken: waitpid() returns 2967 with status 0x0000137f
thekraken: 2967: stopped with signal 0x00000013
thekraken: 2967: Continuing.
thekraken: waitpid() returns 2954 with status 0x0003057f
thekraken: 2954: stopped with signal 0x00000005
thekraken: 2954: cloned 2967
thekraken: 2967: binding to cpu 11
thekraken: 2954: Continuing.
thekraken: waitpid() returns 2968 with status 0x0000137f
thekraken: 2968: stopped with signal 0x00000013
thekraken: 2968: Continuing.
thekraken: waitpid() returns 2954 with status 0x0003057f
thekraken: 2954: stopped with signal 0x00000005
thekraken: 2954: cloned 2968
thekraken: 2968: binding to cpu 12
thekraken: 2954: Continuing.
thekraken: waitpid() returns 2969 with status 0x0000137f
thekraken: 2969: stopped with signal 0x00000013
thekraken: 2969: Continuing.
thekraken: waitpid() returns 2954 with status 0x0003057f
thekraken: 2954: stopped with signal 0x00000005
thekraken: 2954: cloned 2969
thekraken: 2969: binding to cpu 13
thekraken: 2954: Continuing.
thekraken: waitpid() returns 2970 with status 0x0000137f
thekraken: 2970: stopped with signal 0x00000013
thekraken: 2970: Continuing.
thekraken: waitpid() returns 2954 with status 0x0003057f
thekraken: 2954: stopped with signal 0x00000005
thekraken: 2954: cloned 2970
thekraken: 2970: binding to cpu 14
thekraken: 2954: Continuing.
thekraken: waitpid() returns 2971 with status 0x0000137f
thekraken: 2971: stopped with signal 0x00000013
thekraken: 2971: Continuing.
thekraken: waitpid() returns 2954 with status 0x0003057f
thekraken: 2954: stopped with signal 0x00000005
thekraken: 2954: cloned 2971
thekraken: 2971: binding to cpu 15
thekraken: 2954: Continuing.
thekraken: waitpid() returns 2972 with status 0x0000137f
thekraken: 2972: stopped with signal 0x00000013
thekraken: 2972: Continuing.
thekraken: waitpid() returns 2954 with status 0x0003057f
thekraken: 2954: stopped with signal 0x00000005
thekraken: 2954: cloned 2972
thekraken: 2972: binding to cpu 16
thekraken: 2954: Continuing.
thekraken: waitpid() returns 2973 with status 0x0000137f
thekraken: 2973: stopped with signal 0x00000013
thekraken: 2973: Continuing.
thekraken: waitpid() returns 2954 with status 0x0003057f
thekraken: 2954: stopped with signal 0x00000005
thekraken: 2954: cloned 2973
thekraken: 2973: binding to cpu 17
thekraken: 2954: Continuing.
thekraken: waitpid() returns 2974 with status 0x0000137f
thekraken: 2974: stopped with signal 0x00000013
thekraken: 2974: Continuing.
thekraken: waitpid() returns 2954 with status 0x0003057f
thekraken: 2954: stopped with signal 0x00000005
thekraken: 2954: cloned 2974
thekraken: 2974: binding to cpu 18
thekraken: 2954: Continuing.
thekraken: waitpid() returns 2975 with status 0x0000137f
thekraken: 2975: stopped with signal 0x00000013
thekraken: 2975: Continuing.
thekraken: waitpid() returns 2954 with status 0x0003057f
thekraken: 2954: stopped with signal 0x00000005
thekraken: 2954: cloned 2975
thekraken: 2975: binding to cpu 19
thekraken: 2954: Continuing.
thekraken: waitpid() returns 2976 with status 0x0000137f
thekraken: 2976: stopped with signal 0x00000013
thekraken: 2976: Continuing.
thekraken: waitpid() returns 2954 with status 0x0003057f
thekraken: 2954: stopped with signal 0x00000005
thekraken: 2954: cloned 2976
thekraken: 2976: binding to cpu 20
thekraken: 2954: Continuing.
thekraken: waitpid() returns 2977 with status 0x0000137f
thekraken: 2977: stopped with signal 0x00000013
thekraken: 2977: Continuing.
thekraken: waitpid() returns 2954 with status 0x0003057f
thekraken: 2954: stopped with signal 0x00000005
thekraken: 2954: cloned 2977
thekraken: 2977: binding to cpu 21
thekraken: 2954: Continuing.
thekraken: waitpid() returns 2978 with status 0x0000137f
thekraken: 2978: stopped with signal 0x00000013
thekraken: 2978: Continuing.
thekraken: waitpid() returns 2954 with status 0x0003057f
thekraken: 2954: stopped with signal 0x00000005
thekraken: 2954: cloned 2978
thekraken: 2978: binding to cpu 22
thekraken: 2954: Continuing.
thekraken: waitpid() returns 2979 with status 0x0000137f
thekraken: 2979: stopped with signal 0x00000013
thekraken: 2979: Continuing.
thekraken: waitpid() returns 2954 with status 0x0003057f
thekraken: 2954: stopped with signal 0x00000005
thekraken: 2954: cloned 2979
thekraken: 2979: binding to cpu 23
thekraken: 2954: Continuing.
thekraken: waitpid() returns 2980 with status 0x0000137f
thekraken: 2980: stopped with signal 0x00000013
thekraken: 2980: Continuing.
thekraken: waitpid() returns 2954 with status 0x0003057f
thekraken: 2954: stopped with signal 0x00000005
thekraken: 2954: cloned 2980
thekraken: 2980: binding to cpu 24
thekraken: 2954: Continuing.
thekraken: waitpid() returns 2981 with status 0x0000137f
thekraken: 2981: stopped with signal 0x00000013
thekraken: 2981: Continuing.
thekraken: waitpid() returns 2954 with status 0x0003057f
thekraken: 2954: stopped with signal 0x00000005
thekraken: 2954: cloned 2981
thekraken: 2981: binding to cpu 25
thekraken: 2954: Continuing.
thekraken: waitpid() returns 2982 with status 0x0000137f
thekraken: 2982: stopped with signal 0x00000013
thekraken: 2982: Continuing.
thekraken: waitpid() returns 2954 with status 0x0003057f
thekraken: 2954: stopped with signal 0x00000005
thekraken: 2954: cloned 2982
thekraken: 2982: binding to cpu 26
thekraken: 2954: Continuing.
thekraken: waitpid() returns 2983 with status 0x0000137f
thekraken: 2983: stopped with signal 0x00000013
thekraken: 2983: Continuing.
thekraken: waitpid() returns 2954 with status 0x0003057f
thekraken: 2954: stopped with signal 0x00000005
thekraken: 2954: cloned 2983
thekraken: 2983: binding to cpu 27
thekraken: 2954: Continuing.
thekraken: waitpid() returns 2984 with status 0x0000137f
thekraken: 2984: stopped with signal 0x00000013
thekraken: 2984: Continuing.
thekraken: waitpid() returns 2954 with status 0x0003057f
thekraken: 2954: stopped with signal 0x00000005
thekraken: 2954: cloned 2984
thekraken: 2984: binding to cpu 28
thekraken: 2954: Continuing.
thekraken: waitpid() returns 2985 with status 0x0000137f
thekraken: 2985: stopped with signal 0x00000013
thekraken: 2985: Continuing.
thekraken: waitpid() returns 2954 with status 0x0003057f
thekraken: 2954: stopped with signal 0x00000005
thekraken: 2954: cloned 2985
thekraken: 2985: binding to cpu 29
thekraken: 2954: Continuing.
thekraken: waitpid() returns 2986 with status 0x0000137f
thekraken: 2986: stopped with signal 0x00000013
thekraken: 2986: Continuing.
thekraken: waitpid() returns 2954 with status 0x0003057f
thekraken: 2954: stopped with signal 0x00000005
thekraken: 2954: cloned 2986
thekraken: 2986: binding to cpu 30
thekraken: 2954: Continuing.
thekraken: waitpid() returns 2987 with status 0x0000137f
thekraken: 2987: stopped with signal 0x00000013
thekraken: 2987: Continuing.
thekraken: waitpid() returns 2954 with status 0x0003057f
thekraken: 2954: stopped with signal 0x00000005
thekraken: 2954: cloned 2987
thekraken: 2987: binding to cpu 31
thekraken: 2954: Continuing.
thekraken: autorestart: qualifying checkpoint identified (start: 1335859312, now: 1335861202, mtime: 1335861140, conf: 1), restarting.
thekraken: waitpid() returns 2959 with status 0x00000009
thekraken: 2959: got signal 9
thekraken: autorestart enabled, 33 thread(s) left
thekraken: waitpid() returns 2958 with status 0x00000009
thekraken: 2958: got signal 9
thekraken: autorestart enabled, 32 thread(s) left
thekraken: waitpid() returns 2957 with status 0x00000009
thekraken: 2957: got signal 9
thekraken: autorestart enabled, 31 thread(s) left
thekraken: waitpid() returns 2954 with status 0x00000009
thekraken: 2954: got signal 9
thekraken: autorestart enabled, 30 thread(s) left
thekraken: waitpid() returns 2962 with status 0x00000009
thekraken: 2962: got signal 9
thekraken: autorestart enabled, 29 thread(s) left
thekraken: waitpid() returns 2973 with status 0x00000009
thekraken: 2973: got signal 9
thekraken: autorestart enabled, 28 thread(s) left
thekraken: waitpid() returns 2964 with status 0x00000009
thekraken: 2964: got signal 9
thekraken: autorestart enabled, 27 thread(s) left
thekraken: waitpid() returns 2967 with status 0x00000009
thekraken: 2967: got signal 9
thekraken: autorestart enabled, 26 thread(s) left
thekraken: waitpid() returns 2966 with status 0x00000009
thekraken: 2966: got signal 9
thekraken: autorestart enabled, 25 thread(s) left
thekraken: waitpid() returns 2961 with status 0x00000009
thekraken: 2961: got signal 9
thekraken: autorestart enabled, 24 thread(s) left
thekraken: waitpid() returns 2960 with status 0x00000009
thekraken: 2960: got signal 9
thekraken: autorestart enabled, 23 thread(s) left
thekraken: waitpid() returns 2982 with status 0x00000009
thekraken: 2982: got signal 9
thekraken: autorestart enabled, 22 thread(s) left
thekraken: waitpid() returns 2978 with status 0x00000009
thekraken: 2978: got signal 9
thekraken: autorestart enabled, 21 thread(s) left
thekraken: waitpid() returns 2981 with status 0x00000009
thekraken: 2981: got signal 9
thekraken: autorestart enabled, 20 thread(s) left
thekraken: waitpid() returns 2983 with status 0x00000009
thekraken: 2983: got signal 9
thekraken: autorestart enabled, 19 thread(s) left
thekraken: waitpid() returns 2965 with status 0x00000009
thekraken: 2965: got signal 9
thekraken: autorestart enabled, 18 thread(s) left
thekraken: waitpid() returns 2980 with status 0x00000009
thekraken: 2980: got signal 9
thekraken: autorestart enabled, 17 thread(s) left
thekraken: waitpid() returns 2984 with status 0x00000009
