SETI@Home Wow!-Event 2019

[Pokey]

I saw where the WOW event was getting ready to start and decided to find out if my rigs still worked. They do. But out dated, and it is taking some time to get them back in working order. Good to see you all are still puttering around.
I have two more rigs to try and get going before the race starts so I better get busy.

 

[Orange Kid]

"Max concurrent" seems to allow pretty random behavior from what I have seen, it can exclude a GPU in favor of another or in favor of CPU tasks, it's not a great way to control the behavior of multiple devices in a single instance, imo.

...and the better option is? Always willing to learn new stuff.

@Pokey good to see you back.

 

[Modular]

@Assimilator1 you're running W7, right? What does the event manager say after a reboot?

 

[lane42]

Less then an hour to go.
I think well do good

 

[lane42]

And were off
I think some will need this

 

[ao_ika_red]

Fire! Fire everything!!!
(Except if you have more than 100M worth of tasks, I beg you to release them slowly)

 

[StefanR5R]

Hmm, I'm having problems with my rig since I've started running GPU tasks, a problem I had before when I ran GPU tasks but I'd forgotten about!
Anyway, the problem is my PC reboots after an unknown time, but always when I'm not using it!
The CPU has recently passed an OCCT stress test for 33+hrs so it's not that, the grx card isn't o/ced. GPU temp atm is 61-65C atm, & it doesn't overheat playing Elite Dangerous either (or reboot/crash), peak CPU temp's are 61-62C. No invalid tasks for SETI, 2 error tasks but that is due to them expiring.
I'm thinking maybe a voltage issue, but nothing is looking dodgy in speedfan atm, although it can't read the 12v line. This is going to be a PITA to pin down........

Could it be due to amperage spikes through the GPU, which are taxing the PSU too much? Do you know whether your PSU has a single 12 V rail, or several?

On the other hand, it could be purely a software problem (a crash of a kernel-level driver).

 

[lane42]

Just don't run out of work. You wont get new work until you upload
what you have, and if there's a download snag youll run dry.

 

[StefanR5R]

On my 2700x what I have done for this event as I am running both cpu and gpu tasks is create an app_config.xml with the following in it.

<app_config>
<project_max_concurrent>16</project_max_concurrent>
</app_config>

What this does is only allow 16 tasks to run a one time. Either 16 cpu tasks or 15 cpu tasks and one gpu task. Without it Seti will use whatever cpu's are available to run both cpu and gpu. IE: without it, Seti will run 16 cpu tasks and one gpu task, 17 tasks total, not giving a whole cpu to the gpu.
An added benefit is if you are using the computer and the gpu task shuts down a cpu task runs.

"Max concurrent" seems to allow pretty random behavior from what I have seen, it can exclude a GPU in favor of another or in favor of CPU tasks, it's not a great way to control the behavior of multiple devices in a single instance, imo.

I guess @Orange Kid is running S@H exclusively on the machine now, and for that, max_concurrent or project_max_concurrent may work pretty well (better than with a mix of projects). Reasons:

• S@H's task deadlines are very long, relative to the depth of the work buffer that you can normally achieve. Hence it's unlikely that the client would suddenly freak out in belief that some deadlines couldn't be met, and run the wrong jobs.
• If the client handles work from two or more different projects, it has its own twisted ideas about which work is more important. If it is just S@H-CPU vs. S@H-GPU, my guess is that this sort of confusion cannot happen.

Another good use case of (project_)max_concurrent, as I have recently learned from some good advice via PM, is when the user knows better than the client how many GPUs are present in the host. :-)

But apart from cases like these, I too think that (project_)max_concurrent is often less practical than it may seem.

 

[Assimilator1]

@Assimilator1 you're running W7, right? What does the event manager say after a reboot?

Good point, I'll check. I can't find an event manager, but I did find an event viewer, is that what you meant?
Critical - Event ID 41 kernal-power - The system has rebooted without cleanly shutting down first. This error could be caused if the system stopped responding, crashed, or lost power unexpectedly.
After rebooting their is a small window open saying windows has encountered an error. Not much help there.

Could it be due to amperage spikes through the GPU, which are taxing the PSU too much? Do you know whether your PSU has a single 12 V rail, or several?

On the other hand, it could be purely a software problem (a crash of a kernel-level driver).

I can't remember the number of rails it has, I'll check later, along with getting my DMM out to check the 12v rail readings. I do remember it's an old Corsair 650w PSU, more info later....[update] It's a Corsair TX650w, it's got a single 12v rail. Apparently it's MBTF is ~11.4 yrs, I wonder when I got it? I think maybe I got it for when I 1st built my C2 rig(2007)! lol, although it might have been a bit later.

DMM readings from my PSU whilst running SETI on CPU (10 threads) & GPU :-
12v line - wire1 11.95v, wire 2 11.96v, wire 3 11.97v (small plug to mbrd), also seen at the main ATX plug (different WU maybe?) And a steady 12v (11.97-12.03v) to the GPU plug, I watched it on & off over several minutes whilst also looking at GPU-Z to see GPU load.
5v line - 4.86v (main ATX plug), 3v line - 3.32v (main ATX plug).
That good ole PSU is chugging along nicely

I wonder if it's some sort of problem relating to when the monitors go to sleep.....

 

[TennesseeTony]

I think I see a force coming alive, welcome back Pokey old Friend.

I saw where the WOW event was getting ready to start and decided to find out if my rigs still worked. They do. But out dated, and it is taking some time to get them back in working order. Good to see you all are still puttering around.
I have two more rigs to try and get going before the race starts so I better get busy.

WOOOOOO WHOOOOOOOO!!!!!!!

Seriously, I thought you may have gone past your expiration date...so glad to have you still with us, friend!

 

[TennesseeTony]

In regards to experimenting....Regarding running only one task per GPU..........my GTX 1070's are saving 2.5 seconds per cuda101 task by running 2 at a time.... (AMD FX-8350, DDR3-1066, Linux Mint 19.2). I went from a minute.35 to 3 minutes.05. That is 172 extra tasks per 1070 over a 2 week period....

edit: I previously stated I will have only a single Windows 1080Ti to contribute, but that has changed to twin Linux 1070's, for 200% the output. (and about the same energy cost)

 

[lane42]

Seriously, I thought you may have gone past your expiration date...so glad to have you still with us, friend!

Iam an old fart and I think Pokey called me kid once

 

[Markfw]

Iam an old fart and I think Pokey called me kid once

Smoke is older than me....And I am an old fart (65)

In spirit, I am still 18, gets me into trouble at times.

 

[TennesseeTony]

I am crying with JOY just to know you old-sons-of............................................................................-sweet-sweet-ladies are still amongst us.


So who do I get to call kid? 1.) Kiska 2.) Red 3.) Uhmmmmmm...phoenicis? Skillz? Almost 50 here.....

 

[burninatortech4]

I am crying with JOY just to know you old-sons-of............................................................................-sweet-sweet-ladies are still amongst us.


So who do I get to call kid? 1.) Kiska 2.) Red 3.) Uhmmmmmm...phoenicis? Skillz? Almost 50 here.....

I'm 30. I'm pretty sure that counts, right? ;P My dad is almost 78 (ETPhoneHome) and he's just as stoked about this as I am.

 

[crashtech]

Nice to see such great turnout!

 

[Markfw]

ummmm My latest change to the 1600/2060 must have done something. I am at 132k for today so far ! vs 73 k for yesterday.

 

[ao_ika_red]

So who do I get to call kid? 1.) Kiska 2.) Red

It's me obviously, the 26 years old kid. But I do believe @Howdy is younger than me.

 

[Kiska]

I am crying with JOY just to know you old-sons-of............................................................................-sweet-sweet-ladies are still amongst us.


So who do I get to call kid? 1.) Kiska 2.) Red 3.) Uhmmmmmm...phoenicis? Skillz? Almost 50 here.....

It's me obviously, the 26 years old kid. But I do believe @Howdy is younger than me.

I think that would be me. I am 21

 

[Assimilator1]

I'm 21! , yea I was posting when I was a yr old

(btw, no reboots last night [insert confused smilie which doesn't seem to exist now!])

 

[ao_ika_red]

I think that would be me. I am 21

Rare case of young professor.

 

[IEC]

When I was a young pup... err, wait, I still am!

 

[ao_ika_red]

I feel a great disturbance in the Force (seti's server).


zoomed-in

Happened since 1112 UTC (approx)

 

[StefanR5R]

And were off
I think some will need this
View attachment 9611

Obviously, @biodoc at least doesn't have any troubles in this regard (or not anymore).

I keep pointing out to my twin xii5ku that it says on the box "do not swallow", but this guy just doesn't listen.