Paul D. Buck, as usual:

Thanks for the assist!

Also, sorry for lag in reply!

Not exactly a "great" day @ work today & had to work later than usual because I am having problem there too with some methods I know I can use in older languages, but which are 'problematic' when doing WebForms coding (diff. than Windows Forms which is more like Win32 based code really)...

I gotta VENT, lol, because it's PISSING ME OFF!

E.G.-> Passing parameters around between 40 webforms, & from a multi-threaded SINGLE module with 2 subroutines (one is actual subroutine parameter intake + processing engine, & other is a thread creator which spawns said function on its OWN thread which IIS 6.x on Windows Server 2003 runs server-side, talking to SQLServer 2005 (like ISAPI DLL stuff, but NO LEAKS))!

This publicly visible module w/ it's 2 functions are/is the 'reuseable function engine' for 40 class-based VB.NET webforms!

Why am I pissed? WELL, it's a hassle & does NOT work the way it used to in VB6 for instance!

The AddressOf operator for multithread code really does NOT like parameterized subroutines I found out (has NO trouble with ones that are not parm'd/argumented though)...

HOWEVER, I have a 'funky-workaround' one of my co-workers turned me onto (good dude, smarter than hell imo) via passing HttpResponse, HttpRequest, Sessions & global scoped Crystal Reports parameters for firing .rpt reports directly sending them parameters via a single ASP.NET page.

Trouble is, it works on SOME of them, not others - probably my local DSN is messed up or I am passing the wrong info. for some of the Crystal Reports is all! I hope so... lol, this IS my job, & I am like anyone else today:

Only as good as "what have you done for me lately" imo!

Anyhow - on VB.NET:

Still getting "used-to-it" here really! I think what I have is "OK" but is not consistent enough yet, & I would hate to have to do each of the 35 webforms & their reports left as "inlined" separate aspx.vb coded classes.

I would rather have that SINGLE ENGINE multithreaded design, cleanly (no global scope crap) implemented, passing parameters around from its SINGLE point of maintenance. Imo, it's 'better design'...

(Yes, it's a bit tough "adjusting" to this 'new' technology - I am coming out of a VB6/Delphi/Access (C/C++ years ago too) background mostly is why (which, in a way, both of those languages REALLY comprise a great deal of the technology in the .NET programming IDE environs in Visual Studio 2003-2005 so it's not hard getting used to really - same designer for TurboPascal/Delphi for Borland works for Microsoft now, & designed Visual Studio & .NET for them largely is why)).

:)

* Anyhow, sorry for the book above, lol (helps me vent, very therapeutic writing it out too)...

Nice info. archive there as well up @ Wikipedia: Very comprehensive imo @ least!

(Still having some hassles with THIS stuff in BOINC/SETI, but it is just an excuse to try to research how to fix the problem I'm having here is all!)

Right now?

Well, it seems that the MOST easy method would be reinstallation to do it absolutely 110% right, & then just bind myself to the project again & my team though, then lastly install Crunch3r's optimized client build 4.11 which I have always done the best with thusfar performance-wise!

APK

P.S.=> NO biggie, either way - the best/brightest part is, no matter what, I now understand this distributed computing concept suite's functionality better than ever now, & that's always a plus, & not a 'wasted day' imo either... apk

---

http://torry.net/authorsmore.php?id=1781

"The object's hull is made of SOLID neutronium: A single StarShip cannot combat it!" quote Mr. Spock, Star Trek original series, episode title: "The Doomsday Machine"

Hi,

I've been able to upload all day & night until a few hours ago where I get the same problem as yesterday. The upload/download server says its online but it still won't upload.

I've been unable to upload or download for about 5½ hours. I notice on the network traffic graph that the traffic level is still rising so i expect it'll be at least another few hours before it starts to drop off to more normal levels & i'll be able to upload & download again.

---

Grant
Darwin NT

You'll have to wait until someone gets the time to post here or write news on the frontpage about what's going on. I just downloaded a wu (after several retries) but uploads are always fubar'd, because of this ****in' two step connect...

---

Aloha, Uli

Hi,

I've been able to upload all day & night until a few hours ago where I get the same problem as yesterday. The upload/download server says its online but it still won't upload.

Thanks,
Ben

---

* QUESTION - Do any of you guys have any idea of the "mechanics" of the subfolders beneath where the SETI*.exe & BOINC*.exe files are stored & what goes on in them?

Yes, but you will have to do your part ...

There is an awful lot of this material in the Wiki, stilll not as well organized as I would like. In many cases using the google search for the wiki will get you to a good list. But, other times you have to do some digging.

The "participant" side tends to get most of the attention because that is where most people are. The development side is less well done. In many cases we have just the UCB pages with the nomenclature normalized and links added. In other cases I have additional material that is unique to the Wiki and extends the UCB material considerably. BOth sites have had some reorganizations and additions in the last month and I am trying to get work done on some old pending material and to add/condense the "cookbooks" into one set of information instead of asking people to wade through 20 cookbooks that all may be out of date and contridictory.

Anyway, one of the places to start is "BOINC Client Directory", I don't recall if I fully explain the sub-directories or not ... if not complain ...

---

Thanks guys!

I will probably end up having to redo/reinstall the BOINC/SETI client again @ worst it appears is all - this is PROBABLY my fault!

(Thanks for pointing that out I guess, because it probably IS the "culprit" here, lol, my own messing around!)

:)

Still, this "hassle" of mine (and your suggestions as well) made me look thru the folders & files that compose this project's programs suite, & I now have a decent understanding of what is in the .XML & .CFG files that boinc*.exe & seti*.exe use!

(That helps LOADS imo!)

* QUESTION - Do any of you guys have any idea of the "mechanics" of the subfolders beneath where the SETI*.exe & BOINC*.exe files are stored & what goes on in them?

Once I understand that, I can REALLY understand how this program suite works LOCALLY @ least... & that will help me understand it better & be more able to fix it up myself (w/out reinstalling etc.).

APK

P.S.=> All I needed was a little "push" is all, to get off my lazy behind lol, & take a peek inside the .XML & .CFG files that drive/keep state for the executables this suite uses, with notepad.exe!

After ALL that?

It turns up my reinserting Cruncher's 4.11 build of SETI is NOT reflected in many of the .XML files, but instead refs to the 4.18 build I used to use are in them... I edited them & will restart the boincmgr.exe & see how it goes!

Hopefully? That will 'set it right' again... & if not? I reinstall is all, putting in Cruncher's 4.11 build of SETI*.exe again (gives me the best fpops results on the boinc.exe benchmarks @ least, so I go with it!)... apk

---

http://torry.net/authorsmore.php?id=1781

"The object's hull is made of SOLID neutronium: A single StarShip cannot combat it!" quote Mr. Spock, Star Trek original series, episode title: "The Doomsday Machine"

After four hours of nothing uploading, but downloads working very slow, in a fifteen minute span, the almost fifty "uploading" results on five machines started uploading slowly. I hope this is not an example of an "improvement" and the future performance of a recovery.

First of all, thank you for helping.

Your welcome, anything we can do, we try. :-)

Secondly - I am MOST concerned about checking this HOST stuff, because of these messages & I connect to setiboincdata.ssl.berkeley.edu, because I see this in the stderrdae.txt:

Reason being, is because SETI is currently going through a "catch-up" phaze, as they brought things down earlier today, and usually after an outage, it can be anywhere from a couple of hours to a day or so, till things start flowing again. I got 1 WU on this machine that won't upload right now, because of the outage. Its nothing to worry about at this time. If in a few days your still getting the messages, then there might be other problems and things to look at from there.

Just keep an eye on things, see if it improves. My biggest concern now would be to get your app's working right, so you can stop sending back invalid results, and getting credit again. As long as you keep sending bad WU's back, your quota won't increase past 1. For every good one you send back, your quota doubles. So it will go to 2, then 4, and so on.

Jeremy

---

Alec - to the best of my knowledge, you can't change your quota once you get down to 1/day. The only way that will change is by you successfully returning good results. Then your quota should start to increase incrementally.

Patience is a virtue on some days...

<grins>

---

Wendy



Click Here for BOINC FAQ Service

Hey Alec.

First, unless someone else posts, we went to your account by clicking your name, which brings us to Account Data for AlecStaar. Then, we click on Computers. We then click on the host that has a host id of 2130409 which is the computer that was found to have all of these errors.

As MJ stated, lots of errors, which brought your Maximum daily WU quota per CPU = 1/day.

I looked at one of the work units, and I'm taking a wild guess, but I believe you might have messed up with the installing of an optimized app, since you say you played around with them.

My suggestion, is go and download CPUID and download the Version 1.32, and see what your computer is capable of, as far as MMX, SSE, SSE2, SSE3. Then, do a search for "Optimized" on the Seti Boards. Theres many. Crunch3r, Trux, Tej, and a few others have optimized apps. The choice is yours on which one to go with. Then, go to The BOINC WIKI Optimized App Installation Guide and read it and follow the steps exactly as it says.

I hope that helps a bit. Good luck, and let us know if you need any more help.

Web03, KB7RZF, & MJKellher:

First of all, thank you for helping.

Secondly - I am MOST concerned about checking this HOST stuff, because of these messages & I connect to setiboincdata.ssl.berkeley.edu, because I see this in the stderrdae.txt:

"Couldn't connect to hostname [setiboincdata.ssl.berkeley.edu]"

+

"Message from server: (reached daily quota of 2 results)"

QUESTIONS:

1.) By checking my HOST, what is that, exactly?

2.) How can I alter my WU/day setting??

(Disregard question #2 - Apparently, this does not sound like something I can set, per the statements of KB7RZF (which I take as 100% straight-up right) but is the result of the many errs I had today!)

See, after that suggestion by MJKelleher below?

Well, I looked on the user prefs page & updated my prefs (for how SETI@Home uses my BOINC clients etc. from the website) & STILL I get those messages above! I hit the UPDATE button in the Project tab too after that...

E.G.-> Is checking HOST in the interface of boincmgr.exe, or webpage managed by myself on MY end in my preferences (was I in the right spot)??

:)

* Thanks!

APK

P.S.=> If any of you KNOWS how the 'peripheral files & folders' (the .txt, .xml, & other config + err msg files but most importantly, the folders & what they store & do) for this project interact, just a synopsis really, could you tell me more as well? TIA! Because once I know? Well, I too can be of service to others here as well, pulling them out of jams too (as I am having now apparently with what appears to be a 'scrambled config' here now, lol)!

Also - Ah, thanks guys... figures I may have screwed it up myself!

I did reinsert Cruncher's build into the mix here!

In the doing of it, I doubtless/probably "scrambled the eggs" of my SETI@Home folder but good!

See, I have found I like his work the best for a GOOD mix of floating point & integer op power (with stress on FPOP results, because it's really what matters for units processing imo @ least because of the nature of the dataset processed & Cruncher's builds show more/the most profiency in that area vs. others (no offense intended, because Trux's stuff is good too)).

It never did that before, the doing of that mixing in new optimized builds here (Just searching for best possible performance out of it & all that), but there is ALWAYS a first time!

Now, I have to "unscramble" it, & knowing how the files & folders work in this project's files will help a LOAD imo! You can't fix something on computers imo, w/out understanding its process fully... apk

---

http://torry.net/authorsmore.php?id=1781

"The object's hull is made of SOLID neutronium: A single StarShip cannot combat it!" quote Mr. Spock, Star Trek original series, episode title: "The Doomsday Machine"

The slow uploads I was on about actually never made it...they just halted.

Same here.

Just a few minutes ago, my two slowest systems each uploaded one WU result and was able to Report those two. All machines have completed pending downloads. Activity on the Cogent link Graph picked up about that time.

As there are not thousands of posts about not being able to upload, we have to assume that most users are not having any problems at all uploading. And have uploaded all of their results. So there is just a "new" problems that a few of us are having.

Hey Alec.

