I'm not getting points for finished WU's

[Weedo]

I never received the points for 3 WU's completed in the last 3 to 4 days. Is this happening to anyone else? I see the points being updated for others but not mine. I've got 216 points not credited.

[profdlp]

I haven't noticed the problem myself, but some of my rigs are 400 miles away so it is hard to tell.

Sometimes Stanford has a server glitch and there is a delay in points being posted. Usually they will come rolling in after a day or two.

[KingFish]

Don't feel bad Weedo, I believe they owe me 2,000 points for today. :(

[profdlp]

Don't feel bad Weedo...?

Now I feel horrible. :bawling: ;D

2,000 points in one day is but a dream... :respect: :fold:

[Weedo]

Don't feel bad Weedo, I believe they owe me 2,000 points for today. :(

I've been waiting days for the points. I don't think they're coming. I've received points for another subsequent WU.

[KingFish]

Well that's certainly not a good sign. My production may simply be down for the weekend although this weekend it appears to be down more than any other weekend I can remember.

Don't worry prof, a lab or office will fall into your lap one day. Make it your new years resolution or just ask santa.

[bothered]

Have you checked the server status page? Are they all ok?

[Weedo]

Have you checked the server status page? Are they all ok?

Most seem to be working. Seems to me they should have gotten through.

I just got credited for the WU I finished last night and still no sign of the missing points. Geez... what happened?! :scratch:

[pythagoras]

Post the protein and run clone gen numbers and I'll find out for you.

Regards

John.

[Weedo]

Post the protein and run clone gen numbers and I'll find out for you.

Regards

John.

This is what I lost:

P591, P957 and a P587, that's 254 points actually.

Please explain 'running the clone gen numbers'. I don't know what you mean by that.

[pythagoras]

In the log for those wus there should be 3 numbers at the side like these:

[10:36:21] Protein: p694_L939_wt_ext
[10:36:21] - Run: 6 (Clone 18, Gen 25)

regards

John.

[Lord_Night]

Weedo needs the points ;D
but like him I think everyone is missing points.
I would have to check mine not sure which one or if any are missing..
but I know Weedo doesnt want me to have anymore points.... ;D

The whistles blowing.... are you on board

[profdlp]

I haven't noticed the problem myself...
I have now. I went from a 700 point/day average to peanuts for the last 48 hours.

Stanford does seem to have a couple of servers down, though I can't tell for sure if they're the ones taking in the WU's and doling out the points.

[Weedo]

The whistles blowing.... are you on board

No... my foot is caught in the tracks. :eek:

[profdlp]

For what it's worth, my every-few-hour milestone notifications have also dropped off to almost nothing.

The WU's that have appeared are all dinky ones, too. It could be that Stanford is just sending out certain (small) WU's right now. They often change the type of WU being offered when their research leads them to believe that they are hot on the trail of something, so this may be very good news for the project as a whole. :)

That wouldn't explain why completed WU's are not showing up in the point totals, though.

[Weedo]

Here's what happened to the the P957. It happened in the middle of the night. When I saw another P957 in progress I assumed the other finished and a new one began. Why it happened I can't say. The machine didn't shutdown or reboot. The client just shutdown. Maybe it's a Win98 thing.

I couldn't find the log for the others and I don't think the same thing happened in those cases. Different machine.

[00:26:38] Completed 4100000 out of 5000000 steps (82)
[00:35:45] Writing local files
[00:35:45] Completed 4150000 out of 5000000 steps (83)
[00:44:46] Writing local files
[00:44:46] Completed 4200000 out of 5000000 steps (84)

Folding@home Client Shutdown.


--- Opening Log file [December 13 03:20:36]


# Windows Graphical Edition ###################################################
###############################################################################

Folding@home Client Version 4.00

http://folding.stanford.edu

###############################################################################
###############################################################################



