Uhm, from the front-page:
"The waiting for validation/transition counts on the server status page are disabled until we diagnose a recent DB slowdown."

Some info is also posted to "technical news", also mentioning problems handing out work due to backlogged database, and "hastily getting two new big servers on line"...

> Thank you RichardG,
>
> i just love how someone that has been around for one whole week has the gall
> to say what should and should not be done.
> try lurking for a while until you figure out how things work.
>
> as the old saying goes, "better to be thought a fool than to open your mouth
> and remove all doubt."
>
>
>
I don't know if he can read. There were at least 3 threads saying the scheduler or servers were down. I started one last night commenting on the question of the 1 hour being a correct estimate of the downtime.

He needs some anger management consoling.

---

Red Bull Air Racing

Gas price by zip at Seti

>
> > Thank you RichardG,
> >
> > i just love how someone that has been around for one whole week has the
> gall
> > to say what should and should not be done.
> > try lurking for a while until you figure out how things work.
> >
> > as the old saying goes, "better to be thought a fool than to open your
> mouth
> > and remove all doubt."
> >
> >
> >
> I don't know if he can read. There were at least 3 threads saying the
> scheduler or servers were down. I started one last night commenting on the
> question of the 1 hour being a correct estimate of the downtime.
>
> He needs some anger management consoling.

I loved that movie!

> > > They did have a news report on January 3 saying servers will be down
> for
> > 1
> > > hour for database updates.
> > > After about 6 hours they remove the 1 hour message but still had
> the
> > message
> > > being down for database updates.
> > > When they were back up and running they removed the news meessage
> from
> > Jan. 3,
> > > 2005.
> > > The scheduler is swamped with requests.
> > >
> > Thank you RichardG,
> >
> > i just love how someone that has been around for one whole week has the
> gall
> > to say what should and should not be done.
> > try lurking for a while until you figure out how things work.
>
> I've been a SETI@home user for: 5.324 years
> Registered on: Fri Sep 10 06:26:39 1999 UTC
>
> > as the old saying goes, "better to be thought a fool than to open your
> mouth
> > and remove all doubt."
>
> I'm also a big fan of Dick Cheney... and quite certain you won't get the
> joke.
>
> So, go xxxx yourself. Fool. ;)
>
>
Well by the looks of it you did not activate your old account from Seti Classic. If you did, it wouldn't have shown joined Dec28, 2004.

---

Red Bull Air Racing

Gas price by zip at Seti

> > > > They did have a news report on January 3 saying servers will be
> down
> > for
> > > 1
> > > > hour for database updates.
> > > > After about 6 hours they remove the 1 hour message but still
> had
> > the
> > > message
> > > > being down for database updates.
> > > > When they were back up and running they removed the news
> meessage
> > from
> > > Jan. 3,
> > > > 2005.
> > > > The scheduler is swamped with requests.
> > > >
> > > Thank you RichardG,
> > >
> > > i just love how someone that has been around for one whole week has
> the
> > gall
> > > to say what should and should not be done.
> > > try lurking for a while until you figure out how things work.
> >
> > I've been a SETI@home user for: 5.324 years
> > Registered on: Fri Sep 10 06:26:39 1999 UTC
> >
> > > as the old saying goes, "better to be thought a fool than to open
> your
> > mouth
> > > and remove all doubt."
> >
> > I'm also a big fan of Dick Cheney... and quite certain you won't get the
> > joke.
> >
> > So, go xxxx yourself. Fool. ;)
> >
> >
> Well by the looks of it you did not activate your old account from Seti
> Classic. If you did, it wouldn't have shown joined Dec28, 2004.

We all decided to start fresh with seti/boinc... figured that would help in the competition area because some of us were thousands of WU's ahead of the newer members of the team.

Please don't change the text of my posts in the future. Snip lines but don't change the text. Not everyone is offended by common, everyday, colloquial language.

Thank you and... well you know. ;)

> > I've been a SETI@home user for: 5.324 years
> > Registered on: Fri Sep 10 06:26:39 1999 UTC
> >
> > > as the old saying goes, "better to be thought a fool than to open
> your
> > mouth
> > > and remove all doubt."
> >
> > I'm also a big fan of Dick Cheney... and quite certain you won't get the
> > joke.
> >
> > So, go xxxx yourself. Fool. ;)
> >
> >
> Well by the looks of it you did not activate your old account from Seti
> Classic. If you did, it wouldn't have shown joined Dec28, 2004.
>
well, as he noted, here, you show a week ago.
and i am not a fan of cheney, nor bush, nor blair. But you do speak like Cheney (recall the congress episode), and make presumptions such as his republican stature dictates that when one doesn't agree with you, then they must have an IQ of less than 90 (whereas Bush, who really DOES have an IQ about 90 is a brilliant man...)

anyway, away from the politics since this is science here - and politics never like science nor the truth much anyway.

as for the colloquial language, many of us also read these boards at work, not the best thing i want showing in the log if they decide to look.

---

S@h Berkeley's Staff Friends Club

> We all decided to start fresh with seti/boinc... figured that would help in
> the competition area because some of us were thousands of WU's ahead of the
> newer members of the team.
>
> Please don't change the text of my posts in the future. Snip lines but don't
> change the text. [edit] everybody [edit] is offended by common, everyday, colloquial
> language[ edit] that I use when it is an insult [edit].
>
> Thank you and... well you know. ;)
>
>
The work units are keep seperate anyway.
The language may be common in your neck of the woods, but not here.

