see ya!

[Mr.Guvernment]

well guys i have officially said good bye to folding... (just uninstall F@H on some rigs this week and replacing with boinc)

and i see today some others are getting rather tired as well!

this started some time ago, trying to get the SMP client to run properly on a quad core... which it never would, even after following suggestion after suggestion, step by step, going as far as to format a computer!!!

Standford needs to get their butt'ds in gear!


WE are doing THEM a favor running F@H is how i see it, and when i have to spend literally, hours, trying to fix an install that was working fine then decided to just die... sorry, but i have better things to do.


I went back over to Boinc - install an exe - DONE! walk away! finished!, well you can install the optimized client - but again, copy an exe - DONE walk away!

i have enough things in my life frustrating me that i don't need to come to work and need to check on 6 rigs to make sure F@H is running, can connect to get WU... et cetera et cetera...

If F@H ever gets their head on straight and makes F@H as easy as Boinc, which i dont see why they cant? i may come back, but until then.. i have more faith in finding alien life form.


I hit top 100, only for a few hours, but i hit 100th place, i had hoped to go further.. but honest with all the gpu client coming out and the screwing of the points system Standford keeps doing.. i couldn't see myself getting much further

I do understand folding is not for points, it is for the cause, but when their client cant even let you fold for the cause cause your client cant get any WU for days... or your client crashed.. or your client which was at some %90 done decided to corrupt the WU... you get the idea....

Until next time! Good luck everyone else in your folding ventures.

[SniperXX]

Sorry to see you go man. I'm really annoyed with the SMP install also. One of my quads works flawlessly but the system is down temp (not due to FAH) and another one crashes consistently. Its beta but man the one that crashes crashes maybe 10min after I boot it...

[Mr.Guvernment]

it stinks, really does, seems SMP is always going to be in beta, earlier version seemed to run flawlessly, as far as i recall, install them and they were fine, seems once they got into thie 6 series versions, all hell has gone loose!

If Boinc can do a simpe install, that just works. why cant F@H.. i understand they likely calculate differently, and they have some genius minds over there for sure.. why cant they make it work..lol

or even with the SMP, you install it, and then have to find some link to a forum that has a new EXE cause they haven't bothered to update the download with this new exe, so those running gpu can do so with out the fatal error stuff...

[OldSkool]

+1 Mr. G It started as something positive and was going great but they couldn't leave well enough alone and now it's just too much work and a complicated mess. I've moved to Seti as well and haven't had to do anything beyond install one program, sign in and be happy.

[ChasR]

FAH is far more cutting edge science than looking for green men on Boinc. You didn't have problems with the non beta client, did you? You went for big points with beta clients and couldn't get it to work. While you aren't the only one to have problems, the vast majority of folks who try FAH, do get it to work.

[CGR]

I have had no problems with the SMP client on my quad. Even if you cant get two running and only one instance, its still folding.

I fold for the cause though, I could care less about points.

[jonspd]

I fold for the cause though, I could care less about points.:beer:

[SniperXX]

.

I fold for the cause though, I could care less about points.

Same here, my reason being I'm a cancer survivor. :beer:

I'm now looking into getting in on the gpu side of folding (in addition to the regular non-beta cpu clients), seems like the cpus can do more complex stuff much quicker.

[benbaked]

If yourself or anyone else still wants to help medical research but doesn't want to put up with Stanford's clients, then I'd like to suggest rosetta@home. It's similar to folding@home, with more or less the same goals. It's not as fun as seti@home since there are no optimized clients for rosetta, but it is truly a "set it and forget it" system and you can run on a split-time if you still want to crunch seti@home. I'm running a 50/50 time split.

Plus, it's a great time to jump on board the O/C rosetta team, we're riding a wave of overtaking teams as we're currently outputting our highest RAC ever!

