Folding Forum

I hope this isn't a sign of trouble on my end, but I thought all was aok fine with
my PPC Mac Dual Core G5 that has been folding PPC Mac WU's (Gromacs) for years and years
but tonight, for some reason, I am missing in action with 2 WU's completed yesterday
1/25/09, one at 4:41PM PST and another at about 8:30PM PST.

Here it is going on two updates later, and neither of those WU's has been scored
at the FAH Main score keeping system as yet. Weird!?

I checked out the server, and it appears to be healthy, 'Accepting' and not in Reject
status, nothing unusual in its logs that I could point out, but what do I know?

All I can say is that those 2 WU's are not yet scored, and it's very unusual for the PPC
Mac (only one I've got doing Folding, the rest are Intel Macs) to not get scored quickly
and efficiently on that particular Server....?

I honestly have no clue as to what is going on-- if anything at all-- just thought to bring
this up before it gets into another reporting cycle and nothing happens again.

I have double and triple checked my PPC Mac, and Increase 2.4.2 which is maintaining
those twin PPC WU clients, and all seems to be in order, my Prefs are set correctly, my
name is there for the uploader to catch the numbers, Team name is correct, etc etc...

Could somebody take a gander at that Server and see if all is well? It would
be most appreciated, as it's very unusual for that particular Mac to have any
issues at all...it's not that type of situation, just a low-key PPC Mac that is continuously
doing its thing, with 41 consecutive reported WU's on the current run going down smoothly
getting reported correctly, no problems at all...until this little session we have going on now.

Again, the server *LOOKS FINE*, but what do I know? It's not reporting finished WU's for some
reason from my lone PPC Mac doing the Gromacs, 3044, 3045 guys, like that sort of WU, and I
simply find it strange that it's not showing the WU's at FAH Main scoring system at this point, at
least not since 4:41PM PST yesterday, 1/25/09.

I won't have completed WU's to report about for some 33 and 43 hours respectively on the twin
CPU's in that machine, so I figured to bring this up now just in case something weird is going on
with that Server that I can't see, something not obvious to me in its stats...

Kindest regards,
rexrzer

Log file please ? ©

Server status pages indicates that 171.64.65.58 is doing fine ...

That was one of the finished GROMACs at about 8PM last night, still not tallied today.

Here's the 2nd one earlier yesterday at about 4PM PST:

Those are the two that are unnaccounted for thus far...I don't have any other PPC Macs folding
these days to show, but that Server is NOT reporting to FAH Main scoring center, apparently.

Kindest regards,
rexrzer

The displayed WU references are wrong ... that might be the issue that cause the work to no be credited.

Mr. toTOW wrote:

The displayed WU references are wrong ... that might be the issue that cause the work to no be credited.

Mr. Moderator, I just read what you posted...and no comprendo!?

Is this something that is wrong on my end? Is this a FAH issue, a Stanford issue, a problem with that Server, or something else altogether?

Please explain what you wrote because for the life of me I cannot figure this out, nor understand what you wrote about the "WU references" being "wrong".

AFAIK everything is the same on my end, as it has been for the past 3 years with that particular PPC Mac, and the way that I have Increase set up. My Prefs are there with my name, and my Team name, just like they've been for the life of that Mac since I put it online in 2/2006...so I really don't know what to think right now. I'd really appreciate some clarification about the whole deal that is going on with that Mac and that Server.

Muchas Gracias!

Kindest regards,
rexrzer

I'm no moderator, but this does not look right, to me:

[04:31:46] Project: 58379 (Run 0, Clone 64770, Gen 21248)

And I also know what's wrong : it's a little/big endian confusion ... some form PandeGroup will have to check this set of projects.

Well, that's just bloody freakin' beaaauuutiful!

My poor PPC G5 PowerMac works its tail off to squeeze a few extra points for me and the Stanford cause,
and all of the poor sick people on the planet with the various cancers, Parkinson's, Atkinson's, whatever,
and the G5 ends up getting no credit, nor do I, for all that work!

If the nice people at PandeGroup and/or Stanford, and/or BOTH of them would please get this straight and
figure out WTF is going on, and please, kindly give me some credit for all the WU's that I've been doing lately.......
and I guess getting ZERO credit for doing them, that would indeed be commendable!

I swear, it's just my poor PPC G5's bad luck to be working its arse off like it does for a few measly points, and
the project ends up with a "run 0" header, ie that's totally unacceptable...I thought that the nice people running things
over at the Stanford Folding center had it better together than this!

I don't know why I am so "trusting" of the powers-that-be here, or WHY I didn't notice these "run 0" headers on my WU's untiil now!
That's totally nuts!

Holy Moolly for Lord's Sake!

I give up and am pulling the G5 out of service until this gets resolved...this is just totally out to lunch, and I am NOT happy about it.

If I go back into my coding from weeks ago, this is all I am getting, and it's a big all-day SUCKER PUNCH:

[22:35:23] Finished Work Unit:
[22:35:23] - Reading up to 211560 from "work/wudata_05.arc": Read 211560
[22:35:23] - Reading up to 164200 from "work/wudata_05.xtc": Read 164200
[22:35:23] goefile size: 0
[22:35:23] logfile size: 62918
[22:35:23] Leaving Run
[22:35:25] - Writing 508742 bytes of core data to disk...
[22:35:26] Done: 508230 -> 424264 (compressed to 83.4 percent)
[22:35:26] ... Done.
[22:35:26] - Shutting down core
[22:35:26]
[22:35:26] Folding@home Core Shutdown: FINISHED_UNIT
[22:35:26] CoreStatus = 64 (100)
[22:35:26] Unit 5 finished with 97 percent of time to deadline remaining.
[22:35:26] Updated performance fraction: 0.965745
[22:35:26] Sending work to server
[22:35:26] Project: 58379 (Run 0, Clone 46850, Gen 24576)


[22:35:26] + Attempting to send results [January 23 22:35:26 UTC]
[22:35:26] - Reading file work/wuresults_05.dat from core
[22:35:26] (Read 424776 bytes from disk)
[22:35:26] Connecting to http://171.64.65.58:8080/
[22:35:30] Posted data.
[22:35:30] Initial: 0000; - Uploaded at ~103 kB/s
[22:35:30] - Averaged speed for that direction ~90 kB/s
[22:35:30] + Results successfully sent
[22:35:30] Thank you for your contribution to Folding@Home.
[22:35:30] + Number of Units Completed: 40

And here's my PRESENT project that I am now pulling the plug on and stopping the Folding project,
as it now stands, because yuppers, it's got another "run 0" header going for it...!

[04:31:53] Core found.
[04:31:53] Working on queue slot 07 [January 26 04:31:53 UTC]
[04:31:53] + Working ...
[04:31:53] - Calling './FahCore_78.exe -dir work/ -suffix 07 -checkpoint 15 -forceasm -verbose -lifeline 648 -version 620'

[04:31:53]
[04:31:53] *------------------------------*
[04:31:53] Folding@Home Gromacs Core
[04:31:53] Version 1.90 (March 8, 2006)
[04:31:53]
[04:31:53] Preparing to commence simulation
[04:31:53] - Assembly optimizations manually forced on.
[04:31:53] - Not checking prior termination.
[04:31:53] - Expanded 241371 -> 1271729 (decompressed 526.8 percent)
[04:31:53] - Starting from initial work packet
[04:31:53]
[04:31:53] Project: 3045 (Run 0, Clone 464, Gen 97)
[04:31:53]
[04:31:53] Assembly optimizations on if available.
[04:31:53] Entering M.D.
[04:31:59] Protein: p3045_FORMIN BINDING PROTEIN
[04:31:59]
[04:31:59] Writing local files
[04:31:59] Testing CPU type...
[04:31:59] Done testing.
[04:31:59] Extra AltiVec boost OK.
[04:31:59] Writing local files
[04:31:59] Completed 0 out of 1500000 steps (0%)
[04:46:59] Timered checkpoint triggered.
[04:53:18] Writing local files
[04:53:18] Completed 15000 out of 1500000 steps (1%)
[05:09:18] Timered checkpoint triggered.

Kindest regards,
rexrzer

Flathead74 wrote:I'm no moderator, but this does not look right, to me:

[04:31:46] Project: 58379 (Run 0, Clone 64770, Gen 21248)

There is no Project: 58379 listed on the Project Summary page (see link on the top of this page) and a simple endian swap does not yield a valid project number either. This will take some additional research.

Project: 3045 (Run 0, Clone 464, Gen 97) is valid and you'll find 3045 on that project summary page, so it's fine. You should have no trouble with it when it's completed.

Was there another WU involved?

Yo Bruce;

Thanks for piping up about the non-existent WU's!

What does that mean to my last two months worth of those things, then?

That I have been working my G5 PPC Mac for no reason?

That would truly suck rotten eggs, or worse...my Lord, what is going on here my dear Head Moderator/Explainer of Folding@Home?
It's like a bad nightmare of never ending proportions for the Mac involved...I've been wasting its CPU cycles for nothing for months it seems, with those blasted
"Project 58379" WU's which do NOT exist according to your stats page above!

As for what else that G5 was folding, well my dear sir, it was folding a "Project 3044" Gromac like this: "(Run 0, Clone 731, Gen 97)"

So it's not necessarily a bad thing when you see "Run 0" attached to a WU? I thought that was a universally bad sign!?

I honestly have no reason to put that G5 back online again the way things are...I'll just let my Intel Macs do the damage to the proteins for now.

Please do me a favor, and let me know when and if the Pande Group/Stanford has anything to say about what the G5 has been folding the past 2 months... That would be very much appreciated...you can send me a PM if you feel it's not "newsworthy" to the group here, since probably none of them are using PPC Macs for anything right about now.

Fortunately that great machine has an actual use in my photography studio: it's the No.1 production machine I use there, with a 4TB RAID sub-system (FirmTek SeriTek 2eEN4 RAID enclosure, a 2nd RAID enclosure by FirmTek SeriTek [2 TB] that houses my iTunes Library), all eSATA stuff, that generates nice, consistent 240-MB/sec writes and reads even faster than that. But that's neither here nor there.

It's NOT going to be doing any Folding@Home until further notice, I am sorry to say, as it used to be a reliable PPC Mac folder for the Gromacs over @ FAH.

Anyway, please keep me updated about the status of the "investigation" into the 58379 proteins that don't exist, the loves of my PPC Mac's Life for the past couple of months, unfortunately....

Kindest regards,
rexrzer

I have some rather disappointing, complex, and somewhat befuddling news about the "problem" we're encountering with this server.

Yesterday Bruce advised me to go ahead and let my G5 continue folding a project with the header "3405, Formin Binding etc etc", for the type of WU it was...

I went ahead, against my better judgment, and let the CPU finish that project and guess what happened today?

Yup, project "3405, Formin Binding etc etc" WU-protein, when reported by this server, turns that WU into one from "Project 58365[79] etc etc" (Non Existent WU's!!) with a header that begins "Run 0, etc etc etc"...This is the craziest, most complex, "interesting" and NEGATIVE thing that has ever happened to me personally in 5 YEARS of folding for the Stanford FAH Project. It means that every single WU that gets assigned by this server somehow, anti-miraculously, gets turned into a NON-EXISTENT WU starting out with the call numbers "58365[79]"....

Go Figure!

I am stunned by this disclosure, and discovery. There are no other words that fit. I am DEFINITELY pulling the G5 from any further folding for Stanford until this gets resolved because it means that every single WU it gets assigned by the above server is turned into virtually NOTHING AT ALL, every single time, WU after WU, day after day. I won't let this G5 do any further folding until somebody figures out what is going on with this server, its programming, or lack of it, whatever is wrong with it...

It looks like this when it all boils down to a report...REMEMBER, this started out as a legitimate "Project 3405" FAH WU, but when it reports this happens:

[01:40:38] Folding@home Core Shutdown: FINISHED_UNIT
[01:40:39] CoreStatus = 64 (100)
[01:40:39] Unit 7 finished with 98 percent of time to deadline remaining.
[01:40:39] Updated performance fraction: 0.967907
[01:40:39] Sending work to server
[01:40:39] Project: 58635 (Run 0, Clone 53249, Gen 24832)


[01:40:39] + Attempting to send results [January 28 01:40:39 UTC]
[01:40:39] - Reading file work/wuresults_07.dat from core
[01:40:39] (Read 350970 bytes from disk)
[01:40:39] Connecting to http://171.64.65.58:8080/
[01:40:42] Posted data.
[01:40:42] Initial: 0000; - Uploaded at ~114 kB/s
[01:40:42] - Averaged speed for that direction ~97 kB/s
[01:40:42] + Results successfully sent
[01:40:42] Thank you for your contribution to Folding@Home.
[01:40:42] + Number of Units Completed: 42

(I have NEVER seen a WU's data sent in 3 seconds! That is 100% IMPOSSIBLE!...it usually takes many MINUTES of my high-speed 3MB/sec cable modem to send a WU!)

[01:40:43] Trying to send all finished work units
[01:40:43] + No unsent completed units remaining.
[01:40:43] - Preparing to get new work unit...
[01:40:43] + Attempting to get work packet
[01:40:43] - Will indicate memory of 8192 MB
[01:40:43] - Connecting to assignment server
[01:40:43] Connecting to http://assign.stanford.edu:8080/
[01:40:43] Posted data.
[01:40:43] Initial: AB40; - Successful: assigned to (171.64.65.58).
[01:40:43] + News From Folding@Home: Welcome to Folding@Home
[01:40:43] Loaded queue successfully.
[01:40:43] Connecting to http://171.64.65.58:8080/
[01:40:44] Posted data.
[01:40:44] Initial: 0000; - Receiving payload (expected size: 242267)
[01:40:44] Conversation time very short, giving reduced weight in bandwidth avg
[01:40:44] - Downloaded at ~473 kB/s
[01:40:44] - Averaged speed for that direction ~332 kB/s
[01:40:44] + Received work.

(AGAIN, an IMPOSSIBLY SHORT TIME TO RECEIVE DATA)

[01:40:44] Trying to send all finished work units
[01:40:44] + No unsent completed units remaining.
[01:40:44] + Closed connections
[01:40:44]
[01:40:44] + Processing work unit
[01:40:44] Core required: FahCore_78.exe
[01:40:44] Core found.
[01:40:44] Working on queue slot 08 [January 28 01:40:44 UTC]
[01:40:44] + Working ...
[01:40:44] - Calling './FahCore_78.exe -dir work/ -suffix 08 -checkpoint 15 -forceasm -verbose -lifeline 10406 -version 620'

[01:40:44]
[01:40:44] *------------------------------*
[01:40:44] Folding@Home Gromacs Core
[01:40:44] Version 1.90 (March 8, 2006)
[01:40:44]
[01:40:44] Preparing to commence simulation
[01:40:44] - Assembly optimizations manually forced on.
[01:40:44] - Not checking prior termination.
[01:40:44] - Expanded 241755 -> 1271729 (decompressed 526.0 percent)
[01:40:44] - Starting from initial work packet
[01:40:44]

[01:40:44] Project: 3045 (Run 0, Clone 646, Gen 77)

[01:40:44]
[01:40:45] Assembly optimizations on if available.
[01:40:45] Entering M.D.
[01:40:51] Protein: p3045_FORMIN BINDING PROTEIN
[01:40:51]
[01:40:51] Writing local files
[01:40:51] Testing CPU type...
[01:40:51] Done testing.
[01:40:51] Extra AltiVec boost OK.
[01:40:51] Writing local files
[01:40:51] Completed 0 out of 1500000 steps (0%)
[01:55:51] Timered checkpoint triggered.
[02:03:17] Writing local files
[02:03:17] Completed 15000 out of 1500000 steps (1%)
[02:19:17] Timered checkpoint triggered.

That's what happens...and I have no explanation for it whatsoever. My G5 is still the same machine it was 3 years ago. The PPC-server it's been dealing with for many years remains the same, "171.64.65.58"...so nothing has changed that I am personally aware of, nothing at all, just this crazy, zany, inexplicable thing happens: THAT SERVER turns a legitimate, or what APPEARS TO BE LEGITIMATE, WU into a NON-EXISTENT WU when it comes time to report it!! And all of the events that follow make NO SENSE! How can a WU be reported in 3 seconds? How can the data for the following WU be downloaded in NO TIME AT ALL, NOT EVEN 1 SECOND?

Does anybody have any idea why, how, what, when, for what reason, what methodology or programming would cause this to happen? Has anyone ever seen this type of behavior by a server here at the FAH Stanford projects? Ladies and gentlemen, this is REALLY REALLY WEIRD! It means that my G5 gets no points credited to it, the WU turns into a space and time non-existent entity, and the projects continue to be produced in this fashion, appearing to be "normal" projects that ARE LISTED IN THE FAH DATABASE just like all the other legitimate WU's are listed.

I just don't get this at all, no comprendo!?

That is all that I have to say or write about this matter. It is officially OUT OF MY HANDS as of the end of this post, and I will leave it to the powers that be here to figure out all the particulars of the situation, which is self-defeating, discouraging, and 110% NEGATIVE to anyone's efforts at folding. This server continues to output the same type of garbage and non-existent data over, and over, and over, and over again ad finitum. I can't be the only person experiencing this phenomenon...it's got to be happening to other people and computers, and it's just nuts, so overwhelming, totally devastating to one's outlook on this project, that it just makes one want to give up and quit the whole thing.

If anybody in the Administration of the Stanford FAH project, or the Pande Group, the overseers of this project, would care to send me a PM it would be very much appreciated, because my G5 has apparently been getting, processing and reporting, then receiving WU's like this, and repeating this same sequence of events, for MONTHS! That machine's CPU time has been getting wasted and thrown in the trash for a long time, apparently, and I'd like to KNOW WHY THIS IS HAPPENING!

Just look at the sequence of events above in that log txt file, please, and tell me WHY that is happening, with all due respect...

I have been a fairly low-key, happy-go-lucky FAH participant up to now, and this just guts me out 100%, right to the very center of my enthusiasm, takes my whole thinking process about the FAH project, and just devastates it to the core. Can you blame me for thinking like this? How would you feel if you just found out that months and months of folding by a Mac in your stable had been worth NOTHING AT ALL, that you had just found out that your Mac(PPC or Intel, it doesn't really matter, the principle is the same no matter how you look at this situation) had just wasted months and months of CPU time? That it was all for nothing at all?

Well, that is where I am right now with respect to what I've found out since yesterday. I am literally just about physically sick over this episode. That PPC G5 Mac has never been anything but a gem for me, and the Stanford FAH project, folding to maximum capacity day after day, week after week, month after month, since it was put online in February 2006, and here it sits idle now, nothing to do, nothing to look forward to, until this situation is somehow figured out and solved.

Kindest regards,
rexrzer

I'm assuming Project: 3045 (Run 0, Clone 464, Gen 97) that you posted in your earlier post is the one that upon completion, got reported as Project: 58635 (Run 0, Clone 53249, Gen 24832).

Maybe a Mod can check this there is a return against this WU.

It has been returned once by Donator: y Team: 38910 on 2009-01-27 18:09:56 for 314 points of credit.

rexrzer wrote:Yo Bruce;

Thanks for piping up about the non-existent WU's!

What does that mean to my last two months worth of those things, then?

That I have been working my G5 PPC Mac for no reason?

I don't quite understand your sarcasm. I'm just a volunteer, just like you. I do not create WUs. I gather information about bugs and do my best to get them fixed. I have no idea why you've been working on a project that appears to be invalid, I'm simply noting that it is invalid and that the other one that I saw in your log appears valid.

The moderator database does not have any entries for Project: 58635 and if the results were uploaded, I'm sure the server didn't know how to credit you for it. How and when that number got corrupted will need to be investigated by somebody who knows more about it than I do. I can say that it's not a simple endian reversal.

There's nothing wrong with Run 0. All project have a Run 0.

Is it reasonable for me to presume from your attitude that you do not want any help?

I am sorry, but I am not being sarcastic!

I spoke what was on my mind, and if it affronts anyone here, then the apparent waste of time
that I've been experiencing, working my PPC G5 for no apparent reason, for no points, for no credit,
for non-existent WU's, all of that has been lost in the translation.

I have every right to be upset about wasting my CPU time on non-existent projects!
I don't know if that rings a particular bell, but I consider the 'donator's time' to be somewhat important
in the greater scheme of things.

If you can help, if you can do anything to straighten this situation out, or at least explain it for my benefit,
and for this Forum's benefit also (in general), I think that would be a good thing. There is nothing so
defeating, mind-numbing, and crushing as to find out that one's time, energy, and machinery has been used for no
apparent benefit whatsoever. I have no idea why my particular CPU got assigned non-existent projects, why projects
are getting changed from one thing to another, apparently, in the cycle of reporting on that particular server, but it
really smacks of something being very askew, or out of control. It seems to be contrary to the whole spectrum of
activity here in the Folding community as a whole, where being IN control is of absolute necessity, where having a simple
purpose, at least an apparent purpose, is an absolute value of 1 to 1.

If there's anything that I mean to say it's pretty simple, and it's this: Respectfully I didn't mean anything personal toward
any of the people here who are Moderators, or site hosts, or FAH people that know, ostensibly, what
is going on 100% of the time. There was nothing of a personal nature in anything that I wrote, and if someone thinks otherwise
perhaps it would best be communicated to me in a PM rather than the bandwidth of this forum. I happen to know what
sarcasm means, and if my words were taken to be anachronistic and sarcastic, then you were looking for that, when it in
fact was not present at all.

I was, and am genuinely frustrated and perplexed by what is happening with that server, and I am not getting credit for
units obviously, for starters, that my PPC G5 CPU has been completing like a clock because the clock is non-existent!
I have no idea what is going on, and nobody from the Pande group, the FAH group, or anyone of an "official nature" has
stepped up and voiced an opinion about what, and why, and when, and how, and by whom, things with that particular
server are being controlled. There has been no official explanation for the goings on about that server, and why units are
mysteriously being changed into things that receive no credit, and have no apparent rational basis for even being here in the
1st place.

Help? Of course that generic term is appreciated by myself, and if anyone of an official nature here indeed CAN
help explain what is going on with that server, then by all means "help" would be a very, very good thing to have.

All I really know is that my G5, which has folded many, many hundreds of WU's for the FAH cause since 2006, has been used
for no good reason, for no good whatsoever, because just a couple months ago things were NOT like this with that server, things
were in control, and I was getting credit for WU's that were legitimate, that were completed on time, in fact waaaay ahead
of being on time in all cases...and somewhere, somehow, some way, a big fat Monkey Wrench got thrown into the mix of things
with that server and its programming, apparently, and the result is what I've been experiencing for the past couple of months,
apparently: the G5 has been Folding WU's that either don't exist, or that DO exist and are being credited to non-existent 3rd parties
or entities unknown to myself, or to anyone in the FAH community at large.

In short, I have literally NO IDEA what is happening with that server, with my G5's CPU time, with any WU's that are created by
that server, have no idea why or how my G5's WU's are being co-opted by another person, entity, 3rd party, whatever...it is all
100% confused, confusing, and altogether out of control in a very real way, and I cannot participate in the work at FAH with my
G5 any longer until, and if, things get resolved.

If anyone can help with that situation, then by all means fire away, as I am 100% interested in things being explained, and figured out
and would be very pleased to assist in any way, shape, or form that I can, rest assured! Presently, my Intel Macs
are folding tons of WU's of the SMP variety, with great success, and I can't understand why or how the PPC Mac deal is all screwed up
for some reason. It just baffles me why, or how, things could be that messed up while FAH goes on about it business in all other
respects full steam, and speed ahead, no problems at all.

I repeat once again, there was nothing personal, nothing directed toward any Moderator here in any substantive way
by myself, and I have no reason to be thinking or doing anything like that with respect to what is going on with that out-of-control server.
If I can be of any help just ask, and I'll do what needs to be done from my end if I understand it and can carry out
any commands from anyone from the Pande group, the FAH officials, or anyone with knowledge of some concrete nature regarding
what has been going on with that particular server.

It's just a Big Mystery to me personally right now, and I'd like to know why, how, when, by whom, for whom, things became askew and out
of control as they are apparently right now. I cannot trust that server to do anything rational, apparently, and I'd like to be able to trust
in something, anything, anything at all really, with regard to having my PPC G5 back in service doing FAH, as it has been doing since it was "born"
back in early 2006, put on line as my main PPC Mac work machine, and certainly as my No.1 folding machine for PPC WU's.

If I need anything at all right now it's a simple explanation by somebody, any one of you in the power positions here in the forum, in the
Pande group, the FAH group, whomever you know (?)...about what has been going on with that server. And especially WHY it's been askew as it
seems to be right now... that would be really nice, and appreciated.

Kindest regards
rexrzer