Would it help the project if we turn off network access?


log in

Advanced search

Message boards : Number crunching : Would it help the project if we turn off network access?

Author Message
Profile Kirsten
Volunteer tester
Avatar
Send message
Joined: 7 Jul 00
Posts: 190
Credit: 565,264
RAC: 0
Denmark
Message 13433 - Posted: 3 Aug 2004, 9:23:52 UTC

As it is now the servers must deal with numerous and most often futile upload requests that they cannot handle.

So would it give some 'oxygen' to the project that seems to be gasping for air if we turned of the network access in preferences?

STE\/E [BlackOps]
Volunteer tester
Send message
Joined: 29 Mar 03
Posts: 1137
Credit: 3,040,805
RAC: 109
United States
Message 13441 - Posted: 3 Aug 2004, 10:02:16 UTC

The only thing that would help is if the Dev's turned off the Project for good & put everybody out of their misery. I will continue to Hammer the Site Unmercifully with every computer at my disposal until I get everything uploaded & Reported no matter how much air the site is gasping for...

It was to be expected everybody would be hammering the site once they went back on line, if they were not ready for it then the should have just stayed off line until they where...IMO

Profile Kirsten
Volunteer tester
Avatar
Send message
Joined: 7 Jul 00
Posts: 190
Credit: 565,264
RAC: 0
Denmark
Message 13445 - Posted: 3 Aug 2004, 10:31:14 UTC - in response to Message 13441.
Last modified: 3 Aug 2004, 10:37:21 UTC

> The only thing that would help is if the Dev's turned off the Project for good
> & put everybody out of their misery.

I have read some of your previous posts and I must say I don't give them much attention anymore.

I had hoped for a sober answer from a user not full of I don't know what it is? Hate? Malice?

STE\/E [BlackOps]
Volunteer tester
Send message
Joined: 29 Mar 03
Posts: 1137
Credit: 3,040,805
RAC: 109
United States
Message 13451 - Posted: 3 Aug 2004, 12:20:27 UTC
Last modified: 3 Aug 2004, 15:07:36 UTC

I'm sorry you feel that way Kristen about my posting but I think if you would read all my posts you would not feel that way about me.

I have at times been helpful to people when they needed help. I have also given the Dev's their due when I felt they deserved it, and on occasion even came to their defense when I felt the poster was totally wrong with what they were saying about the Dev's.

I have no Hate or Malice toward the Project & sincerely hope someday it flies like it is supposed to. But you tell me, why is the site even back online. Uploading is painfully slow if even at all. Nobody or very few people can get any work & then it's just 1 or 2 WU's at the most. Just replying to your post has took me over 1 hr to finally get through long enough to attempt replying to you.

There are 40,000 to 50,000 clients out there with many of them having multiple Computers that are out of work & have WU's already processed that they want to Upload & Report before the Deadlines hit them. What did the Project think, everybody would be sleeping when they went back online & they wouldn't get hit so hard...???

All the Project has done is add to the peoples frustration with it by going back online when obviously they were not ready to accept any sort of load yet from the clients.

I can come off as a very blunt person at times because I'm willing to say what a lot of other people would like to say but are afraid of stepping on anyones toes. Well I'm past that age of worrying about who's toes I step on & it's the way I have always been and always will be...If I have stepped on your toes then I apologize but all I can say is go get some steel toed ones and it won't hurt as much the next time... ;)

ric
Volunteer tester
Avatar
Send message
Joined: 16 Jun 03
Posts: 482
Credit: 666,047
RAC: 0
Switzerland
Message 13455 - Posted: 3 Aug 2004, 14:44:38 UTC - in response to Message 13451.
Last modified: 3 Aug 2004, 14:46:59 UTC

yes it would help, if we ALL disable the network connection in the prefs.

If everybody will do this, no need to down/upload, therefore, no
"stress" for berkeley.

A football-player can also be asked not to play football..

Is this the answer U seeking for?

Yes we are a part of the problem(s) because we are so in need to have WU's..


For me it's like a game, level by level must walked trough and each level
has it's own "dangerious holes". At the filan paa of the level, point, sorry credits are spoken (sometimes)


well, we all are parts and members of this "club", we all have the same "rights" to express our feeling knowledge or our sadness or fun, our suffering.

For myself, I enjoy the multiple writtings... *lagest possible smile*


ric



Profile xi3piscium
Volunteer tester
Avatar
Send message
Joined: 17 Aug 99
Posts: 287
Credit: 26,674
RAC: 0
China
Message 13456 - Posted: 3 Aug 2004, 15:58:14 UTC

I have to agree with PoorBoy, they knew the servers
would get hammered, and are still getting hammered as
I write this, what disappoints me most is the lack of
communication we as participants get from the DEV team.
I know we are all volunteering our computers and time,
we chose this project, and there may be more bumps in
the road until the project smoothes out. It is just
frustrating after this outage that things are really
looking bad, but in due course the DEV team will right
the ship. We have to give credit to the DEV team when
they are working diligently to keep the project going
along, but we also must be allowed to speak our minds
when we feel the need to do so when things are going
sour and we hear nothing from the DEV's. I'm not really
one to complain, as so many here are whiners, but I
(speaking for myself only) feel a bit frustrated at this
point. Just hoping things start getting back to normal in
the next few days. No malice intended to anyone...

Xi3

Profile Darth Dogbytes™
Volunteer tester
Send message
Joined: 30 Jul 03
Posts: 7512
Credit: 2,021,148
RAC: 0
United States
Message 13457 - Posted: 3 Aug 2004, 16:04:25 UTC

Although the website is up (although rather slow), it appears that the schedulers/servers are still off line. The system is going to have to get used to some rough treatment sooner or later, if for no better reason than to test its' capabilities and compacity. Personally, I'm staying on line and not disabling the network connection to Boinc. However, do what ever you personally think is best for yourselves. Have a good day...




Alpha Tester
____________
Account frozen...

Profile huns0004
Send message
Joined: 14 Jun 01
Posts: 46
Credit: 1,790,888
RAC: 849
United States
Message 13460 - Posted: 3 Aug 2004, 16:53:42 UTC - in response to Message 13456.

> We have to give credit to the DEV team when
> they are working diligently to keep the project going
> along, but we also must be allowed to speak our minds
> when we feel the need to do so when things are going
> sour and we hear nothing from the DEV's. I'm not really
> one to complain, as so many here are whiners, but I
> (speaking for myself only) feel a bit frustrated at this
> point. Just hoping things start getting back to normal in
> the next few days. No malice intended to anyone...

It would be easier to give the developers more credit if they gave us more information. By constantly hitting "retry now" for the past hour, I've managed to upload four WUs. Why is this happening? What's the bottleneck? What are they doing to try to resolve the problem? Is it a programming issue, a hardware issues, a RTFM issue? At this point, I am beginning to think that they are running the entire project off of the 486 DX2 computer I got over ten years ago.

Yes, everybody will once again say "if the developers take the time to provide updates that's less time for fixing," but there are 50,000 people here who want to know what's going on. Based on how long the recent outage was (from about Wednesday to Sunday), it's hard to picture them working 24/7 to fix things. They obviously have some time to do two things: first, tell us how things are working right now, in terms of hardware and software. Second, as new things happen, post updates. If I know what's going on, I feel less frustrated.

Also, they should get these message boards off the project servers. There is no reason why everything has to go down, including message boards completely unrelated to SETI science, when the "project" is down.

Profile Toby
Volunteer tester
Avatar
Send message
Joined: 26 Oct 00
Posts: 1005
Credit: 5,597,723
RAC: 2,207
United States
Message 13467 - Posted: 3 Aug 2004, 17:18:41 UTC

I'm sure they knew very well that they were going to get hammered. But what could they do? You can't "prepare" for it without better hardware. The main problem it seems is their disk I/O. They ordered some new, beefier hardware for their database server but the supplier messed up and I believe it won't arrive until Friday. What could they do in this case? Just not turn it back on? That would probably piss people off more than it running badly. At least now there is hope of getting work. And I think everyone was able to upload at least a few work units over the past day. I wouldn't be surprised if most of us get a work unit or two by the end of today.

PoorBoy: "The only thing that would help is if the Dev's turned off the Project for good..."

-------------------------------------------
- A member of The Knights Who Say NI!
Possibly the best stats site in the universe:
http://boinc-kwsn.no-ip.info

Profile StokeyBob
Avatar
Send message
Joined: 31 Aug 03
Posts: 848
Credit: 2,218,691
RAC: 0
United States
Message 13473 - Posted: 3 Aug 2004, 17:36:16 UTC - in response to Message 13433.

In this great age of communication SETI isn't th only one with problems.

In the olden days if we wanted to tell someone something we called them on the phone, they would pick it up, you would say your piece and you would both hang up.

Now if you need to tell someone something, our ask a question, you call them on the phone and get a machine. The machine gives you a list of selections to chose from and you push a button. Then the machine gives you a list of selections to chose from and you push a button. Then the machine gives you a list of selections to chose from and you push a button.

I'm old and I often give up. I see some of the young people that don't know any better that have no problem with this. You see them all the time on the freeway going about their lives with phones pasted to their heads.

SETI would not be getting hammered right now if their was someway for someone to just answer the phone.

To get work takes numerous tries. To upload takes numerous tries. Multiply that by thousands and the problem gets astronomical.

ric
Volunteer tester
Avatar
Send message
Joined: 16 Jun 03
Posts: 482
Credit: 666,047
RAC: 0
Switzerland
Message 13476 - Posted: 3 Aug 2004, 17:42:38 UTC - in response to Message 13460.

> By constantly hitting "retry now" for the past hour, I've managed to upload four WUs

are uploads working from time to time?

that makes 4 uploads in 1 hour. Poorboy managed yesterday about 80 Wus in 2 hours. In every case a poor rating.


Still sitting on over 440 Wu's, I can't and I don't will put so much time and efforts to constantly hitting "retry now".

I still let them "auto retry" and keep the hope, "normal" returns
(what ever normal is)

More information will help to all of us, understanding actual problem.

Reading the lot of postings, "asking for more infos" and viewing what was going on in the past 3-4 weeks, I don't believe, they will increase there communication.

>>I'm sure they knew very well that they were going to get hammered. But what could they do? You can't "prepare" for it without better hardware. The main problem it seems is their disk I/O.

maybe wrong, sorry, but they could easy put information about they basic services like what is up or down, what is online/offline and if working on it.
that's (my opinion) not a question of hardware or $$.

I wish all reader, they could upload all their work within the 14 day "delay"

ric



Profile StokeyBob
Avatar
Send message
Joined: 31 Aug 03
Posts: 848
Credit: 2,218,691
RAC: 0
United States
Message 13480 - Posted: 3 Aug 2004, 17:50:54 UTC - in response to Message 13473.

Something I have noticed in the last twenty four hours is that my faster machine seems to be able to upload better.

I don't know much about computers but it seems to be able to beat the time-out better. Or loose them in cyberspace faster.

Pascal, K G
Volunteer tester
Avatar
Send message
Joined: 3 Apr 99
Posts: 2343
Credit: 150,491
RAC: 0
United States
Message 13481 - Posted: 3 Aug 2004, 17:53:04 UTC - in response to Message 13473.

> In this great age of communication SETI isn't th only one with problems.
>
> In the olden days if we wanted to tell someone something we called them on the
> phone, they would pick it up, you would say your piece and you would both hang
> up.
>
> Now if you need to tell someone something, our ask a question, you call them
> on the phone and get a machine. The machine gives you a list of selections to
> chose from and you push a button. Then the machine gives you a list of
> selections to chose from and you push a button. Then the machine gives you a
> list of selections to chose from and you push a button.
>
> I'm old and I often give up. I see some of the young people that don't know
> any better that have no problem with this. You see them all the time on the
> freeway going about their lives with phones pasted to their heads.
>
> SETI would not be getting hammered right now if their was someway for someone
> to just answer the phone.


