• 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.

Notfreds problem

Kendrak

[H]ard|DCer of the Year 2009
Joined
Aug 29, 2001
Messages
21,141
[07:44:34] Timered checkpoint triggered.
[07:44:38] Writing local files
[07:44:38] Completed 2700000 out of 5000000 steps (54 percent)
[07:59:39] Timered checkpoint triggered.
[07:59:45] Writing local files
[07:59:45] Completed 2750000 out of 5000000 steps (55 percent)
[08:14:46] Timered checkpoint triggered.
[08:14:49] Writing local files
[08:14:49] Completed 2800000 out of 5000000 steps (56 percent)
[08:26:48] Warning: long 1-4 interactions
[08:26:48]
[08:26:48] Folding@home Core Shutdown: INTERRUPTED
[08:26:52] CoreStatus = 66 (102)
[08:26:52] + Shutdown requested by user. Exiting.***** Got a SIGTERM signal (15)
[08:26:52] Killing all core threads

Folding@Home Client Shutdown.

This happens on both of my headless systems from time to time. Just one of the instances (I am running 2) What gives?

I'm not telling the cores to shut down :confused:

 
This is often a indication of a borderline overclocking. Try to back down the clock a notch (the error happened probably because the master core crashed so the client thought it was a shutdown request).

 
This is often a indication of a borderline overclocking. Try to back down the clock a notch (the error happened probably because the master core crashed so the client thought it was a shutdown request).


/sigh

Got some work to do today

Headless/ diskless systems are great, untill you have to work on them ;)

 
This is often a indication of a borderline overclocking. Try to back down the clock a notch (the error happened probably because the master core crashed so the client thought it was a shutdown request).


Its not always unstable OC - I have Q6600 @3.2Ghz which is running folding CD 1 month 24/7 without problem but when I tried to use it with XP with folding CD in VMware I had some error at least 1x every week.

First problem was same as Kendrak´s.

Second problem was that after it finished WU and sent it it didnt continue and I had to restart it.

 
OK, where's nomad8u (only kidding :p), he seems to be an authority on the notfred CD and I've benefited from his and many others (Tigerbiten, Xilikon, SmokeRngs, Kendrak, et al) on any problems I've had with a notfred CD, especially with the VMWare image (vmx ?)

FOLD ON!

Stay safe everyone in the path of "Gustav" (especially to our brother folder [XC] Synthetickiller from Xtremesystems.org):D

 
If anyone whips out a stable FredCD that is ready to "boot&fold", I'm up for it. I've not had luck with it, so I play games with self boot Ubuntu CD's & FAH installs(A real pain when I need to reboot the various PC's.

I realize an edit to the config file will be required, but saves me the PIA factor of d/l'ing same client over and over....
 
OK, where's nomad8u (only kidding :p

Right here dude! :D Just popped on and saw this.

I had the same problem a couple times in the past and it was indeed a borderline OC on the processor. Just a tad more Vcore seemed to fix that issue up most ricky-tick. I was also running all my Notfred VM's @ 512m of ram. I found that increasing them to 768 helped stability when picking up the 2665 WU's and didn't hurt with the 2662's as well. If you have enough ram, a meg per VM is even better but I haven't had an issue with the 768 yet.

I had the same issue (as have others without using Notfreds) as Rebel44 on the 2662 WU as well where the WU would complete and not upload. Several sat for a day like that. If you have a USB backup stick configured, killing and restarting the VM seems to flush it up but that requires babysitting.

A recent issue this week on 2 machines (both running GPU + dual Notfred VM's caused me to drop the second VM on both machines and convert a core to run WCG.
Both picked up 2665 WU and wouldn't start. I killed the VM's on both machines twice and tried to restart with no luck. I killed the VM's twice again on both machines and flushed the USB sticks to flush the WU and re-download. Both machines picked up 2665's again and had the same issue of not starting the WU. I tried a third time on my "daily driver" machine with flushing the WU but d/l'd another 2665 and hung again. That just tried my patience too far..

So I killed the second VM on both Quads and setup WCG with a Notfred VM and GPU client on each one. All good so far. I kind of went back and forth with myself on whether to kill the 2nd VM on both machines since it limits the amount of work I'm contributing to F@H but honestly, if I have to constantly babysit the boxen and spend several hours trying to massage issues with Stanfords junk, am I really worse off? I don't think so. I'm now contributing some work to other worthwhile projects and still giving F@H some love. As far as I'm concerned, me not contributing 100% of my resources to F@H is their own damn fault. If I'm going to have my machines crunching 7x24 I want them working 7x24 not sitting there hung or crashing regularly.

Stay safe everyone in the path of "Gustav" (especially to our brother folder [XC] Synthetickiller from Xtremesystems.org):D

+1


 
I had the same problem a couple times in the past and it was indeed a borderline OC on the processor. Just a tad more Vcore seemed to fix that issue up most ricky-tick. I was also running all my Notfred VM's @ 512m of ram. I found that increasing them to 768 helped stability when picking up the 2665 WU's and didn't hurt with the 2662's as well. If you have enough ram, a meg per VM is even better but I haven't had an issue with the 768 yet.

I remember that few months ago there was big problem with WU which required almost 1GB RAM. Since then all my VMs have 1+GB (its not broblem for me because my PC has 8GB RAM :D

Stay safe everyone in the path of "Gustav" (especially to our brother folder [XC] Synthetickiller from Xtremesystems.org):D


+1

 
This may sound like a silly answer to your collective problem but why not install VMware server (free) and simply install a known distro of Linux?

You need to know NOTHING about Linux to make this work. Right next to the down load icon as a book looking icon. Please click it.

You will find not so much instructions but a line by line cut and paste right into a terminal window and boom, you have a working copy of folding. The only thing you might actually have to type on the command line is “sudo” then cut and paste each line of text until you are up and running.

Just down load a distro, leave it as an ISO and point the CD/DVD ROM (the one in VM) at the file, Linux reads an ISO file as well as windows reads text.

In the end it’s much harder to screw up, honest.;)




 
[CUT]
Headless/ diskless systems are great, untill you have to work on them ;)
[/CUT]

QFT, when my server was screwing up I had to lug out my old 19 inch crt and find a working keyboard to fix it, it sucked majorly.
 
It's not always consistently OC's that cause these errors. My Sig Rig was at Stock Clocks for a while when I was tinkering with notfred's and I still got EUE's. I ditched it for now and am running dual SMP clients native in Vista x64. When I have time again, I'll take a look at again.

 
im using his tftp solution and i love it for machines that cant be fully dedicated. they boot it run and back up to the tftp server every 5 minutes and then can be restarted when they actually need use and go back to whatever OS is needed.
 
I agree with BillR on this one. Installing Ubuntu or OpenSuse is a no brainer and using finstall to get folding is even easier. And on a dedicated folder, theres no need to even bother with VM's. With a real Linux install, you don't get very many of these "something went wrong but it wont tell me what" situations.


 
And on a dedicated folder, theres no need to even bother with VM's. With a real Linux install, you don't get very many of these "something went wrong but it wont tell me what" situations.

True enough.. but doesn't help you much in a dedicated boxen with GPU. Not YET anyway. In that case if you don't want to run the native Windows SMP client, VM is the only solution.


 
I agree with BillR on this one. Installing Ubuntu or OpenSuse is a no brainer and using finstall to get folding is even easier. And on a dedicated folder, theres no need to even bother with VM's. With a real Linux install, you don't get very many of these "something went wrong but it wont tell me what" situations.



I do so on several boxen. I want GPU suppor tunder linux, native..... nooooooooooowwwww
 
Well, the GPU2 client does run under WINE if you wish to dabble a bit with dark magic. However, I am a self professed Linux geek and thats a bit more than even I feel like trying to undertake right now. :eek:


216
 
I got home and both my notfreds vms were doing nothing. I migrated one VM to an Arch linux install, and the other one decided to start doing work while I was setting up the first one, so I let it do its thing. Next time it hangs its going to Arch as well. I have horrible luck with the smp clients for whatever reason. The GPU clients run all day without issue for me.
 
Back
Top