[Mr.Guvernment]

will check out rosetta, have seen its name mentioned here and there.

yes ChasR i did jump on the Beta Clients, to get the most out of my dual and quad core rigs, it is what every said to do, so i did it. and with the 5 versions they ran great! then they dropped support for them with little notice (at least i saw) so i upgrade the new 6 versions, that is where hell started.. :( :(

Now it seems the extended the expiration of the 5 clients - if they had of just left them.

but it just gets frustrating when you take the time, set everthing up, everything seems stable and good, and then you come into a rig and see either "no work to be done " " cant connect to get work " "fatal error" or some other problem and then spend all day trying to fix it, because simply reinstalling doesnt seem to do the trick.

i am sure plenty of people don't have problems once installed, but i am sure just as many people do have problems, especially those who run alot of clients on various machines.

[ChasR]

Guv,
HayesK and I run over one hundred clients in almost every conceivable configuration in three different locations. It'll be difficult to convince me that the client doesn't work. Do we have some problems? Of course, but nothing like you describe. Most clients are set and forget. Is FAH perfect, no. Does it require maintenance and vigilance? Yes, if you want the big points. Would I ever consider spending the $ it costs to fold, hoping to find the cause and perhaps the cure for some nasty diseases, to find alien life forms? No way in he**, but one thing I truly believe, to each his own.

[grunjee]

Hopefully you'll come back someday assuming they get the SMP client fixed. As others have said there's always GPU folding as well.

[Sorin]

Hopefully you'll come back someday assuming they get the SMP client fixed. As others have said there's always GPU folding as well.

How is GPU2? I tried regular GPU a little bit after it came out on my X1950XTX and decided to call it quits. The temperatures were out of hand, the manual setting of the 3D clocks, and the really weak points compared to the science being done.

The biggest thing though was how a minor instability could cause you to rack up thousands of EUEs in a day until you noticed something wasn't right with the folding.

I really like the points of GPU2, plus the amount of science being done, but have the myriad of other problems been solved? Manual clock setting, instabilities, potential for zillions of EUEs? I'd obviously be going Nvidia this time around (I never liked ATi and only got it for the folding).

[Mr.Guvernment]

ChasR if i had the amount of clients you had i think i would of died by now...LOL

as said i will certain come back to F@H once they get a nice stable SMP client out, i love the GPU client, i have had zero issues with it on my 4850, they seemed to manage to perfect that with out any problems.....

[Adak]

*Sorin, the GPU2 client is vastly improved over the first GPU client. Really, a home run for FAH's programmers.

The SMP client is the black sheep of the family of clients. Charles and HayesK get along with it OK, but let's be honest, they have a HUGE amount of experience setting up FAH clients, and I'm sure, they take a very professional approach as well.

Consider OSUmaxx's problem - He was signed into the administrators account with the right password, but was somehow invoking the command line *without* being in that account, with that password. (this is in Vista).

I didn't even know this was possible! :eek:

[Jolly-Swagman]

Guv ,, sorry to hear about your Leaving the Team , but as Stated above GPU2 client is allot better now and you could always just run them ,, Until you see that FAH becomes more stable and better

[ChasR]

I think one of the keys to success with FAH SMP may be avoiding Vista. HayesK and I have 1 Vista machine between us.

[harlam357]

Hate to hear of your troubles Guv... and even though I only have one WinSMP client running right now, it's just as stable as v5.91... drop in replacement really.

I'd be curious to know the ratio of folks having trouble on Vista vs. XP with the new v6 SMP betas. I've never thought it to be more of a Vista problem... but that indeed may be the case.

I can understand if things are just getting too frustrating... but please stick with us and Fold on your GPU and do some other projects using your CPU - at least until Stanford gets us a more stable SMP client. :)

[CJ-5]

I have Vista on a laptop running SMP. I never worry about it's reliability as it has given me no reason to. The only hitch was when the newest client came out and I had to turn off "User Account Control".

I will say, however, it is the most basic of installs. There is no vm ware or gpu folding in addition to the one SMP wu.

Linux, on the other hand, has to be checked every day to see if it hangs after finishing a wu. I used to leave them running when I went out of town but now with the newest version there is no point in that. I lose several hours of folding each day do to them hanging and requiring a restart.

[grunjee]

I think one of the keys to success with FAH SMP may be avoiding Vista.


I agree 100%. In my experience Vista is much more problematic than XP. To have any chance with SMP in Vista, you need to do a few things... disable UAC, do not put the FAH folder under Program Files etc. Even then I've had problems and almost gave up on it.

[Mr.Guvernment]

I think one of the keys to success with FAH SMP may be avoiding Vista. HayesK and I have 1 Vista machine between us.

4 of my rigs are server 2008 also, which ideally is the same as vista, i cant really avoid it since these are work rigs, my home rig is also vista and my work rig is vista as well :(

i will run the GPU client as i only have a4850 at home, but i am shutting my rig off during the day to get my power bill down, or try to (trying to see if my landlord is leeching power from my place for something), i am looking to grab an 8800GT for GPU2 to put into a server here that does have a spare pcie x16 slot.

[pik4chu]

I think one of the keys to success with FAH SMP may be avoiding Vista. HayesK and I have 1 Vista machine between us.

So very true. just to see what a new machine we got in the office could do I tried to setup FAH on it. gave up after an hour. stoopid vista -_- But I have it running fine on my xp pro and server 2003 standard/enterprise servers at home. What keeps getting me is the expiration of the client lol...

someone ask God to give us 36 hour days so I have time to do everything >.>

[Mr.Guvernment]

then i guide my frustraion also at Vista for what ever reason it does not like F@H and at Standford for what ever reason they cant get SMP stable on Vista :)

[ChasR]

There are beta Linux clients out or about to come out that eliminate the hang on completion problem. Removing the -forceasm flag greatly reduces the problem on the 6.02 client for some unknown and unexplainable reason.

[Hack30]

I understand your frustration. Maybe when they get it sorted you'll try again?

TTFN:attn:

[Adak]

someone ask God to give us 36 hour days so I have time to do everything >.>

I keep asking Him, He keeps ignoring me. Says something about not having enough hours in the day to get to my request. :p :p :p

[silent bob]

yeah as my client get borked , they stop folding and I dont restart them , well I do til they bork again then off , got a q6600 doing nothing at all not a ddam thing , 8800 gs , doing nothing .
I am down to 2 gpu and 1 smp, and when those get borked , they will be gone too

[harlam357]

I always say... "don't blame FAH just because your machine isn't stable". ;) :p

What do you mean by borked bob? Definition please... maybe we can help you... or did the FAH clients just spontaneously combust? :D In which case I would say that it's borked. ;)

[Audioaficionado]

I think one of the keys to success with FAH SMP may be avoiding Vista. HayesK and I have 1 Vista machine between us.
:shrug: Vista works alright for me. The only thing that still peeves me is that 16+ GB memory lock when I first start a VMWare Server VM. The second one opens just fine without that issue.


I keep asking Him, He keeps ignoring me. Says something about not having enough hours in the day to get to my request. :p :p :p
Hmmm... He always has time for me and often initiates our conversations.

[silent bob]

I always say... "don't blame FAH just because your machine isn't stable". ;) :p

Bone stock NO overclock wont fold . My machines are stable , hell they are stock and FaH still doesnt like it . Seti doesnt have a problem , they are stable enuff for that .

I dont remember exactly when the problems started , its been a while , but IO Errors and Unstable Machine Errors were the norm .When I wanted max ppd it was constant maintenence , Time is becoming a valued asset to me , and to spend an evening fixing a client , at one time used to be fun , its not anymore .

I just spent 750 bux on a few parts for another FaH rig . I will set it up with FRESH everything , even new never used OS and hd , we will see what happens . oh a P5E and an 8400 Wolf with 8800ssc and 4 gig mushkin with modular psu .

[Adak]

Hmmm... He always has time for me and often initiates our conversations.

That's ME initiating the conversations, right after the orderly has given you your meds, silly! :clap: :clap:

[Shiggity]

I agree with Guv, sure you CAN get SMP setup with some effort. Most people want an effortless installation though, and the easy solution there is BOINC.

I know Stanford wants to do cutting edge research and that is great, but you can't let your clients get as complicated as they have gotten. Yeah sure it's great for the tiny amount of people that are dedicated enough to go through the hassle of setting them up, but overall I think it's hurting them.

I'm sure in the future they'll have an all purpose GPU2 / SMP client that can do anything you want and is easy to setup. When that time comes, I'll probably come back to F@H full time.