With you on all points, GREY POWER>>>>>> lol

>
> To get work takes numerous tries. To upload takes numerous tries. Multiply
> that by thousands and the problem gets astronomical.
>
>

Just my opinion, not UCBs, BOINCs or Seti@H or the little old lady down the road.
M7

Profile Qui-Gon
Volunteer tester
Avatar
Send message
Joined: 15 May 99
Posts: 2901
Credit: 6,266,050
RAC: 2,062
United States
Message 13492 - Posted: 3 Aug 2004, 20:17:05 UTC

>By constantly hitting "retry now" for the past hour, I've managed to
>upload four WUs

Is this the right thing to do? WE DON'T KNOW BECAUSE THERE HAS BEEN NO ADVICE FROM BERKELEY!

>Something I have noticed in the last twenty four hours is that my faster
>machine seems to be able to upload better.

>I don't know much about computers but it seems to be able to beat the
>time-out better. Or loose them in cyberspace faster.

Are the few results we manage to (maybe) upload getting through, or are they going to be lost like those we were told about last Thursday? WE DON'T KNOW BECAUSE THERE HAS BEEN NO ADVICE FROM BERKELEY!

>Yes, everybody will once again say "if the developers take the time to
>provide updates that's less time for fixing," but there are 50,000 people
>here who want to know what's going on.

Amen, but WE DON'T KNOW BECAUSE THERE HAS BEEN NO ADVICE FROM BERKELEY!

I find myself complaining (so don't respond to this by telling me something I already know). It is my hope that someone from the project reads this and starts telling us what is going on (thanks for the very short post on the home page), and what we can do.

Profile Paul D. Buck
Volunteer tester
Send message
Joined: 19 Jul 00
Posts: 3898
Credit: 1,158,042
RAC: 0
United States
Message 13496 - Posted: 3 Aug 2004, 20:42:01 UTC - in response to Message 13492.

> >By constantly hitting "retry now" for the past hour, I've managed to
> >upload four WUs
>
> Is this the right thing to do? WE DON'T KNOW BECAUSE THERE HAS BEEN NO ADVICE
> FROM BERKELEY!
>
> >Something I have noticed in the last twenty four hours is that my faster
> >machine seems to be able to upload better.
>
> >I don't know much about computers but it seems to be able to beat the
> >time-out better. Or loose them in cyberspace faster.
>
> Are the few results we manage to (maybe) upload getting through, or are they
> going to be lost like those we were told about last Thursday? WE DON'T KNOW
> BECAUSE THERE HAS BEEN NO ADVICE FROM BERKELEY!
>
> >Yes, everybody will once again say "if the developers take the time to
> >provide updates that's less time for fixing," but there are 50,000 people
> >here who want to know what's going on.
>
> Amen, but WE DON'T KNOW BECAUSE THERE HAS BEEN NO ADVICE FROM BERKELEY!
>
> I find myself complaining (so don't respond to this by telling me something I
> already know). It is my hope that someone from the project reads this and
> starts telling us what is going on (thanks for the very short post on the home
> page), and what we can do.

Ok, here is the advice, though not official, and not from berkeley. Let the system try to connect and if it does not, let the BOINC Software back-off and try after the time interrval. The point of that automatic mechanism is to avoid the "rush" problems where the scarce resource is swamped with requests.

In practical terms it is like rush hour with the weavers (the ones that dart in and out of traffic) making them able to increase their distance traveledl but only at the expense of everyone else on the road. By sending in lots of requests you will get more uploaded, but only for yourself. Everyone else pays with increased delays because the miniscule amount of time it takes to tell you that you cannot be serviced is at the expense of anyone else making a similar request.

Epiration of the work units is going to happen. And you will lose the credit. It is part of the rules as posted on the web site here. You did read them, didn't you?