First, unless someone else posts, we went to your account by clicking your name, which brings us to Account Data for AlecStaar. Then, we click on Computers. We then click on the host that has a host id of 2130409 which is the computer that was found to have all of these errors.

As MJ stated, lots of errors, which brought your Maximum daily WU quota per CPU = 1/day.

I looked at one of the work units, and I'm taking a wild guess, but I believe you might have messed up with the installing of an optimized app, since you say you played around with them.

My suggestion, is go and download CPUID and download the Version 1.32, and see what your computer is capable of, as far as MMX, SSE, SSE2, SSE3. Then, do a search for "Optimized" on the Seti Boards. Theres many. Crunch3r, Trux, Tej, and a few others have optimized apps. The choice is yours on which one to go with. Then, go to The BOINC WIKI Optimized App Installation Guide and read it and follow the steps exactly as it says.

I hope that helps a bit. Good luck, and let us know if you need any more help.

---

but it tries to upload more than the size of the file (like it goes to 106%) lol, very strange.

Seeing that same thing also, which is normal. But it is NOT completing the transfer (or at least not getting the acknowledgement back), so it retires.

After almost three hours of trying, not one single upload on five systems with lots of pending uploads. Downloads have been VERY slow, much slower that in all prior recovery, but all machines have been able to slowly download most of their pending downloads.

So something is now broken with uploads.
I have stopped and restarted BOINC.
I have rebooted the system.
I have power cycled the router and system.
I etc. .... still does not upload.

Im finding that units that are not complete, when completed are able to be reported (at least from 2 pc's i`ve checked) so it may be a problem with units completed before a certain time. But, all units that are ready to send back from earlier on can't be reported. The slow uploads I was on about actually never made it...they just halted.

---

"I'm not worried, I have a days worth of cache :)" - Michael Buckingham

Yea, lol, neither was I... until it ALL processed out on me! I set my system to only talk to the servers once every 3 days, so it usually hauls down TONS of units to process.

You've got something else holding you back, APK.. check your host, your quota is down to 1/CPU/day, and you've got a round 100 results returned with Client errors today, starting well before the server outage...

MJ

Ah, thank you for responding, first of all - I would like to be "back-on-track" smashing thru more units, you know?

First off - Please try to understand:

I am TRULY only an "amateur" user of this project's tools & such!

I also have done a LOT of 'mixing/matching' optimized build clients, & perhaps this HAS 'backfired' on me? It NEVER did before... but, there is always a first!

And, I really do not know much of how ALL of its parts interact!

(E.G. -> The subfolders especially & files in them, PLUS the files other than the .exe's & .dll's such as the .xml files (many of these), & the .cfg files (they probably store configuration data of somekind, but I feel like when I first started out using Windows years ago, & trying to understand .ini files etc. & later the registry))...

So, that said?

First, I need to know a some things/could use your help:

1.) By checking my HOST, what is that, exactly?

2.) How can I alter my WU/day setting?? I looked on the user prefs page & updated my prefs (for how SETI@Home uses my BOINC clients etc. from the website) & STILL I get this message:

& I connect to setiboincdata.ssl.berkeley.edu, because I see this in the stderrdae.txt:

"Couldn't connect to hostname [setiboincdata.ssl.berkeley.edu]"

+

"Message from server: (reached daily quota of 2 results)"

E.G.-> Is it in the interface of boincmgr.exe, or webpage managed by myself on MY end in my preferences (was I in the right spot)??

:)

* Thanks!

APK

P.S.=> If any of you KNOWS how the 'peripheral files & folders' for this project interact, just a synopsis really, could you tell me more as well?

Thanks!

Because once I know? Well, I too can be of service to others here as well, pulling them out of jams too (as I am having now apparently with what appears to be a 'scrambled config' here now, lol)! apk

---

http://torry.net/authorsmore.php?id=1781

"The object's hull is made of SOLID neutronium: A single StarShip cannot combat it!" quote Mr. Spock, Star Trek original series, episode title: "The Doomsday Machine"

"I'm not worried, I have a days worth of cache :)" - Michael Buckingham

Yea, lol, neither was I... until it ALL processed out on me! I set my system to only talk to the servers once every 3 days, so it usually hauls down TONS of units to process.

