Posts by Jimbocous

21) Message boards : Number crunching : The Server Issues / Outages Thread - Panic Mode On! (119) (Message 2043635)
Posted 8 Apr 2020 by Profile Jimbocous Project Donor
Post:
Well, the replica has finally caught up and is back in synch.

Meow meow meow.

Wee Haw meow!
22) Message boards : Number crunching : The Server Issues / Outages Thread - Panic Mode On! (119) (Message 2043615)
Posted 8 Apr 2020 by Profile Jimbocous Project Donor
Post:
Another system (from another thread).

In progress 25281
And it's been a week since last contact with the server.

And with the user being Anonymous, for all we know they could have several systems in the same state.

Given it's a spoofed GPU system on 7.16.7, shouldn't be too hard to figure out who it belongs to. Creative bunkering, anyone?
23) Message boards : Number crunching : Setting up Linux to crunch CUDA90 and above for Windows users (Message 2043471)
Posted 7 Apr 2020 by Profile Jimbocous Project Donor
Post:

the ONLY problem I ever had was after running the coolbits command and writing an xorg.conf file, sometimes the bus ID gets changed of the GPU driving the monitor. ... just manually edit the xorg.conf file to return device/screen 0 to being the bus ID back to the GPU that shows the BIOS at boot. 

sudo nano /etc/X11/xorg.conf
--make changes to file in nano--
[Ctrl]+X to save changes
reboot

That's been the issue that's plagued me on my HP boxes since day 1. Can't ever count on where the desktop is actually going to show up. The times I tried to change the xorg.conf file I ended up unbootable, and had to go back to the backup.
Must be about time for me to tackle that issue again, now that SETI is wound down and most crunchers are turned off, and figure out how I'm doing wrong something that should be so simple.
For anyone who is interested, I finally figured out my problem with trying to modify xorg.conf, and it was a strange one.
When trying to sort out which GPU was which, as it turns out nvidia writes PCI bus ID numbers in xorg.conf as decimal numbers, but it reports them via nvidia-smi in hexadecimal. On the HP boxes, the dec numbers of one GPU match the hex numbers of a different one, so each time I was editing ID 28h I was changing 40dec, as opposed to 28dec which was of course 1Ch. Wasn't immediately obvious, given that the first 4 GPUs were devs 04-05-06-07, which of course are the same either way.
Always the little things ...
24) Message boards : Number crunching : The Server Issues / Outages Thread - Panic Mode On! (119) (Message 2043468)
Posted 7 Apr 2020 by Profile Jimbocous Project Donor
Post:
Because of systems like this one.
Last contact, April 1
Tasks in progress, 14,409!
Something really messed up here, and I don't think it's the client, but rather the DB.
. Error tasks that timed out in 24 hours?
. MB tasks with a 14 day timeout?
. 14k tasks in progress on a machine with 1 GPU?
25) Message boards : Number crunching : Anything relating to AstroPulse (3) tasks (Message 2043420)
Posted 7 Apr 2020 by Profile Jimbocous Project Donor
Post:
31=4
1=1
2=4
26) Message boards : Number crunching : Setting up Linux to crunch CUDA90 and above for Windows users (Message 2043283)
Posted 6 Apr 2020 by Profile Jimbocous Project Donor
Post:
Do you ever use the LInux box to remote into windows? I am having a heck of a time......
OK, so just for grins I checked this out.
At least on my 18.04, Remmina is part of the distributions and is ready to use. Near as I can tell, this has been around and included since Ubuntu 12 or so.
1) Open a terminal window.
2) Just invoke it: $ remmina
3) Pops up GUI window, offers choice of RDP, SSH and VNC connections. Select RDP. (Cute, I'll have to see how SSH does to another Linux box. Useful)
4) Enter Computer Name of remote Win box. ( You did enable remote access, right? Do select Network Level Authentication when enabling it in Windows Settings.)
5) Enter your username on that machine and, if needed, password.
6) You may need to adjust color settings if you'll get an error. The default H264 generated an error. I found 32 bit Truecolor to work.
7) Save this as default.
8) Connect.