[Mr.Guvernment]

For sure, if they get a stable SMP client going, i am back into F@H to claim back a top 100 spot, but like others, i just dont have the time to mess with a client if it gets borked, i have left some of the older single clients running on 2 or 3 systems right now, so i havent stopped %100.

[WarriorII]

well guys i have officially said good bye to folding... (just uninstall F@H on some rigs this week and replacing with boinc)

and i see today some others are getting rather tired as well!

this started some time ago, trying to get the SMP client to run properly on a quad core... which it never would, even after following suggestion after suggestion, step by step, going as far as to format a computer!!!

Standford needs to get their butt'ds in gear!


WE are doing THEM a favor running F@H is how i see it, and when i have to spend literally, hours, trying to fix an install that was working fine then decided to just die... sorry, but i have better things to do.


I went back over to Boinc - install an exe - DONE! walk away! finished!, well you can install the optimized client - but again, copy an exe - DONE walk away!

i have enough things in my life frustrating me that i don't need to come to work and need to check on 6 rigs to make sure F@H is running, can connect to get WU... et cetera et cetera...

If F@H ever gets their head on straight and makes F@H as easy as Boinc, which i dont see why they cant? i may come back, but until then.. i have more faith in finding alien life form.


I hit top 100, only for a few hours, but i hit 100th place, i had hoped to go further.. but honest with all the gpu client coming out and the screwing of the points system Standford keeps doing.. i couldn't see myself getting much further

I do understand folding is not for points, it is for the cause, but when their client cant even let you fold for the cause cause your client cant get any WU for days... or your client crashed.. or your client which was at some %90 done decided to corrupt the WU... you get the idea....

Until next time! Good luck everyone else in your folding ventures.





I'm speechless..... :(

[Audioaficionado]

That's ME initiating the conversations, right after the orderly has given you your meds, silly! :clap: :clap:
Were you around 7-21-1976? ;)

Back on topic: I'm starting to get very irritated with nix SMP_FAH too.
http://www.ocforums.com/showthread.php?p=5809679#post5809679

[Adak]

Were you around 7-21-1976? ;)

I had graduated before 7-21-1976! :)

[harlam357]

Bone stock NO overclock wont fold . My machines are stable , hell they are stock and FaH still doesnt like it . Seti doesnt have a problem , they are stable enuff for that .

I dont remember exactly when the problems started , its been a while , but IO Errors and Unstable Machine Errors were the norm .When I wanted max ppd it was constant maintenence , Time is becoming a valued asset to me , and to spend an evening fixing a client , at one time used to be fun , its not anymore .

I just spent 750 bux on a few parts for another FaH rig . I will set it up with FRESH everything , even new never used OS and hd , we will see what happens . oh a P5E and an 8400 Wolf with 8800ssc and 4 gig mushkin with modular psu .

I have no clue if seti is more or less stressful than FAH, but I'm going to guess that it's less stressful. Also, just because a rig is set stock doesn't mean that it's stable... likely that it is, but there's no way to tell unless one runs some extended tourture test other than FAH.

I feel for ya bob... I really do... I wish I had some kind of answer for you. :(

Good luck with the new build... hopefully it will turn out to be a winner with FAH. :)

[visionco]

I have no clue if seti is more or less stressful than FAH, but I'm going to guess that it's less stressful. Also, just because a rig is set stock doesn't mean that it's stable... likely that it is, but there's no way to tell unless one runs some extended tourture test other than FAH.

I feel for ya bob... I really do... I wish I had some kind of answer for you. :(

Good luck with the new build... hopefully it will turn out to be a winner with FAH. :)

I am having the same problem on New IBM Workstations running a C2D 2.0 stock. Ever since the new client they will fold fine for a couple of days and then EUE out so I restart the client and they will fold for a couple of more days. This is starting to get old for me since I have 5 of them on top of all the other rigs.

[Mr.Guvernment]

I'm speechless..... :(


I know m8, we had a good run going on! neck and neck, so .. for now... i will let you control page 1 ;)

[Audioaficionado]

I had graduated before 7-21-1976! :)So that makes you an old fart too :p

[WarriorII]

I know m8, we had a good run going on! neck and neck, so .. for now... i will let you control page 1 ;)


I will guard the Gates at the 100th spot.

