Seems like these are old wu's
Yup. Now I can't tell which WU my machine FUBAR and which were FUBAR when I got 'em.

My theory is that we've drained Arecibo dry, so Berkeley's sending us these... uhm... "rejects"?

---

> > ...
> > Of course they doing it when the system is having problems.
> >
> LOL, thanks for mentioning it...
> If i would have said that, i would be disrupted, as usual ;)
>
>

it seems, that for one who so likes to use sarcasm and speak tongue in cheek, that you take any such inferences from others personally.
we are not out to get you Uli.
and i actually agree with that statement above too, but 'stuff' happens.

---

S@h Berkeley's Staff Friends Club

> My connect is set up for the max, every 10 days.
> Still only lets me download 4 work units.
> Something amiss here.
> 4.19 was fine. Would even let me bank up to 30 wu at a time.
> 4.25 has some issues. It won't allow me and 35 others now more than 4 per day.
> And we are all set to 10 days between connect time.
> Seen this exact same issue on 35 computers now.
> Thanks,
> John
>

i downloaded boinc v4.25 on march 8. cache 8 days worth of workunits. seti runs the same as on v4.19, no problems.

different topic, today i have been able to u/l but not "report" wus and no d/l, at all. :)

RNV

---

To make the scenario I am taling about clear.
I can upload an ddownload and "update project" just fine with climate predictor, on all of my computers.
Yet myself, and 35 others are not able to download any new work units, nor are we able to "update project" on seti. WE ARE able to upload work units as they are completed.
So we are basically in the same boat as many others.

Any word on what is the cause? out of work units? If so, that is OK, we can come back later.
If not? ????

All are running the same computers we had with the same setups we had before upgrading from 4.19 to 4.25. The issues started there. 4 wu per day limit, now unable to download, etc.

again, no sarcasm from here, just wanting to know when we will be able to contribute again. we all know this is volunteer, and for fun.

Thanks,
John

---

Thanks, and Keep on crunchin'
John Henry KI4JPL
Sevierville TN

I started with nothing,
and I still have some of it left.
<img src="http://www.boincstats.com/stats/banner.php?cpid=989478996ebd8eadba8f0809051cdde2">

> it seems, that for one who so likes to use sarcasm and speak tongue in cheek,
> that you take any such inferences from others personally.

Sorry, that once again came down the wrong way. I don't take that personally :)

> we are not out to get you Uli.
I'm not quite sure about that :fear:
BTW: You won't get me either ;)

> and i actually agree with that statement above too, but 'stuff' happens.
>
Exactly, what i wanted to express. Goin' out now to Biergarten with friends enjoying the spring weather %)

---

Aloha, Uli

> Sorry, that once again came down the wrong way. I don't take that personally
> :)
>
cool =)

> > we are not out to get you Uli.
> I'm not quite sure about that :fear:
> BTW: You won't get me either ;)
>
yeah, would be difficult. went through Frankfort airport twice, but don't speak any german, and knowledge of the local geographies is nil.
well, other than ich bin ein auslander.

> > and i actually agree with that statement above too, but 'stuff' happens.
> >
> Exactly, what i wanted to express. Goin' out now to Biergarten with friends
> enjoying the spring weather %)
>
ah yes, very nice here today too.
i watched a little bundesliga this weekend, funny to see the shortsleeves when it was just 3 weeks ago they were playing in snow

happy crunching (and back on topic maybe)

Micah

---

S@h Berkeley's Staff Friends Club

> Just letting you know the status of several boinxsters that are now down due
> to connection issue at BOINC, and a crazy new limit of 4 w/u.

BOINC has some new code to measure the relative "reliability" of various hosts.

When a host is deemed unreliable, it lowers the limit to favor other hosts that do return work reliably.

I see on both of your machines that there are several "lost" work units. That might be the reason.

As you return more work and are granted credit, the limit will increase.

You may also want to modify your "connect every 'x' days" parameter so that you'll cache more work once the limit starts to rise....

---

I guess that a limit on reliablity is good.
But one has to remember, the only two times I had to reset the project were due to the seti software and people telling me to do so.
Those were only two instances since starting with seti, and I usually return more than 8 per day from the computers I am running on.
But also have to remember, there are 35 other people in the same scenario, where they were told by the seti sw and by the seti staff to reset projects, based on problems that were in seti, and not their own.
Not a good thing to tell 35 people that they are unreliable, when they were only following what seti told them to do when seti programs were having problems.
Unfortunately, some are dropping boinc seti now, and dedicating their time to others that have not imposed an unreliable limit on them.

---

Thanks, and Keep on crunchin'
John Henry KI4JPL
Sevierville TN

I started with nothing,
and I still have some of it left.
<img src="http://www.boincstats.com/stats/banner.php?cpid=989478996ebd8eadba8f0809051cdde2">

I think I see what Digger sees - data upload fine, but schedulers
quit responding sometime in the early am of Mon (CA time),
thus no credit being granted. It doesn't seem that problems with
bad units or anything like that would cause schedulers to not even
respond. Also, it looks like when they rolled back from 4.10 to
4.09, sometime early am Sun., that the data link got hammered for
as while, see

http://fragment1.berkeley.edu/~cricket/inr-668-interfaces.html

as upgraded clients had to re-download 4.09,
but later on Sunday it had recovered, and I was seeing no failures
at all, and I was uploading and reporting 4.09 units just fine
after the switch back to 4.09. I've got some 4.10 units, but
haven't run any yet due to caching. Hope they are not bombs waiting
to explode (I never use graphics, at least).

So it would seem that early Mon. am, either devs started working on
something that affects the schedulers, or (more likely, I think)
something unrelated just blew up of it's own accord. (But in such
a way that the server status still sees the scheduler status as
"online".)

hmmph. We'll see. This is why we have WU caches!

Dave

> I guess that a limit on reliablity is good.
> But one has to remember, the only two times I had to reset the project were
> due to the seti software and people telling me to do so.
> Those were only two instances since starting with seti, and I usually return
> more than 8 per day from the computers I am running on.
> But also have to remember, there are 35 other people in the same scenario,
> where they were told by the seti sw and by the seti staff to reset projects,
> based on problems that were in seti, and not their own.
> Not a good thing to tell 35 people that they are unreliable, when they were
> only following what seti told them to do when seti programs were having
> problems.
> Unfortunately, some are dropping boinc seti now, and dedicating their time to
> others that have not imposed an unreliable limit on them.

I've been around the forums for a while, and I would say that 95% of the time when people say "reset the project" that the recommendation is from someone in the user community, not an official representative of the project.

... and in my opinion, some are too quick to reset -- waiting often works fine.

But, that said, the same mechanism that lowered the limit will raise the limit as work is returned.

So, you've got a choice: you can get angry and go away because you think the project has treated you badly, or you can relax, let things run well, and watch the limit go back up.

Now, what I would do -- representing myself only and my comments are not binding on SETI, BOINC, Berkeley or any other project:

Don't stop crunching SETI. Add another project or two so that you'll always have something to crunch, and hopefully you won't be penalized by the other projects for problems. You can "fill in" with E@H while you wait for your reliability to increase at SETI. You can even set BOINC to do 90% SETI and 10% E@H if you like SETI best.

This is a volunteer project. I'm frequently amazed how often people get upset when a project is (again, IMHO) working perfectly -- how they constantly complain when something is down when BOINC does not require 100% uptime. How they complain about how work is distributed, etc.

If you have 35 people running BOINC on their desktops, then we're talking about CPU cycles that would have gone unused -- why get upset over what is essentially a zero-cost resource?

---

> But also have to remember, there are 35 other people in the same scenario,
> where they were told by the seti sw and by the seti staff to reset projects,
> based on problems that were in seti, and not their own.

As Ned said, most comments are posted by people NOT working on the project directly. You can look under the name of the person that posted to see if they are "oficial" or not ... Like I am not connected directly to any BOINC Powered Project other than as a volunteer scribe.

> Unfortunately, some are dropping boinc seti now, and dedicating their time to
> others that have not imposed an unreliable limit on them.

As the BOINC system matures we are slowing improving the *SYSTEM* reliability by adjusting the operational parameters. Some computers return a lot of bad results. This causes problems, so, the scheduler has been changed to try to reduce that problem. In my case, for example, I have been having some trouble with the Macintosh version of BOINC and its ability to reliably do work for CPDN. For awhile I removed CPDN from the Mac just because of this problem. BUt I continued to do work for other projects because they don't seem to have the same issues.

---

Hi,
Only reason they and myself get frustrated, is that we were told in a post in november by a seti at home worker to reset our projects, so we did.
Now, with the limit of 4, and going through 8-15 a day, we have to manually try to reload work units many times during the day.
Compounded with the fact that all 35 of us can now not download any new work units, nor can we even update our status, that adds to another level of frustration. There is no response from the server, even though the server status is up. I know things can get backlogged and bogged down, but this has been going on now for almost 5 days. We waited it out to try to wait for things to catch up, and still as of this moment, I just tried and got the no response message.
I can only speak for myself in that I will wait for the download of work units issue to be resolved, then I wil continue.
One person is frustrated because she downloaded and processed over 400 work units, and posted a message 2 mos ago about a problem, and was told on this message board by a real seti at home person, at least they said they were, that she should reset, she asked if any ramifications, such as future limits, and the person said flatly, NO, especially since this is purely a volunteer basis and they wanted to keep volunteers, even through bad times due to any reason.
Then she asked if others in our group had the same problem, should they reset, and the answer was YES.
So, they all had the same problem, and each person lost on an average of 10 units that show up as no response or other such client issue.
One person, with over 400 units returned, has one reset project of 4 units lost, and another rest project of 8 units lost. ALL TOLD BY SETI TO RESET.
THey first waited, and could not get any new until those were reset, so there was no choice. It was a problem with the servers in November that caused this.
I hope you can read this, understand why these people feel the way they do, and see what can be done to help out people who followed official instructions in the future.
It is an improvement opportunity, not a point of contention and should not make people feel defensive or rejected, learn from mistakes, and go on.

Thanks,
John

---

Thanks, and Keep on crunchin'
John Henry KI4JPL
Sevierville TN

I started with nothing,
and I still have some of it left.
<img src="http://www.boincstats.com/stats/banner.php?cpid=989478996ebd8eadba8f0809051cdde2">

Second that about the resets, Ned. I'm with you.

In running sah/boinc for
quite some time now (some Beta, and then public), and on a motley
assortment of HW and SW (from .2GHz to 2.8GHz, and matching vintage
WinOS's, 4 of them) I can only recall one instance in which I really
had to end up dumping units, and only recently did I have (only the
fastest) one machine run out of work for a bit.

Dave

> Hi,
> Only reason they and myself get frustrated, is that we were told in a post in
> november by a seti at home worker to reset our projects, so we did.
> Now, with the limit of 4, and going through 8-15 a day, we have to manually
> try to reload work units many times during the day.

Am I the only one who thinks that having all these people with a daily D/L limit of 4 is a little strange. I would like to think I've been here (forums) long enough to have heard of this "Limit". I mean, I don't think that there was even one posting in the last 4 months that I haven't read. I don't ever recall hearing of a Daily D/L limit being this low.

I find this strange. I think John and his friends are being misled by someone. John, do you remember who told you that all 36 of you had a limit of 4?

tony

No person said this originally.
THe software told me that when I extended my day connect again to 10 days, and tried to do an update.
I already had 4 wu downloaded, and tried an update. The software then told me in the messages that I had a 4 w/u limit.
Several others in the group (note, not a registered group, just a bunch of hackers working thoruhg issues as they come up together), several others in our group had the same response. 35 out of 75 in our forum had this.
Only confirmation of such a limit was by Ned today in this same thread.
I fully agree with a "dead responder" limiter, but there should be some criteria where the reason should be known.
Thanks,
John

---

Thanks, and Keep on crunchin'
John Henry KI4JPL
Sevierville TN

I started with nothing,
and I still have some of it left.
<img src="http://www.boincstats.com/stats/banner.php?cpid=989478996ebd8eadba8f0809051cdde2">

> Hi,
> Only reason they and myself get frustrated, is that we were told in a post in
> november by a seti at home worker to reset our projects, so we did.
> Now, with the limit of 4, and going through 8-15 a day, we have to manually
> try to reload work units many times during the day.
> Compounded with the fact that all 35 of us can now not download any new work
> units, nor can we even update our status, that adds to another level of
> frustration. There is no response from the server, even though the server
> status is up. I know things can get backlogged and bogged down, but this has
> been going on now for almost 5 days. We waited it out to try to wait for
> things to catch up, and still as of this moment, I just tried and got the no
> response message.
> I can only speak for myself in that I will wait for the download of work units
> issue to be resolved, then I wil continue.
> One person is frustrated because she downloaded and processed over 400 work
> units, and posted a message 2 mos ago about a problem, and was told on this
> message board by a real seti at home person, at least they said they were,
> that she should reset, she asked if any ramifications, such as future limits,
> and the person said flatly, NO, especially since this is purely a volunteer
> basis and they wanted to keep volunteers, even through bad times due to any
> reason.
> Then she asked if others in our group had the same problem, should they reset,
> and the answer was YES.
> So, they all had the same problem, and each person lost on an average of 10
> units that show up as no response or other such client issue.
> One person, with over 400 units returned, has one reset project of 4 units
> lost, and another rest project of 8 units lost. ALL TOLD BY SETI TO RESET.
> THey first waited, and could not get any new until those were reset, so there
> was no choice. It was a problem with the servers in November that caused
> this.
> I hope you can read this, understand why these people feel the way they do,
> and see what can be done to help out people who followed official instructions
> in the future.
> It is an improvement opportunity, not a point of contention and should not
> make people feel defensive or rejected, learn from mistakes, and go on.
>
> Thanks,
> John