If you're running BOINC on the Win box, you need to understand that GPU processing will stop during the RDP session. This is a Windows issue, in that the RDP handler has its own video driver, and shuts down the GPU driver while the RDP session is active. Always has, which is why when all my crunchers were still Win7/10 I never used RDP from the daily driver to do stuff on them.
There may be a way to save the connection info for the next session, but if so I didn't find it. Looks like the info can be found in remmina --help, though

Hope this is useful. Thanks for asking the question, glad I learned this for that one time I'll need it. :)
Later, Jim ...
27) Message boards : Number crunching : Anything relating to AstroPulse (3) tasks (Message 2043279)
Posted 6 Apr 2020 by Profile Jimbocous Project Donor
Post:
4 turned up here for the 31st UTC.
Ditto, so far ....
28) Message boards : Number crunching : Don't know where it should go? Stick it here! (Message 2043278)
Posted 6 Apr 2020 by Profile Jimbocous Project Donor
Post:
It may have taken a PhD to write BOINC, but it shouldn't require one to configure it.
This thing load-shares like three teenagers and a dog in the back seat of the car.
Just sayin' ...
I suspect that when developing it they had no idea of the ways in which people would try to get things to run. And many of the options can conflict with other options, combined with Anonymous Platform support, the way Resource sharing is determined, variable processing times & hugely varying deadlines and hardware requirements along with incredibly small to ridiculously huge caches and people using ncpus & other such settings in ways the developers never imagined in their wildest dreams (worst nightmares?) results in things not happening in quite the way people might expect them too, and certainly not within their expected time frame- which generally seems to be immediately, if not retrospectively.

As I said, like 3 teenagers and a dog. :)
29) Message boards : Number crunching : Don't know where it should go? Stick it here! (Message 2043259)
Posted 6 Apr 2020 by Profile Jimbocous Project Donor
Post:
It may have taken a PhD to write BOINC, but it shouldn't require one to configure it.
This thing load-shares like three teenagers and a dog in the back seat of the car.
Just sayin' ...
30) Message boards : Number crunching : Setting up Linux to crunch CUDA90 and above for Windows users (Message 2043256)
Posted 6 Apr 2020 by Profile Jimbocous Project Donor
Post:

the ONLY problem I ever had was after running the coolbits command and writing an xorg.conf file, sometimes the bus ID gets changed of the GPU driving the monitor. ... just manually edit the xorg.conf file to return device/screen 0 to being the bus ID back to the GPU that shows the BIOS at boot. 

sudo nano /etc/X11/xorg.conf
--make changes to file in nano--
[Ctrl]+X to save changes
reboot

That's been the issue that's plagued me on my HP boxes since day 1. Can't ever count on where the desktop is actually going to show up. The times I tried to change the xorg.conf file I ended up unbootable, and had to go back to the backup.
Must be about time for me to tackle that issue again, now that SETI is wound down and most crunchers are turned off, and figure out how I'm doing wrong something that should be so simple.
31) Message boards : Number crunching : Setting up Linux to crunch CUDA90 and above for Windows users (Message 2043207)
Posted 5 Apr 2020 by Profile Jimbocous Project Donor
Post:
Do you ever use the LInux box to remote into windows? I am having a heck of a time......
Can't say I've had a need to do that, though I can see a couple ways it could be done. There's a remote Desktop app for Linux out there called Remmina Remote Desktop Client. Haven't tried it, but it looks promising .
I tried to set up XRDP for GUI access via RDP to Ubuntu, but had no luck with it due primarily, I believe, to interactions with the NVidia drivers or my poor understanding of XOrg. That's what led me to what I detailed above.
32) Message boards : Number crunching : Setting up Linux to crunch CUDA90 and above for Windows users (Message 2043202)
Posted 5 Apr 2020 by Profile Jimbocous Project Donor
Post:
If you are going "around and around" with Ubuntu 18.4.x for anything from the logon loop to not being able to get the Nvidia drivers to install via "Launchpad" then you might want to try Zorin (core).
Now that I've refreshed my ux head a bit, I can't say I've found much grief with 18.04.
There are a few things I do, however, that make life easier.
1) Load Samba, so I can use my Win10 box across the network for basic file edits and admin, especially LAMP admin.
2) Enable local SSH and load PuTTY on the Win box, for places where Win across the network can't handle the job (e.g. ownership, permissions, symlinks, etc.)
3) Forget Gnome/Nautilus exist. Not needed with 1,2 above in place.
4) Do use dos2unix across SSH where needed after edits.
5) Do updates and installs across SSH from a terminal window. Easier than GUI anyway.
These days, about the only time I go downstairs is when I actually need to touch the box itself.
33) Message boards : Number crunching : The Server Issues / Outages Thread - Panic Mode On! (119) (Message 2043068)
Posted 4 Apr 2020 by Profile Jimbocous Project Donor
Post:

Thanks for the tip but I can't do that.
Remember due my unique client, what you see is not what it rely is.