Best Wishes,

You know where to find us. :rolleyes:

:attn:

[Adak]

So that makes you an old fart too :p

You have to be old as the hills to make your buds think you're god, when they're on their meds! :clap:

[dark bishop]

ill keep smp folding, most points ive gotten other than gpu. took me 3 minutes to setup again.

[Mr.Guvernment]

Well not all hope is lost!

i just installed the GPU client on my 4850 again, but it is taking up %100 of one core on my q6600 @ 3ghz.. seems a little high no?

also just set up SMP under my ubuntu VM install, but only has access to 2 cores... so see how it does... (taking a screen shot on ubuntu is nice also, so easy!)

[ihrsetrdr]

If you're greedy you can set up a second vm running a SMP client in Linux- in Task Manager set priority=low for each vmware-vmx.exe process & uncheck the CPU3 core's box for affinity for both vmware-vmx.exe processes so you don't starve the GPU client.

[WarriorII]

Yeah,,,,,,, :)

[harlam357]

Rock on Guv! :attn: Glad to see you back in action! :)

Not sure about the cpu usage on the ATI card... are you running XP as the host OS? If so, such usage is normal on the nVidia side, so I wouldn't be surprised to see the same for ATI.

[Mr.Guvernment]

using vista...


so far i am %50 done on a 25000 unit on the SMP using 2 cores... it should finish in time no?

[AlabamaCajun]

I need to figure out whats borked here. Since the last shutdown for RL I've been forced to abandon the old SMP and "ALL" folding rigs are blowing chunks. I'll read through all the threads and see whats up when I get some time.

[drshivas]

I need to figure out whats borked here. Since the last shutdown for RL I've been forced to abandon the old SMP and "ALL" folding rigs are blowing chunks. I'll read through all the threads and see whats up when I get some time.

Are these VMs? If so, allocate more than 800MB + for each VM. The new WUs are memory hogs. This has been the advice for what, a few months now? Not sure if it's changed.

Interestingly though, thanks to a Ballistix failure the past weekend, I was forced to run two boxes with 1GB with VMs @768MB, but they haven't crashed yet.

[SF101]

i wont be checking the thread back on responses for this. but i too went hardcore (as hardcore as i could ) and setup 7-8 pc's Dual / quads a-couple single core pc's and 2x PS3's for FAH.

As Mr G stated clients would stop working ect.. after weeks of this randomly happening on pc's that i did not frequently monitor i just ended up uninstalling the clients off every machine i had.

that and having my ps3's stop responding and not auto folding after i stopped using them kind of pushed me off it entirely.

If F@H come out with something like bonic where i could just install it .. setup the optimizer for my processor .. minimize it and never have to look at it other than to track my pc's last uploads / points contributions to make sure they are all up.. I would most likely start using them again. but for now I'm sorry i just don't have the time or the access to be running around every morning figuring out why 4 machines stopped uploading just to see the client stuck and not recovering.

Also being a Tech i really don't want to have to trouble shoot clients ect at the end of my day. i just want it to work my pc's are there to do the work if the right client is around to do it hassle free. my2cents

[Mr.Guvernment]

well hit almost 4k for one day, so getting back up there already with just one SMP and one GPU client going... not too shanny, maybe i will hit top 100 again :D


^^^ i felt your pain m8! i cant wait until the day stanford makes it easy and stable!

[Mr.Guvernment]

Ubuntu VM's set up, 2 of them on my Q6600 @ 3.35Ghzan d running good, taking about 15mins to complete a piece on each one..

and a 9650 coming to replace my Q6600 and my Q6600 going into my HTPC which may allow for some folding now and then, but trying to keep my hydro bill down!

Top 100, i am coming back for you!! OH YAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA

[eddyg519]

hell i should move over to linux for folding im lucky to get 2500 points from my 9800GTX+ AND Q8200 LOLz

[klear]

hell i should move over to linux for folding im lucky to get 2500 points from my 9800GTX+ AND Q8200 LOLz

9800GTX+ should get 5k alone with GPU2 client!

Mr.Guvernment, you've got some catching up to do :burn:

[Mr.Guvernment]

i know! i let myself fall behind too much :(, oh well once my Q9650 is here and nicely clocked i will get my 4850 also folding ;)

[science man]

MR.GOVERNMENT I feel ur pain man believe me I know what ya mean. The programmers over there are such rookies. It's like you to have skill to get the clients going. I'll let ya know if they ever actually decide to use their heads and make the reasonably easy. Peace for now. :)

[AlabamaCajun]

Are these VMs? If so, allocate more than 800MB + for each VM. The new WUs are memory hogs. This has been the advice for what, a few months now? Not sure if it's changed.

Interestingly though, thanks to a Ballistix failure the past weekend, I was forced to run two boxes with 1GB with VMs @768MB, but they haven't crashed yet.

Here it is, lost track of this thread (bump).
Now, I'm still old scule with rigs and towers just running the classic SMP FAH.
These things ran fine for several weeks then just chucked again all about the same time like a bad batch that they all get sick and just stop. Happens on different machines with 1G to 4G and NO VM. The unit completes, says its finalizing I think then dies with the error has occured. Restarting it, or attempting to send has no effect. I have to delete the queue and all that precious work and start over.

[Adak]

Here it is, lost track of this thread (bump).
Now, I'm still old scule with rigs and towers just running the classic SMP FAH.
These things ran fine for several weeks then just chucked again all about the same time like a bad batch that they all get sick and just stop. Happens on different machines with 1G to 4G and NO VM. The unit completes, says its finalizing I think then dies with the error has occured. Restarting it, or attempting to send has no effect. I have to delete the queue and all that precious work and start over.

No, no!

Don't restart them if they die! That will just wind up thrashing the queue. Instead, use the qfix utility. It will give you some bogus message about nothing being wrong with the queue.

Edit: I wrote this while in the midst of some programming, and gently "slaughtered" the procedure. :p ChasR's has the procedure correctly listed below.

Now restart your client, and it will "see" the WU waiting to be sent, and send that baby on in to "Mama" Stanford! :)

It's not 100% a sure thing, but it has worked 100% of the time when I've used this little trick. If you get the latest version of the client, you'll have fewer of these "die on the vine" incidents.

[ChasR]

To send a Linux SMP WU hung at completion is slightly more complicated than Adak's post, at least the 100% "official" method is. :D
Stop the client. Run qfix in the FAH directory. It should find a full queue slot and say the file is OK but it really isn't. If, perchance, qfix were to report it had fixed the file and requeued the WU for upload, you could go ahead and start FAH. This has never happened for me. Run fah6 with the delete flag and the 2 digit queue # (leading zero): ./fah6 -delete 0X.

The client will report after several minutes that it wasn't able to delete the WU and close. Run qfix again and it will report the file fixed and the WU requeued for upload.

Restart FAH and your WU will be sent.

In my experience, p5101 and p5102 are most likely to hang followed by p2665. If you have one of those you need to watch to be sure it gets sent in. Nothing worse than missing the deadline due to a days inattention.

[Adak]

Coincidentally, I just was hit with three bad boy WU's that all folded to 100%, and then wouldn't send! Whatta *****!

This is for Ubuntu Linux, but the example works almost the same for a Windows box.

I d/l'ed qfix for Debian (same kernel as Ubuntu), from:
http://linuxminded.xs4all.nl/?target=so ... s.plc#qfix

then put it in the same folder as the offending FAH client. (so nice to just drag icons)

And stopped the stuck client, if it hadn't stopped itself already: Crtl+C, of course.

Had to click on the the qfix icons and select "properties" and check the "allow this file to execute" box, and close it. ( You can also enter:
chmod +x qfix in the terminal window, to allow qfix to execute., which is faster)

entered ./qfix in the terminal window and got this output:

adak@adak-Rocketfish:~/fah1/FAH6.02-Linux$ ./qfix
entry 4, status 0, address 171.108.34:8080
entry 5, status 0, address 171.108.34:8080
entry 6, status 0, address 171.108.34:8080
entry 7, status 0, address 171.65.36:8080
entry 8, status 0, address 171.65.36:8080
entry 9, status 0, address 171.108.34:8080
entry 0, status 0, address 171.65.36:8080
entry 1, status 0, address 171.65.36:8080
entry 2, status 0, address 171.108.34:8080
entry 3, status 1, address 171.108.34:8080
Found results <work/wuresults_03.dat>: proj 2671, run 28, clone 18, gen 76
-- queue entry: proj 2671, run 28, clone 18, gen 76
-- queue entry isn't empty
File is OK

Now we get FAH to delete it - almost! :)

adak@adak-Rocketfish:~/fah3/FAH6.02-Linux$ ./fah6 -smp -verbosity 9 -delete 3

And in a minute or two FAH says it can't delete the WU.