[03:20:36] - Ask before connecting: No
[03:20:36] - User name: Weedo (Team 93)
[03:20:36] - User ID = 4C1ED7156340A09B
[03:20:36] - Machine ID: 1
[03:20:36]
[03:20:37] Loaded queue successfully.
[03:20:37] Initialization complete
[03:20:37] + Benchmarking ...
[03:20:43]
[03:20:43] + Processing work unit
[03:20:44] Core required: FahCore_78.exe
[03:20:44] Core found.
[03:20:44] Working on Unit 04 [December 13 03:20:44]
[03:20:44] + Working ...
[03:20:47]
[03:20:47] *------------------------------*
[03:20:47] Folding@Home Gromacs Core
[03:20:47] Version 1.70 (October 24, 2004)
[03:20:47]
[03:20:47] Preparing to commence simulation
[03:20:47] - Ensuring status. Please wait.
[03:21:04] - Looking at optimizations...
[03:21:04] - Working with standard loops on this execution.
[03:21:04] - Previous termination of core was improper.
[03:21:04] - Files status OK
[03:21:04] - Expanded 51971 -> 272981 (decompressed 525.2 percent)
[03:21:05] - Checksums don't match (work/wudata_04.xtc)
[03:21:05] - Starting from initial work packet
[03:21:05]
[03:21:05] Project: 957 (Run 94, Clone 51, Gen 18)
[03:21:05]
[03:21:05] Entering M.D.
[03:21:11] Protein: p957_pa_drf2
[03:21:11]
[03:21:11] Writing local files
[03:21:11] Using table 3
[03:21:13] Writing local files
[03:21:13] Completed 0 out of 5000000 steps (0)
[03:44:14] Writing local files
[03:44:14] Completed 50000 out of 5000000 steps (1)
[04:01:39] Writing local files
[04:01:39] Completed 100000 out of 5000000 steps (2)
[04:16:46] Writing local files
[04:16:46] Completed 150000 out of 5000000 steps (3)
[04:31:53] Writing local files

[Straight_Man]

Weedo, there is a column called WU RCVD on the server stats page. There are a bunch of WUs on acceptance\data servers that have not been transferred to stats. They total to over a thousand WUs waiting to hit stats. There is a note by the WU RCVD info down at the bottom of the page that says if there is a problem and a lot have gathered they will have to be credited maunally. That, unfortunately, means Monday or Tuesday (ever noticed there is a points spike on Monday or Tuesday for this team fairly regularly???). SO, it is quite possible that you have a WU on server .148 and server .153. What happened to project 591 I have no idea, it is not being fed right now as a regular WU (COULD be a late beta WU, this has been known to happen). BUT, one of the data servers is also down.

The only other thing I can say is this: IF a WU computation does an Early_End you get partial credit for the completed part proportional to the total points for 100% complete (so if a WU Early_Ended at 67%, you would get .67 of total points for that WU). You might check your FAHlog-Prev.txt file and see if any of those WU calcs did exit with an early end and at the same time you might check the gen and runs for your WUs you are missing credit for.

As of 15 minutes ago, there are over 1500 WUs received but not applied to stats at a very QUICK look.

Below is a link to the server summary page for you, if you need one:

http://vspx27.stanford.edu/serverstat.html

[Straight_Man]

Concerning your log, can you run Client 5.03 or 5.02 on that computer??? BOTH are stable here, on XP and Linux, and do not have issues here that Client 4.0 has had. Client 4.00 had some issues with some WUs, and Client versions 5.02 and 5.03 seem to have fewer issues with same WUs and Gro Core that Client 4.0 had problems with. Also, for some reason I am running Gromacs Core 1.68. I will see if a newer core is out offically by triggering a redownload of Gro Core and thanks for showing me that, you alerted me to something which I need to check out. BUT, the latest Gro Cores like the latest client better on XP and Linux here than Client 4.0 did.

[Lord_Night]

No... my foot is caught in the tracks. :eek:


Must be why I am sitting in place now and not getting points
am waiting for you to get your foot unstuck......


;D

[Weedo]

Straight_Man, I tried v5 on my Win98 machines but it wouldn't install. I'll check that stats page link when I get the chance though. Thanks for that heads up. I currently run GUI v4 on the 2 98's and one of my XP machines. On the other XP machine I run Console v5 as a service with Electron Microscope (the only way to go). When I do have problems it's almost always with the 98 machines. Anytime there is a power blip they restart their WU's from the beginning. I've been burned like that many times on big Gromacs and it's always very costly. Also unexplainable restarts and hangups and stuff. I can't get away from 98 on those 2 machines for reasons so I guess I'll have to accept those occasional losses. Just when I get really going good is when it happens.

Lord_Knight... maybe you're caught in a Gromac storm. I hate it when that happens to me. Puts my production in the dumper. I recently spent a whole week on a p1412 and around 5 days on a P220 Murea. Man that was killer. You gained a lot of ground on me then. That's what I mean by 'foot caught in the tracks'. Seem to be getting back on a roll now though.

[profdlp]

Weedo, I don't know of any free ones off the top of my head, but if you could find some automated backup software to backup your Folding folder several times a day maybe you could at least limit your losses to a few hours work, instead of perhaps days.

About a year ago I had a computer that went about a month rarely finishing a WU. I'd usually get partial points, but it was a lucky day to actually complete a full WU. I tried everything I knew (stopped o/c'ing, reinstalled numerous times, fiddled with all sorts of new drivers, etc.), then the problem seemed to go away as mysteriously as it came. Hope you get it sorted out. :)

[Weedo]

Thanks Prof. I'll look into that. I could live with a few hours.

Mysteriously going away would be even better. :thumbsup: