-Bigadv problems again

[T4rd]

For the last couple weeks, my Linux VM hasn't been able to pull WUs. This happened a month or so ago too, so I used a new image from the bigadv guide in the sticky up top and that worked for a few weeks, it seemed. So I tried to re-download a new image from the same guide/link (maybe it's been updated?) and it receives WUs now, but before it finishes its first step, in the VM is gives me an error to something of the effect of "ran out of memory" or some kind of memory error that made me think that (sorry, I'm at work now and can't post the log, I'll try again later and paste it if need be).

I had originally forgotten to change the memory allocation to 5 GBs (what I had it set on before with no probs), so I changed it from 4600 MBs to 5100 MBs; same error. Then I maxed it out to 5500 MBs, still get a memory error. I haven't changed my OC settings or anything, I know I'm stable. So is there something I'm doing wrong now?

I haven't been browsing the DC forums as much as I used to a few months ago, so I've kind of been out of the loop for new distros and clients. But I think I faintly remember reading about a Notfreds release of -Bigadv. If this is exists, I think I would much prefer it to messing with this Linux clients configuration and random problems. Notfreds was way too easy to configure and maintain.

So can someone point me in the right direction to getting my extra ~20k PPD back so I can finally get in the top 100?

 

[APOLLO]

I don't know what the issue is but it cannot be memory allocation. I have two clients running only 4GB each in native Linux since October. I have noticed for some reason that clients in VMs need more memory but I have run them successfully with a tad under 5GB. Right now, my VM clients have 5400MB and less. I am suspecting something else to be the cause.

 

[nitrobass24]

What are your OC settings?

 

[T4rd]

What are your OC settings?

Off the top of my head:

180 BLCK x 20 Multi
vCore @ 1.28 (forgot last two digits).

I've been folding with this OC since October with no issues until it stopped pulling WUs.

 

[nitrobass24]

are you running -smp 7 or -smp 8?
I was reading on FF that SMP 7 does not work anymore. that you must have 8 cores to pull bigadvs.

 

[T4rd]

I've always ran -smp 8 and just set priority (with the "priority grabbed" or whatever the command is in the .vmx file) to idle.

I'm just about to leave work, I'll try to run it again when I get home and post the log error.

 

[Cypher-]

are you running -smp 7 or -smp 8?
I was reading on FF that SMP 7 does not work anymore. that you must have 8 cores to pull bigadvs.

You just scared the crap outta me. I just went and looked it up and to clarify this is what is says:

Have a i7 860 folding bigadv units using -bigadv -smp 7 and getting 48,000+ bonus from it.
Have a i7 950 folding bigadv units using -bigadv -smp 7 and getting 52,000+ bonus from it.

So think the VMX file needs to report 8 cores available, but you can downsize to -smp 7 in the actual folding.

Tks
Rick

 

[T4rd]

Of course when I get home and run it so I can post the log, it doesn't error out .

It's only done one step so far though. I'll post back in the morn to update you all on how it goes. I got a funny story and a protip for you all though from just a little bit ago;

I am RMAing my OCZ Vertex drive because of this (Genmay access required). So I formatted it and went to pull the drive out WHILE my PC was running (FYI: Depending on your mobo/HDD controller, SATA is hot swappable. I've did it before with no probs) and somehow I determine my Intel X-25, which is my OS drive, is the Vertex and pulled the cable - INSTANT BSOD (as expected) lol. So I almost shit my pants and punch my face, then reconnect the drive. Amazingly enough, it stilled booted and I was back on my desktop in its usual 10-15 second boot time . No probs whatsoever! I was folding on both cards and the VM at the moment too! I had to delete my VMs folder and reconfigure the client because it said it was "still in use" when I went to boot it again and it wouldn't let me take ownership of it. No biggie though since I had just started that WU.

I just thought that was amazing that Win7 could recover so easily from that. I KNOW if that had been a pre XP OS (prolly XP too though), I would have been F'd in the A , hah.

 

[T4rd]

Well I checked on it again this morning and it had got 18 steps since I started last night. So hopefully all is good now, heh. I dunno WTF was going on and I hate it when shit magically fixes itself because when it happens again, idk what the hell is going on, heh. Oh well, thanks anyways, guys.

One of my questions still stands though; are these Linux clients the only methods to folding bidadv WUs still? I it would be really nice if there was just a Notfreds VM I could run with -bigadv so there's no PEBKAC issues, hah.

 

[nitrobass24]

there is a prebuilt linux VM for bigadv.

 

[APOLLO]

One of my questions still stands though; are these Linux clients the only methods to folding bidadv WUs still? I it would be really nice if there was just a Notfreds VM I could run with -bigadv so there's no PEBKAC issues, hah.

There's no Windows -bigadv support yet, but there is a Notfreds -bigadv VM as nitro pointed out. I never tried it and cannot comment how good it is. Are you using the EVGA image now?

 

[T4rd]

There's no Windows -bigadv support yet, but there is a Notfreds -bigadv VM as nitro pointed out. I never tried it and cannot comment how good it is. Are you using the EVGA image now?

I was before when it stopped receiving WUs. Now I'm using whatever image is linked on the bigadv guide in the sticky. Its a direct download link and I didn't check the URL. I would now if I wasn't on my phone.

Thanks for the notfreds info though. I'll prolly try it out if I have another problem with this image. But I won't be home for a few days (on a business trip right now) to check on it. I'm sure if it got through the night without an error though, I should be straight for a while.