Now restart the FAH client, and the bad unit will be found, and sent if done.

[17:17:16] Connecting to http://assign.stanford.edu:8080/
[17:17:19] Posted data.


And it worked for all three of these rascal WU's! :beer: :clap: :clap: :clap:

[AlabamaCajun]

No, no!

Don't restart them if they die! That will just wind up thrashing the queue. Instead, use the qfix utility. It will give you some bogus message about nothing being wrong with the queue.

Edit: I wrote this while in the midst of some programming, and gently "slaughtered" the procedure. :p ChasR's has the procedure correctly listed below.

Now restart your client, and it will "see" the WU waiting to be sent, and send that baby on in to "Mama" Stanford! :)

It's not 100% a sure thing, but it has worked 100% of the time when I've used this little trick. If you get the latest version of the client, you'll have fewer of these "die on the vine" incidents.

Dang [sound=traincrash.mp3] wish I had been paying more attention as I've dumped a dozen or so in the last month. Oh wait, I see a sleeping layer (er rather quiet in here), off to go give this a shot.

[Mr.Guvernment]

well, got my Q9650 in my main rig now @ 4Ghz and just got my Q6600 down into my HTPC< just got my first VMset up, working on 2nd one


Scary part is, i dont even check the forum threads now for install the SMP under linux, i can do it by heart..lol


so i will not have a

Q9650 @ 4Ghz running 2 VM SMP's
Q6600 (only @ 2.7 right now) running 2 VM SMP's but not 24/7.

:) this linux stuff is funnnnnnnnnnnn

[Mr.Guvernment]

well popping out from 7-9K a day now! but going to see how much my powerbill goes up this month.. if alot i wont be runnng my htpc 24/7

[the_cultie]

Glad to see your still with us :)

[Mr.Guvernment]

yuppers, that i am, after going the VM+SMP route it has been alot easier with far less problems then windows +smp.

[Surferseth]

I have no problems with Vista, and find it better to fold with than XP. Main rig is Vista 64bit. I have UAC off and Windows Firewall off. I can run a SMP client and multiple GPU2 clients with out having to set CPU priority and dont notice any PPD hit on the GPU2 or SMP clients.
On my wife's rig (Vista 32bit), I have UAC on and the windows firewall off. I have no problems with that machine either.

I did run into some problems running the SMP client in Win 7, but I have not killed UAC yet. If I cant get my head around it I will post up the error text.

[Mr.Guvernment]

under vista 64 the SMP client was always crapping out, corrupting and not working for me, that was why i was leaving, UAC off and firewall as well, seems for some it worked okay, others it is a nightmare.

[Zerix01]

:) this linux stuff is funnnnnnnnnnnn

I've been folding for many years now but always in Linux. I tried setting up SMP on my g/f's new XP based computer and spent an hour swearing at it and gave up. It's just running two uni clients and one GPU. Probably for the better since she won't run any of it 24/7 so it might not make the SMP deadlines anyway.

For a long time I didn't know what all the fuss was about with the Windows client.

[SF101]

yes i pulled off my farm because it frustrated me so much( 3 core2 Duals a ps3, 3xP4's a machine with a 8800GTS and my main rig) because i hate babysitting programs.

[Edward2]

I've only used the VM/Ubuntu SMP client and the MPICH SMP client, but I have not had any problems with either one. It's time consuming to set these clients up, but the PPD production is worth the effort.

I almost forgot. I've also used the not-freds USB flash drive client and it works fine also.

[Mr.Guvernment]

maybe some hope.. http://www.overclockers.com/forums/showthread.php?p=6377758#post6377758