You've got something else holding you back, APK.. check your host, your quota is down to 1/CPU/day, and you've got a round 100 results returned with Client errors today, starting well before the server outage...

MJ

---

but it tries to upload more than the size of the file (like it goes to 106%) lol, very strange.

Seeing that same thing also, which is normal. But it is NOT completing the transfer (or at least not getting the acknowledgement back), so it retires.

After almost three hours of trying, not one single upload on five systems with lots of pending uploads. Downloads have been VERY slow, much slower that in all prior recovery, but all machines have been able to slowly download most of their pending downloads.

So something is now broken with uploads.
I have stopped and restarted BOINC.
I have rebooted the system.
I have power cycled the router and system.
I etc. .... still does not upload.

My PC's show 'uploading' where they should show 'ready to report' even though they're not even trying to upload - when I press 'update' the client reports that there is no work to report when there are at least 5 units on each pc to send :S

Go look at the Transfer tab,and hit "Retry Communications" fromthe "COmmands" drop down menu item.

Thanks. That started to send very slowly 0.14KB/s and seems to have stopped at 0.28KB sent. THat's just one out of 7 on this pc that actually tried to report. The rest say retry in x minutes.

edit - I can upload (occasionally) but it tries to upload more than the size of the file (like it goes to 106%) lol, very strange.

---

My PC's show 'uploading' where they should show 'ready to report' even though they're not even trying to upload - when I press 'update' the client reports that there is no work to report when there are at least 5 units on each pc to send :S

Go look at the Transfer tab,and hit "Retry Communications" fromthe "COmmands" drop down menu item.

---

My PC's show 'uploading' where they should show 'ready to report' even though they're not even trying to upload - when I press 'update' the client reports that there is no work to report when there are at least 5 units on each pc to send :S

---

"I'm not worried, I have a days worth of cache :)" - Michael Buckingham

Yea, lol, neither was I... until it ALL processed out on me! I set my system to only talk to the servers once every 3 days, so it usually hauls down TONS of units to process.

(@ least so far, the more I study HOW this system works it appears to be that way)

Now though?

Well, it's telling me "communication deferred" & it's been this way for 3 days now. This delay MAY be the result of my setting my BOINC client prefs to ONLY COMMUNICATE WITH THE SERVERS EVERY 3 DAYS.

I am sort of "worried" - BUT, I dunno, you guys tell me, on that account. Most of you understand this system a HELL of a lot better than I do.

(But, as to worries? WELL, not really! You know - it's over nothing crucial really. It's not my job or food in my fridge & bills paid, you know?)

However, I am 'worried' that my "RAC" which has gone way, WAY, down from the 535++ area to under 300 now.

That's ok, not a "BIG" deal. BUT, I am wondering when my system will retrieve more units is all.

So, same here fellas...

E.G.-> I had 2 days of work tab status messages of "ready to upload" units in my work queue tab, but no uploads.

LOL, then, when it DID upload? I got LOADS of errors for some reason.

Not meaning to complain either (in my above statements), or being rude/facetious... because "shit happens".

Personally, I am sure this will be "ironed-out" by the folks that run the back-end of this distributed computing concept BOINC system!

(See, I try to realize a few things about it. First of all - it's young still & Secondly? I imagine it is getting flooded by new users (I have @ least 3-4 folks I know that want to get into it once they get systems, after I told them about SETI@Home in fact), & the SETI round #1 folks joining it as well probably).

Maybe it's just TOO MUCH @ ONCE & the system has to be 'mended' @ the backend DB level for the new larger loads?

I dunno... just guessing!

* Anyhow - like the outages we had, & other things? It will be mended I am sure, & the units will come out again like before, once things are straightened out.

:)

APK

---

http://torry.net/authorsmore.php?id=1781

"The object's hull is made of SOLID neutronium: A single StarShip cannot combat it!" quote Mr. Spock, Star Trek original series, episode title: "The Doomsday Machine"