thekraken: 2984: got signal 9
thekraken: autorestart enabled, 16 thread(s) left
thekraken: waitpid() returns 2972 with status 0x00000009
thekraken: 2972: got signal 9
thekraken: autorestart enabled, 15 thread(s) left
thekraken: waitpid() returns 2975 with status 0x00000009
thekraken: 2975: got signal 9
thekraken: autorestart enabled, 14 thread(s) left
thekraken: waitpid() returns 2970 with status 0x00000009
thekraken: 2970: got signal 9
thekraken: autorestart enabled, 13 thread(s) left
thekraken: waitpid() returns 2974 with status 0x00000009
thekraken: 2974: got signal 9
thekraken: autorestart enabled, 12 thread(s) left
thekraken: waitpid() returns 2986 with status 0x00000009
thekraken: 2986: got signal 9
thekraken: autorestart enabled, 11 thread(s) left
thekraken: waitpid() returns 2987 with status 0x00000009
thekraken: 2987: got signal 9
thekraken: autorestart enabled, 10 thread(s) left
thekraken: waitpid() returns 2985 with status 0x00000009
thekraken: 2985: got signal 9
thekraken: autorestart enabled, 9 thread(s) left
thekraken: waitpid() returns 2979 with status 0x00000009
thekraken: 2979: got signal 9
thekraken: autorestart enabled, 8 thread(s) left
thekraken: waitpid() returns 2977 with status 0x00000009
thekraken: 2977: got signal 9
thekraken: autorestart enabled, 7 thread(s) left
thekraken: waitpid() returns 2976 with status 0x00000009
thekraken: 2976: got signal 9
thekraken: autorestart enabled, 6 thread(s) left
thekraken: waitpid() returns 2971 with status 0x00000009
thekraken: 2971: got signal 9
thekraken: autorestart enabled, 5 thread(s) left
thekraken: waitpid() returns 2969 with status 0x00000009
thekraken: 2969: got signal 9
thekraken: autorestart enabled, 4 thread(s) left
thekraken: waitpid() returns 2968 with status 0x00000009
thekraken: 2968: got signal 9
thekraken: autorestart enabled, 3 thread(s) left
thekraken: waitpid() returns 2956 with status 0x00000009
thekraken: 2956: got signal 9
thekraken: autorestart enabled, 2 thread(s) left
thekraken: waitpid() returns 2955 with status 0x00000009
thekraken: 2955: got signal 9
thekraken: autorestart enabled, 1 thread(s) left
thekraken: waitpid() returns 2963 with status 0x00000009
thekraken: 2963: got signal 9
thekraken: autorestart enabled, 0 thread(s) left
thekraken: waitpid() returns 2953 with status 0x00000009
thekraken: 2953: got signal 9
thekraken: main process exited, autorestarting...
thekraken: Forked 3003.
thekraken: child: ptrace(PTRACE_TRACEME) returns 0
thekraken: child: Executing...
thekraken: waitpid() returns 3003 with status 0x0000057f
thekraken: 3003: stopped with signal 0x00000005
thekraken: 3003: initial attach
thekraken: 3003: Continuing.
thekraken: waitpid() returns 3004 with status 0x0000137f
thekraken: 3004: stopped with signal 0x00000013
thekraken: 3004: Continuing.
thekraken: waitpid() returns 3003 with status 0x0003057f
thekraken: 3003: stopped with signal 0x00000005
thekraken: 3003: cloned 3004
thekraken: 3004: binding to cpu 0
thekraken: 3003: Continuing.
thekraken: waitpid() returns 3005 with status 0x0000137f
thekraken: 3005: stopped with signal 0x00000013
thekraken: 3005: Continuing.
thekraken: waitpid() returns 3003 with status 0x0003057f
thekraken: 3003: stopped with signal 0x00000005
thekraken: 3003: cloned 3005
thekraken: 3003: Continuing.
thekraken: waitpid() returns 3006 with status 0x0000137f
thekraken: 3006: stopped with signal 0x00000013
thekraken: 3006: Continuing.
thekraken: waitpid() returns 3003 with status 0x0003057f
thekraken: 3003: stopped with signal 0x00000005
thekraken: 3003: cloned 3006
thekraken: 3003: Continuing.
thekraken: waitpid() returns 3007 with status 0x0000137f
thekraken: 3007: stopped with signal 0x00000013
thekraken: 3007: Continuing.
thekraken: waitpid() returns 3004 with status 0x0003057f
thekraken: 3004: stopped with signal 0x00000005
thekraken: 3004: cloned 3007
thekraken: 3007: binding to cpu 1
thekraken: 3004: Continuing.
thekraken: waitpid() returns 3008 with status 0x0000137f
thekraken: 3008: stopped with signal 0x00000013
thekraken: 3008: Continuing.
thekraken: waitpid() returns 3004 with status 0x0003057f
thekraken: 3004: stopped with signal 0x00000005
thekraken: 3004: cloned 3008
thekraken: 3008: binding to cpu 2
thekraken: 3004: Continuing.
thekraken: waitpid() returns 3009 with status 0x0000137f
thekraken: 3009: stopped with signal 0x00000013
thekraken: 3009: Continuing.
thekraken: waitpid() returns 3004 with status 0x0003057f
thekraken: 3004: stopped with signal 0x00000005
thekraken: 3004: cloned 3009
thekraken: 3009: binding to cpu 3
thekraken: 3004: Continuing.
thekraken: waitpid() returns 3010 with status 0x0000137f
thekraken: 3010: stopped with signal 0x00000013
thekraken: 3010: Continuing.
thekraken: waitpid() returns 3004 with status 0x0003057f
thekraken: 3004: stopped with signal 0x00000005
thekraken: 3004: cloned 3010
thekraken: 3010: binding to cpu 4
thekraken: 3004: Continuing.
thekraken: waitpid() returns 3011 with status 0x0000137f
thekraken: 3011: stopped with signal 0x00000013
thekraken: 3011: Continuing.
thekraken: waitpid() returns 3004 with status 0x0003057f
thekraken: 3004: stopped with signal 0x00000005
thekraken: 3004: cloned 3011
thekraken: 3011: binding to cpu 5
thekraken: 3004: Continuing.
thekraken: waitpid() returns 3012 with status 0x0000137f
thekraken: 3012: stopped with signal 0x00000013
thekraken: 3012: Continuing.
thekraken: waitpid() returns 3004 with status 0x0003057f
thekraken: 3004: stopped with signal 0x00000005
thekraken: 3004: cloned 3012
thekraken: 3012: binding to cpu 6
thekraken: 3004: Continuing.
thekraken: waitpid() returns 3013 with status 0x0000137f
thekraken: 3013: stopped with signal 0x00000013
thekraken: 3013: Continuing.
thekraken: waitpid() returns 3004 with status 0x0003057f
thekraken: 3004: stopped with signal 0x00000005
thekraken: 3004: cloned 3013
thekraken: 3013: binding to cpu 7
thekraken: 3004: Continuing.
thekraken: waitpid() returns 3014 with status 0x0000137f
thekraken: 3014: stopped with signal 0x00000013
thekraken: 3014: Continuing.
thekraken: waitpid() returns 3004 with status 0x0003057f
thekraken: 3004: stopped with signal 0x00000005
thekraken: 3004: cloned 3014
thekraken: 3014: binding to cpu 8
thekraken: 3004: Continuing.
thekraken: waitpid() returns 3015 with status 0x0000137f
thekraken: 3015: stopped with signal 0x00000013
thekraken: 3015: Continuing.
thekraken: waitpid() returns 3004 with status 0x0003057f
thekraken: 3004: stopped with signal 0x00000005
thekraken: 3004: cloned 3015
thekraken: 3015: binding to cpu 9
thekraken: 3004: Continuing.
thekraken: waitpid() returns 3016 with status 0x0000137f
thekraken: 3016: stopped with signal 0x00000013
thekraken: 3016: Continuing.
thekraken: waitpid() returns 3004 with status 0x0003057f
thekraken: 3004: stopped with signal 0x00000005
thekraken: 3004: cloned 3016
thekraken: 3016: binding to cpu 10
thekraken: 3004: Continuing.
thekraken: waitpid() returns 3017 with status 0x0000137f
thekraken: 3017: stopped with signal 0x00000013
thekraken: 3017: Continuing.
thekraken: waitpid() returns 3004 with status 0x0003057f
thekraken: 3004: stopped with signal 0x00000005
thekraken: 3004: cloned 3017
thekraken: 3017: binding to cpu 11
thekraken: 3004: Continuing.
thekraken: waitpid() returns 3018 with status 0x0000137f
thekraken: 3018: stopped with signal 0x00000013
thekraken: 3018: Continuing.
thekraken: waitpid() returns 3004 with status 0x0003057f
thekraken: 3004: stopped with signal 0x00000005
thekraken: 3004: cloned 3018
thekraken: 3018: binding to cpu 12
thekraken: 3004: Continuing.
thekraken: waitpid() returns 3019 with status 0x0000137f
thekraken: 3019: stopped with signal 0x00000013
thekraken: 3019: Continuing.
thekraken: waitpid() returns 3004 with status 0x0003057f
thekraken: 3004: stopped with signal 0x00000005
thekraken: 3004: cloned 3019
thekraken: 3019: binding to cpu 13
thekraken: 3004: Continuing.
thekraken: waitpid() returns 3020 with status 0x0000137f
thekraken: 3020: stopped with signal 0x00000013
thekraken: 3020: Continuing.
thekraken: waitpid() returns 3004 with status 0x0003057f
thekraken: 3004: stopped with signal 0x00000005
thekraken: 3004: cloned 3020
thekraken: 3020: binding to cpu 14
thekraken: 3004: Continuing.
thekraken: waitpid() returns 3021 with status 0x0000137f
thekraken: 3021: stopped with signal 0x00000013
thekraken: 3021: Continuing.
thekraken: waitpid() returns 3004 with status 0x0003057f
thekraken: 3004: stopped with signal 0x00000005
thekraken: 3004: cloned 3021
thekraken: 3021: binding to cpu 15
thekraken: 3004: Continuing.
thekraken: waitpid() returns 3022 with status 0x0000137f
thekraken: 3022: stopped with signal 0x00000013
thekraken: 3022: Continuing.
thekraken: waitpid() returns 3004 with status 0x0003057f
thekraken: 3004: stopped with signal 0x00000005
thekraken: 3004: cloned 3022
thekraken: 3022: binding to cpu 16
thekraken: 3004: Continuing.
thekraken: waitpid() returns 3023 with status 0x0000137f
thekraken: 3023: stopped with signal 0x00000013
thekraken: 3023: Continuing.
thekraken: waitpid() returns 3004 with status 0x0003057f
thekraken: 3004: stopped with signal 0x00000005
thekraken: 3004: cloned 3023
thekraken: 3023: binding to cpu 17
thekraken: 3004: Continuing.
thekraken: waitpid() returns 3024 with status 0x0000137f
thekraken: 3024: stopped with signal 0x00000013
thekraken: 3024: Continuing.
thekraken: waitpid() returns 3004 with status 0x0003057f
thekraken: 3004: stopped with signal 0x00000005
thekraken: 3004: cloned 3024
thekraken: 3024: binding to cpu 18
thekraken: 3004: Continuing.
thekraken: waitpid() returns 3025 with status 0x0000137f
thekraken: 3025: stopped with signal 0x00000013
thekraken: 3025: Continuing.
thekraken: waitpid() returns 3004 with status 0x0003057f
thekraken: 3004: stopped with signal 0x00000005
thekraken: 3004: cloned 3025
thekraken: 3025: binding to cpu 19
thekraken: 3004: Continuing.
thekraken: waitpid() returns 3026 with status 0x0000137f
thekraken: 3026: stopped with signal 0x00000013
thekraken: 3026: Continuing.
thekraken: waitpid() returns 3004 with status 0x0003057f
thekraken: 3004: stopped with signal 0x00000005
thekraken: 3004: cloned 3026
thekraken: 3026: binding to cpu 20
thekraken: 3004: Continuing.
thekraken: waitpid() returns 3027 with status 0x0000137f
thekraken: 3027: stopped with signal 0x00000013
thekraken: 3027: Continuing.
thekraken: waitpid() returns 3004 with status 0x0003057f
thekraken: 3004: stopped with signal 0x00000005
thekraken: 3004: cloned 3027
thekraken: 3027: binding to cpu 21
thekraken: 3004: Continuing.
thekraken: waitpid() returns 3028 with status 0x0000137f
thekraken: 3028: stopped with signal 0x00000013
thekraken: 3028: Continuing.
thekraken: waitpid() returns 3004 with status 0x0003057f
thekraken: 3004: stopped with signal 0x00000005
thekraken: 3004: cloned 3028
thekraken: 3028: binding to cpu 22
thekraken: 3004: Continuing.
thekraken: waitpid() returns 3029 with status 0x0000137f
thekraken: 3029: stopped with signal 0x00000013
thekraken: 3029: Continuing.
thekraken: waitpid() returns 3004 with status 0x0003057f
thekraken: 3004: stopped with signal 0x00000005
thekraken: 3004: cloned 3029
thekraken: 3029: binding to cpu 23
thekraken: 3004: Continuing.
thekraken: waitpid() returns 3030 with status 0x0000137f
thekraken: 3030: stopped with signal 0x00000013
thekraken: 3030: Continuing.
thekraken: waitpid() returns 3004 with status 0x0003057f
thekraken: 3004: stopped with signal 0x00000005
thekraken: 3004: cloned 3030
thekraken: 3030: binding to cpu 24
thekraken: 3004: Continuing.
thekraken: waitpid() returns 3031 with status 0x0000137f
thekraken: 3031: stopped with signal 0x00000013
thekraken: 3031: Continuing.
thekraken: waitpid() returns 3004 with status 0x0003057f
thekraken: 3004: stopped with signal 0x00000005
thekraken: 3004: cloned 3031
thekraken: 3031: binding to cpu 25
thekraken: 3004: Continuing.
thekraken: waitpid() returns 3032 with status 0x0000137f
thekraken: 3032: stopped with signal 0x00000013
thekraken: 3032: Continuing.
thekraken: waitpid() returns 3004 with status 0x0003057f
thekraken: 3004: stopped with signal 0x00000005
thekraken: 3004: cloned 3032
thekraken: 3032: binding to cpu 26
thekraken: 3004: Continuing.
thekraken: waitpid() returns 3033 with status 0x0000137f
thekraken: 3033: stopped with signal 0x00000013
thekraken: 3033: Continuing.
thekraken: waitpid() returns 3004 with status 0x0003057f
thekraken: 3004: stopped with signal 0x00000005
thekraken: 3004: cloned 3033
thekraken: 3033: binding to cpu 27
thekraken: 3004: Continuing.
thekraken: waitpid() returns 3034 with status 0x0000137f
thekraken: 3034: stopped with signal 0x00000013
thekraken: 3034: Continuing.
thekraken: waitpid() returns 3004 with status 0x0003057f
thekraken: 3004: stopped with signal 0x00000005
thekraken: 3004: cloned 3034
thekraken: 3034: binding to cpu 28
thekraken: 3004: Continuing.
thekraken: waitpid() returns 3035 with status 0x0000137f
thekraken: 3035: stopped with signal 0x00000013
thekraken: 3035: Continuing.
thekraken: waitpid() returns 3004 with status 0x0003057f
thekraken: 3004: stopped with signal 0x00000005
thekraken: 3004: cloned 3035
thekraken: 3035: binding to cpu 29
thekraken: 3004: Continuing.
thekraken: waitpid() returns 3036 with status 0x0000137f
thekraken: 3036: stopped with signal 0x00000013
thekraken: 3036: Continuing.
thekraken: waitpid() returns 3004 with status 0x0003057f
thekraken: 3004: stopped with signal 0x00000005
thekraken: 3004: cloned 3036
thekraken: 3036: binding to cpu 30
thekraken: 3004: Continuing.
thekraken: waitpid() returns 3037 with status 0x0000137f
thekraken: 3037: stopped with signal 0x00000013
thekraken: 3037: Continuing.
thekraken: waitpid() returns 3004 with status 0x0003057f
thekraken: 3004: stopped with signal 0x00000005
thekraken: 3004: cloned 3037
thekraken: 3037: binding to cpu 31
thekraken: 3004: Continuing.
Please don't take this as a complaint, it's not. I really respect and admire you for all the selfless work you've provided on the behalf of your Folding brothers. This weekend I'll be building another 4P G34 and am already planning on making use of the Next Generation overclocking resources you and your mad gang have graciously provided. :):):)
 
Linden, the Kraken and DLB are two different things. The Kraken gives NUMA capabilities to the F@H core thereby improving performance on multi-socket/multi-node systems. What Tear has done is added some code to the Kraken which restarts the WU after the first check point to attempt to kick on DLB. DLB is part of the F@H core and as this point has no absolute way to always engage it. The code merely tries to get it to engage without much effect on the running work unit. The Kraken will restart a work unit that has already engaged DLB, and most of the time the DLB continues after the restart.

DLB is an ongoing research project that some here are working on. It is free performance built into the F@H core. It would be nice it there was a flag we could set in the configuration to force DLB always, such as the -forceasm, or -smp XX flag to force a certain thread count the core supports. However, as usual, we are left to our own means to find a way to improve performance anyway we can.
 
Linden said:
The current work unit is folding with both Kraken and DLB engaged properly, so I don't think the current Kraken log would be of use, as there should be nothing out of order.
Click to expand...

Actually, it was useful. It verified that the autorestart feature is on:

thekraken: autorestart is on
Click to expand...


Also, it verified that the kraken restarted the FAH client:

thekraken: autorestart: qualifying checkpoint identified (start: 1335859312, now: 1335861202, mtime: 1335861140, conf: 1), restarting.
thekraken: waitpid() returns 2959 with status 0x00000009
thekraken: 2959: got signal 9
...
thekraken: main process exited, autorestarting...
thekraken: Forked 3003.
thekraken: child: ptrace(PTRACE_TRACEME) returns 0
thekraken: child: Executing...
thekraken: waitpid() returns 3003 with status 0x0000057f
thekraken: 3003: stopped with signal 0x00000005
thekraken: 3003: initial attach
thekraken: 3003: Continuing.
...
Click to expand...


The main purpose of The Kraken is to set CPU affinity of the FahCore threads. These are the threads that do the computations. Setting CPU affinity will cause each thread to run on its own CPU core. e.g. thread 1 is bound to core 1, thread 2 is bound to core 2, etc. This helps to speed up FAH calculations and therefore give reduced frame times.

As for DLB, originally it was hit or miss for it to engage during a WU's processing. Someone (I think it was sfield) observed that DLB is more likely to engage when a unit is restarted from a checkpoint. It is still not guaranteed, but improves the chances of DLB engaging. Based on this observation, Tear added the autorestart option to The Kraken. With the autorestart option enabled, The Kraken will restart the WU after the first checkpoint is written. This will cause the WU to resume from a checkpoint and increases the chances of DLB engaging. There is no guarantee that DLB will engage.

There is a good post by bruce over at FF that gives more details on Dynamic Load Balancing.
 
My request was actually directed as Quisarious.

Kraken's job ends the moment FahCore is restarted. Then all bets are off.
You get DLB? Sweet. You don't? Oh poo.

It's best effort approach. That's it. It was never marketed as reliable DLB "enabler".

Quisarious says he didn't see autorestart occur. THAT is worth looking into.
 
Punchy said:
Apparently theKraken isn't working correctly on SLES11, if you look at the thread at amdzone, though that's not the issue here.

There was a request on ff.org in February 2010 to be able to force DLB on. Maybe in a few more years it will be added to the client... http://foldingforum.org/viewtopic.php?f=44&t=13324
Click to expand...
It's the first time I hear this. Is there a report somewhere that I missed?
 
Also, Linden, I hope documentation does quite good job at explaining what autorestart/DLB are about:
Code: 
6.3. Autorestart feature

    Background: GROMACS employs Dynamic Load Balancing (DLB)
    feature that aims at improving performance.

    GROMACS configuration used by FahCores enables DLB the moment
    cumulative performance loss due to load imbalance exceeds 5%.

    When enabled, DLB reduces times of bigadv units by noticable
    amount of time. Reports include reduction of 30s with P6903
    and 45 seconds with P6904 (sometimes more).

    Depending on WU and system configuration (or even system state),
    DLB gets enabled in a way that may appear random (sometimes it's
    several minutes into WU; at other times it may be as late
    as 90% into WU, sometimes it doesn't engage at all).

    It has been determined that restarting WU from a checkpoint
    significantly increases probability of almost-instantaneous
    DLB engagement (with P6903 and P6904 units).

    Autorestart feature, when enabled, makes The Kraken restart
    FahCore upon completed write of first checkpoint (15 minutes
    in typical configuration).

    To enable autorestart feature add '-c autorestart=1' parameter
    to the command line, when installing, e.g. 'thekraken -i -c autorestart=1'.
    If already installed, uninstall, then install with '-c autorestart=1'.
    Stopping the client is not required.

    NOTE: when enabled, FahCore will appear to have "started twice"
          or restarted without user interaction; this is expected
          and normal

    NOTE: autorestart feature isn't guaranteed; DLB may not always engage

    NOTE: DLB enagagement on units other than P6903 and P6904
          is rare
 
firedfly, tear, once again, thanks for you patience. Sorry you had to explain again for me (three times, as I had previously read the README, but I guess, maybe not thoroughly enough.). :)
 
I reinstalled FAH, so don't have the logs, so cannot be 100% sure, but there was no break in the log (i.e. started WU, 1% done, 2% done.....completed). This was with the 6.34 client, all install done per Musky's guide. Whenever I manually stopped the client and then restarted it, DLB always kicked in after the next % finished.