My host production is in the range of 50-70 WU each 5 min.
Now with the 10 min interval i'm sending 100-140 WU on each request.


Does not matter.

Set NNT and restrict max tasks to a low amount and try.

If it works you do not have to wait just hit the update button again.

The wait was only introduced to stop people swamping the server with work requests.

Seems to me I remember Richard saying something about the "sweet spot" for max reporting being 64 for Win and 128 for Linux. After a lot of experimenting, I found that to be a good compromise here.
34) Message boards : Number crunching : Don't know where it should go? Stick it here! (Message 2043056)
Posted 4 Apr 2020 by Profile Jimbocous Project Donor
Post:

Interesting it's the first time I've heard of the power company contacting someone to see if something is wrong (said with tongue-in-cheek)
LOL, mine sends me a graph each month showing me how much more energy I use than my neighbors. Wonder if that will now end?
35) Message boards : News : SETI@home hibernation (Message 2043052)
Posted 4 Apr 2020 by Profile Jimbocous Project Donor
Post:
Einstein@home really uses your pc ram. All of it if you let it on some work units but it doesn't seem to cause a lot of heat in the cpu even though all cors at 100%
On my Linux box, this is my CPU RAM:
              total        used        free      shared  buff/cache   available
Mem:            23G        5.6G         16G         24M        1.0G         17G
This is with 3xGPUs each running 1 task, and 11 CPU tasks running.

On the Windows box, again the RAM usage seems reasonable:
Total Physical Memory:     12,271 MB
Available Physical Memory: 6,920 MB
Virtual Memory: Max Size:  24,559 MB
Virtual Memory: Available: 16,731 MB
Virtual Memory: In Use:    7,828 MB
This is with 4 CPU tasks, 2xGPUs, and my normal "daily driver" stuff running on Win10Prox64.

Note:
I only run the Gamma-Ray Pulsar Searches on CPU and GPU.
The Gravity Wave stuff is indeed a "heavy hitter" in resource needs.
Perhaps that's what you're referring to?
l8r, Jim ...
36) Message boards : Number crunching : The Server Issues / Outages Thread - Panic Mode On! (119) (Message 2043043)
Posted 4 Apr 2020 by Profile Jimbocous Project Donor
Post:
Mine's 15kb

Thanks. That's solve the puzzle.

Something was changed and the servers not handle anymore my large file.

Back to the drawing board. Will remain running with NNT until i think on something.

Thanks again for the help.

No problem.
Wondering, if that was deleted would BOINC just rebuild it on restart? Would think so.
37) Message boards : Number crunching : The Server Issues / Outages Thread - Panic Mode On! (119) (Message 2043040)
Posted 4 Apr 2020 by Profile Jimbocous Project Donor
Post:
Mine's 15kb.
38) Message boards : Number crunching : The Server Issues / Outages Thread - Panic Mode On! (119) (Message 2043038)
Posted 4 Apr 2020 by Profile Jimbocous Project Donor
Post:
04-Apr-2020 18:43:29 [SETI@home] Scheduler request completed: got 1 new tasks


another one ;)


. . Braggart :)

Stephen

:(

Could you make me a favor?

Can you look and post the size of your setiathome.berkeley.edu.xml file? Its located on the boinc directory (the one where the boinc.exe program is placed)

Thanks in advance.

if you're looking for account_setiathome.berkeley.edu.xml, mine's 6kb.
As I recall, only the first three lines, 
    <master_url>,
    <authenticator>,
    <project_name>,

are needed to connect. The remainder gets filled in from the website from your preferences as defined there.
39) Message boards : Number crunching : Anything relating to AstroPulse (3) tasks (Message 2043032)
Posted 4 Apr 2020 by Profile Jimbocous Project Donor
Post:
28th=51
29th=55
30th=24 45?
40) Message boards : News : SETI@home hibernation (Message 2043010)
Posted 4 Apr 2020 by Profile Jimbocous Project Donor
Post:


Only Einstein@home supports GPU though.

~Yav
And MilkyWay does also.


Previous 20 · Next 20


 
©2025 University of California
 
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.