Your last point is valid, and we can hope. But, don't expect much. Reality is that this is a Beta test still, they just are not calling it that anymore. It is like a Microsoft Operating System. Don't buy it untill service pack one is out. I am fully back to SETI@Home classic in that the BOINC system is not scaling up (which I prdicted back in beta, not that anyone cares) and it is now complicated with the fact that they are now in the bad situation where they now have "legacy" code (and database schema) and changes are hard to make because of simple schema changes can and will break code and oh-by-the-way, requires the migration of data from the "old" system to the new.

And migration of data is a simple and straight-forward task as we have just seen this weekend



Profile K3UZZ
Avatar
Send message
Joined: 29 May 99
Posts: 175
Credit: 67,332
RAC: 0
United States
Message 13503 - Posted: 3 Aug 2004, 21:10:14 UTC - in response to Message 13441.
Last modified: 3 Aug 2004, 21:54:04 UTC

Profile Qui-Gon
Volunteer tester
Avatar
Send message
Joined: 15 May 99
Posts: 2901
Credit: 6,266,050
RAC: 2,062
United States
Message 13512 - Posted: 3 Aug 2004, 21:55:49 UTC - in response to Message 13496.

Thanks Paul. You said:

> Ok, here is the advice, though not official, and not from berkeley. Let the
> system try to connect and if it does not, let the BOINC Software back-off and
> try after the time interrval. The point of that automatic mechanism is to
> avoid the "rush" problems where the scarce resource is swamped with requests.
>
That's pretty much what I have been doing, but it seems that everytime I let the software try to connect, no schedulers respond, and I am deferred for another hour or more. I have had more success re-trying until a unit uploads (I still have lots to go). If this causes others to have difficulty connecting, then the people from Berkeley, not you, should be telling all of us that. Indeed, your assumption may be completely wrong -- the connection difficulties may be something completely different.

> Epiration of the work units is going to happen. And you will lose the
> credit. It is part of the rules as posted on the web site here. You
> did read them, didn't you?
>

Yes, I did read the rules. There was nothing that talked about the loss of credit we experienced last Thursday due to a software problem on the BOINC platform. And if I can't upload my results before they expire through no fault of my own -- because something at Berkeley is not functioning -- then what is the point of doing the work in the first place?

> Your last point is valid, and we can hope. But, don't expect much. Reality
> is that this is a Beta test still, they just are not calling it that anymore.

My last point was that I hope someone from the project reads this and starts telling us what is going on. Whether this project is or is not in "Beta" testing is not the point. The folks at Berkeley should always listen to the problems being posted here, and they should always have the courtesy to keep us informed.

ric
Volunteer tester
Avatar
Send message
Joined: 16 Jun 03
Posts: 482
Credit: 666,047
RAC: 0
Switzerland
Message 13513 - Posted: 3 Aug 2004, 22:04:57 UTC - in response to Message 13512.

update on the home page:

August 3, 2004
While the project is up and functioning normally, the database and web servers are overloaded. Because of this you may have noticed difficulty in transferring work and loading certain web pages. We're tuning these servers and trying to find the bottlenecks now.

Profile xi3piscium
Volunteer tester
Avatar
Send message
Joined: 17 Aug 99
Posts: 287
Credit: 26,674
RAC: 0
China
Message 13524 - Posted: 3 Aug 2004, 23:02:11 UTC - in response to Message 13481.

> > In this great age of communication SETI isn't th only one with problems.
> >
> > In the olden days if we wanted to tell someone something we called them
> on the
> > phone, they would pick it up, you would say your piece and you would both
> hang
> > up.
> >
> > Now if you need to tell someone something, our ask a question, you call
> them
> > on the phone and get a machine. The machine gives you a list of
> selections to
> > chose from and you push a button. Then the machine gives you a list of
> > selections to chose from and you push a button. Then the machine gives
> you a
> > list of selections to chose from and you push a button.

LOL, thnx Pascal, I needed that, best I've felt since the site came
back...

Regards,

Xi3

Message boards : Number crunching : Would it help the project if we turn off network access?

Copyright © 2014 University of California