I've just finished re-installing, so will post a log if same occurs.



tear said:
The Kraken, when properly configured, should restart after first checkpoint.

Please provide thekraken.log from run you didn't observe autorestart so we can tell
more...

If The Kraken determines there is more than one incomplete units in work/
directory it will refuse to autorestart.
Click to expand...
 
I have the same issue and I have 0.6 of the Kraken. Sometimes load balancing works, sometimes it doesn't kick in for a wu or two.
 
Posidon42 said:
I have the same issue and I have 0.6 of the Kraken. Sometimes load balancing works, sometimes it doesn't kick in for a wu or two.
Click to expand...

Dynamic Load Balancing is not garanteed, please read some of the above posts. The Kraken cannot always engage DLB. That is not something the Kraken is capable of doing. The Kraken can only attemp to get the client to engage DLB.
 
UTO, it took them a while to beat it into my head: Kraken is OK, you are OK, it's just the nature of the (DLB) beast vis a vis Folding.
 
This has been beaten to death here but I wanted to document my experience.
I have two nearly identical 4P rigs. Both 6166HE, OC BIOS, Ubuntu 10.10, running theKraken, same MB, same HDD, identical ram, same PSU.
One system restarts and DLB activates 90+% of WUs. If I need to stop and restart for some unrelated reason it appears to always activate DLB.
The other system, just the opposite. Never restarts on new WU and I have only seen it activate DLB a few times over the past month.
Just getting the 3rd rig up and running but it appears to restart and activate DLB most of the time.
Based on this it does not seem so random (luck) but I'm at a loss for what differences there are between my systems.

 
Core32 said:
This has been beaten to death here but I wanted to document my experience.
I have two nearly identical 4P rigs. Both 6166HE, OC BIOS, Ubuntu 10.10, running theKraken, same MB, same HDD, identical ram, same PSU.
One system restarts and DLB activates 90+% of WUs. If I need to stop and restart for some unrelated reason it appears to always activate DLB.
The other system, just the opposite. Never restarts on new WU and I have only seen it activate DLB a few times over the past month.
Just getting the 3rd rig up and running but it appears to restart and activate DLB most of the time.
Based on this it does not seem so random (luck) but I'm at a loss for what differences there are between my systems.

Click to expand...

When did you install Ubuntu, and did you follow my guide the last time? I may be able to explain it.
 
On the first two systems, over a month ago, and yes I followed your guide pretty closely.
This third system I just installed this past week.
 
If you installed thekraken after 9:32 PM CST yesterday from the guide instructions, that would explain it. If you installed it last week some time, I have no idea.
 
Core32 said:
Never restarts on new WU
Click to expand...

On the "problematic" machine --

if you examine contents of thekraken.cfg you should
see 'autorestart=1' line... if you don't, Kraken hasn't
been configured properly.

Feel free to paste thekraken.log as well; it will help
us determine why it's not doing what we think
it should be doing....
 
musky said:
If you installed thekraken after 9:32 PM CST yesterday from the guide instructions, that would explain it. If you installed it last week some time, I have no idea.
Click to expand...

Definitely all three installed before that day and time.

tear said:
On the "problematic" machine --

if you examine contents of thekraken.cfg you should
see 'autorestart=1' line... if you don't, Kraken hasn't
been configured properly.
Click to expand...

All three system .cfg files have this line, correct.

tear said:
Feel free to paste thekraken.log as well; it will help
us determine why it's not doing what we think
it should be doing....
Click to expand...

I'll get those up tonight.
Thanks.
 
tear said:
On the "problematic" machine --
Feel free to paste thekraken.log as well; it will help
us determine why it's not doing what we think
it should be doing....
Click to expand...

What portion do you need to see?
The file is quite long and I'm hesitant to post it in the thread.
Thanks.
 
Kraken log file from worst offender:

Code: 
Core32-H8QG6:~/fah$ cat thekraken-prev.log 
thekraken: The Kraken 0.6 (compiled Mon Feb 20 21:12:14 EST 2012 by Core32@Core32-H8QG6)
thekraken: Processor affinity wrapper for Folding@Home
thekraken: The Kraken comes with ABSOLUTELY NO WARRANTY; licensed under GPLv2
thekraken: PID: 31561
thekraken: launch binary: ./thekraken-FahCore_a5.exe
thekraken: config file: ./thekraken.cfg
thekraken: config: autorestart=1
thekraken: autorestart: examining work directory...
thekraken: autorestart is on
thekraken: autorestart slot: 0
thekraken: Forked 31562.
thekraken: autorestart: filesystem time offset: 0 seconds
thekraken: child: ptrace(PTRACE_TRACEME) returns 0
thekraken: child: Executing...
thekraken: waitpid() returns 31562 with status 0x0000057f
thekraken: 31562: stopped with signal 0x00000005
thekraken: 31562: initial attach
thekraken: 31562: Continuing.
thekraken: waitpid() returns 31564 with status 0x0000137f
thekraken: 31564: stopped with signal 0x00000013
thekraken: 31564: Continuing.
thekraken: waitpid() returns 31562 with status 0x0003057f
thekraken: 31562: stopped with signal 0x00000005
thekraken: 31562: cloned 31564
thekraken: 31564: binding to cpu 0
thekraken: 31562: Continuing.
thekraken: waitpid() returns 31565 with status 0x0000137f
thekraken: 31565: stopped with signal 0x00000013
thekraken: 31565: Continuing.
thekraken: waitpid() returns 31562 with status 0x0003057f
thekraken: 31562: stopped with signal 0x00000005
thekraken: 31562: cloned 31565
thekraken: 31562: Continuing.
thekraken: waitpid() returns 31566 with status 0x0000137f
thekraken: 31566: stopped with signal 0x00000013
thekraken: 31566: Continuing.
thekraken: waitpid() returns 31562 with status 0x0003057f
thekraken: 31562: stopped with signal 0x00000005
thekraken: 31562: cloned 31566
thekraken: 31562: Continuing.
thekraken: waitpid() returns 31565 with status 0x00000009
thekraken: 31565: got signal 9
thekraken: autorestart enabled, 2 thread(s) left
thekraken: waitpid() returns 31564 with status 0x00000009
thekraken: 31564: got signal 9
thekraken: autorestart enabled, 1 thread(s) left
thekraken: waitpid() returns 31566 with status 0x00000009
thekraken: 31566: got signal 9
thekraken: autorestart enabled, 0 thread(s) left
thekraken: waitpid() returns 31562 with status 0x00000009
thekraken: 31562: got signal 9
thekraken: main process exited, autorestarting...
thekraken: Forked 32296.
thekraken: child: ptrace(PTRACE_TRACEME) returns 0
thekraken: child: Executing...
thekraken: waitpid() returns 32296 with status 0x0000057f
thekraken: 32296: stopped with signal 0x00000005
thekraken: 32296: initial attach
thekraken: 32296: Continuing.
thekraken: waitpid() returns 32300 with status 0x0000137f
thekraken: 32300: stopped with signal 0x00000013
thekraken: 32300: Continuing.
thekraken: waitpid() returns 32296 with status 0x0003057f
thekraken: 32296: stopped with signal 0x00000005
thekraken: 32296: cloned 32300
thekraken: 32300: binding to cpu 0
thekraken: 32296: Continuing.
thekraken: waitpid() returns 32296 with status 0x0003057f
thekraken: 32296: stopped with signal 0x00000005
thekraken: 32296: cloned 32301
thekraken: 32296: Continuing.
thekraken: waitpid() returns 32301 with status 0x0000137f
thekraken: 32301: stopped with signal 0x00000013
thekraken: 32301: Continuing.
thekraken: waitpid() returns 32296 with status 0x0003057f
thekraken: 32296: stopped with signal 0x00000005
thekraken: 32296: cloned 32302
thekraken: 32296: Continuing.
thekraken: waitpid() returns 32302 with status 0x0000137f
thekraken: 32302: stopped with signal 0x00000013
thekraken: 32302: Continuing.
thekraken: waitpid() returns 32302 with status 0x00000009
thekraken: 32302: got signal 9

And here is the -prev.log file:

Code: 
Core32-H8QG6:~/fah$ cat thekraken-prev.log 
thekraken: The Kraken 0.6 (compiled Mon Feb 20 21:12:14 EST 2012 by Core32@Core32-H8QG6)
thekraken: Processor affinity wrapper for Folding@Home
thekraken: The Kraken comes with ABSOLUTELY NO WARRANTY; licensed under GPLv2
thekraken: PID: 31546
thekraken: launch binary: ./thekraken-FahCore_a5.exe
thekraken: config file: ./thekraken.cfg
thekraken: config: autorestart=1
thekraken: autorestart: examining work directory...
thekraken: autorestart is on
thekraken: autorestart slot: 9
thekraken: Forked 31547.
thekraken: autorestart: filesystem time offset: 0 seconds
thekraken: child: ptrace(PTRACE_TRACEME) returns 0
thekraken: child: Executing...
thekraken: waitpid() returns 31547 with status 0x0000057f
thekraken: 31547: stopped with signal 0x00000005
thekraken: 31547: initial attach
thekraken: 31547: Continuing.
thekraken: waitpid() returns 31547 with status 0x00006500
thekraken: 31547: exited with 101
 
Core32 said:
Kraken log file from worst offender:
Click to expand...

How are you running the fah client (which flags are in use)? To me, it looks like you are only running on one core. The kraken is only picking up and setting affinity on one fah worker thread.
 
Core32 said:
-smp 48 and -big......
Definitely all cores are running.
Click to expand...

In that case, you will have to wait for tear to comment. The kraken is only picking up on one fah worker thread which is odd for -smp 48 (I've never seen that before).
 
Just FWIW, tear had me remove the ## from the -smp 64 flag on my 4p machine. Tested with regular old -smp and it works just fine. (IIRC it was Windows that needed the number of cores specifically called out for anything over 32) Have the latest Kraken running on both 4P's and two 2P's and it's working great. DLB turns on everytime, where as before it was off more often than not.

I'm still very green at Linux, but removed with thekraken -u, deleted the old Kraken folder, created the new one, extracted files, cd into kraken, make, su make install, thekraken -i and off to the races.

Maybe, if worse comes to worse, create a new FAH directory and start from there?
 
sbinh said:
maybe thekraken didnot wrap the core client (A3/A5) ?
Click to expand...

Dang it! That would appear to be the case in this instance......
I've checked "top" a dozen times I would swear over the last week.
This time it was showing the Fahcore.... running instead.
Just re-ran "thekraken -i" again and started up.
It now shows in "top" so we'll see if it restarts the client.
Not sure how thekraken did not start onn the last boot up.

:eek: :eek: :eek: :eek: :eek: I am......
 
To check if the FAH core is wrapped, in the FAH folder, look for the file, "thekraken-FahCore_a5.exe."
 
The wrap files are there. At one time this was functioning fine, as I always checked top to see after a re-start.
Now, obviously, senility has set in and I dream my way through WUs. :rolleyes:
 
You must log in or register to reply here.
Back
Top