This message is pretty jumbled, so it's hard to respond intelligently.

I didn't look at things happening months ago -- any limit is based on recent history (as in hours and days, not weeks or months). I do see work from you that was not returned in the recent past.

I of course can't comment about some unnamed person or the 35 others without looking at their machines.

... and as for getting work right now that isn't anything that reflects on you. There is something going on that is causing congestion.

My best recommendation: drink less coffee. This isn't about you (or me) and at the end of the day, we're still talking about using waste CPU cycles.

If we're using something that was going to disappear unused anyway, why sweat it when a little bit gets lost.

If you are using an older version of BOINC, you might consider upgrading.

---

> I fully agree with a "dead responder" limiter, but there should be some
> criteria where the reason should be known.
> Thanks,
> John
>
YOu are not a "WU" trasher, my friend. I went through all your results and up until you reset seti at the end of March, you'd been a PERFECT cruncher. I wish I had that record.

On April first(immediately after the reset) you downloaded 6 wus. On April second you downloaded 7, and on April third you downloaded 9. If this trend continues you should have no problem in a couple days.

tony

[edit] I only looked at your big machine [end edit]

> YOu are not a "WU" trasher, my friend. I went through all your results and up
> until you reset seti at the end of March, you'd been a PERFECT cruncher. I
> wish I had that record.
>
> On April first(immediately after the reset) you downloaded 6 wus. On April
> second you downloaded 7, and on April third you downloaded 9. If this trend
> continues you should have no problem in a couple days.
>
> tony
>

Hiya Tony, hope you're having a good day.

If i understand John correctly, he is currently under a 4 WU restriction which limits the number he can download and crunch at any given time. To get those extra work units that his account shows, he is having to babysit his machine and manually ask for more units throughout the day as he finishes crunching the previous ones. I can see how this could be frustrating, but i have no suggestions but to just hang in there. Would hate to lose a valuable cruncher. :)

Dig

---

> Hiya Tony, hope you're having a good day.

I'd be having a better one is my puppy would let me get more than 1/2 hour nap. I've tried 3 times. Each time... Whine....Lick..lick. then I get up and she goes to sleep.
>
> If i understand John correctly, he is currently under a 4 WU restriction which
> limits the number he can download and crunch at any given time. To get those
> extra work units that his account shows, he is having to babysit his machine
> and manually ask for more units throughout the day as he finishes crunching
> the previous ones.

I hadn't remembered he typed that, thanks. I was under the impression that the limit applied to a 24 hour period, If it were 4 then it should limit out in successive attempts.

> I can see how this could be frustrating, but i have no
> suggestions but to just hang in there. Would hate to lose a valuable
> cruncher. :)
>
> Dig
>
thanks Dig
PS. if he quits, I hope he give me his puters

Or he can detach and reattach using my account key....LOL.. then he'd get more work

> If i understand John correctly, he is currently under a 4 WU restriction which
> limits the number he can download and crunch at any given time. To get those
> extra work units that his account shows, he is having to babysit his machine...

Digger,

This is the part that fascinates me: you say he has to babysit the machine.

... and I don't see why he has to do that.

I very much agree with you that we don't want to lose valuable crunchers but I don't think crunching is such a scarce resource that people (who are a valuable resource) must monitor their machines and coddle them into doing something like picking up work.

If he just leaves it alone, the problem will work itself out.

... or he could add another project and let 'em both crunch. Put SETI at 90% and E@H at 10% and give SETI the lion's share -- if he wishes.

But babysitting the machine? Life is too short to get this upset over spilled clock-cycles.

---