---

Red Bull Air Racing

Gas price by zip at Seti

> Please don't change the text of my posts in the future. Snip lines but don't
> change the text. Not everyone is offended by common, everyday, colloquial
> language.

No, not everyone is offended by such language. However, there are many participants on these boards that are quite young, so toning it down might be best.

Also, these boards have generally stayed away from political discussions.

> Also, these boards have generally stayed away from political discussions.

I noticed they avoid religious topics too. Yeah right.

> > Also, these boards have generally stayed away from political
> discussions.
>
> I noticed they avoid religious topics too. Yeah right.
>
>

Well, I should have said on all the boards excluding the cafe.

> > > Also, these boards have generally stayed away from political
> > discussions.
> >
> > I noticed they avoid religious topics too. Yeah right.
>
> Well, I should have said on all the boards excluding the cafe.

I made the mistake of reading a couple posts over there before I posted here. I'm still working on the anger management thing. I kind of side with that old serpent in the garden... he was framed, you know.

The scheduler seems to be responding very good now.

---

Red Bull Air Racing

Gas price by zip at Seti

Sorry to say. But I checked the status and mine is still not working.

---

The BOINC server is still down and I cant get any more work units to Crunch. I can still get SETI Classic work units however. I guess they use a different server. I guess I will Crunch for Classic until they get BOINC back up again.

---

> The BOINC server is still down and I cant get any more work units to Crunch.
> I can still get SETI Classic work units however. I guess they use a different
> server. I guess I will Crunch for Classic until they get BOINC back up again.

Downloaded a few dozen mew workunits to crunch about 45 minutes ago.

---

Grant
Darwin NT

After a long period (the optimistic hour) where the server did not respond I grabed 2 WUs today but the server is again down or not responding. I'm in the UK if that makes a difference.

What I did wonder was whether BOINC projects could force a client to cache extra work prior to any server upgrades or expected outage?

In my case I am crunching one or two WUs cached between server down \ up loads.

This is not optimising the space I have allocated as my SETI BOINC folder stands around 6 MB on average where as I have offered between 100 and 300MB.

Whilst I could change my server updates from 0.1 to 4 or so days I prefer to be available to SETI on a more frequent basis.

I wonder if the SETI server could be more pre-emptive or flexible and use all of the parameters I have accepted which includes HD space available?

Perhaps this is already the case? I don't know..

I do realise I could just decrease the frequency, do an update and then increase the frequency but there is in theory lots of slack space out there where the more frequent updating crunchers are concerned..

I also realise I have lots to learn and may be barking up the wrong tree as I still haven't got my head round the process used by the software.

Are the downloading of data and uploading of results done within the same time slot (for dialup = connection)?

Pete

>What I did wonder was whether BOINC projects could force a client to cache
> extra work prior to any server upgrades or expected outage?

Doesn't do this. Not a good idea either, the user might not want the extra WU's. If running out of WU's is a problem then increase your "connect every...." number to keep more work queued.


> Whilst I could change my server updates from 0.1 to 4 or so days I prefer to
> be available to SETI on a more frequent basis.
>

Up it to .5 and run for awhile. Up it again if you process all those WU's within a day, lower it if it doesn't. Your results show your system is processing a WU in like 5-6 hours so you could keep 4 or so in the queue while processing one and still return them all within a day.

Ideally you could set it to some value and only connect like twice a day - once in the morning to upload results and again before you leave in the evening to get enough work for overnight. And keep enough work so you could miss the evening connection and still have enough queued to run something overnight.

> I wonder if the SETI server could be more pre-emptive or flexible and use all
> of the parameters I have accepted which includes HD space available?
>
> Perhaps this is already the case? I don't know..
>

It is the case. Take a look at BOINC's "Disk" tab and you'll see how its tracking disk usage for BOINC and SETI.


> Are the downloading of data and uploading of results done within the same time
> slot (for dialup = connection)?
>

No. Uploading is independent of the scheduler. Reporting results (CPU time) and downloading new work is done within the same "slot" as they're done by the scheduler.

If your "connect every..." time is small enough (like your .1 days), it'll appear that all three happen together as it'll connect to the scheduler to report on WU's and request more work as soon as you connect (dialup). But if you set it to connect every 4 days then dialing into internet will only upload completed WU's and not report them or get new work.

> What I did wonder was whether BOINC projects could force a client to cache
> extra work prior to any server upgrades or expected outage?
>
> In my case I am crunching one or two WUs cached between server down up
> loads.

Pete,

This is the essence of the "Connect every 'x' days" setting. We are in an "awkward" time in the life of BOINC where we are still in the begining. As an artifact of this point in time is that it is difficult to truly see BOINC working as advertised. There was a point last summer where it was easier to see the flexibility and robustness of the BOINC System. I am of course talking of the time when there were more than two projects live and serving work.

Unfortunatly, we have only the two projects and the ability to get work is compromised slightly with only SETI@Home and cp.net alive. Hopefully when Predictor@Home gets thier current server problems solved, LHC@Home restarts issuing work, and Einstein@Home opens for buisness we will see fewer problems on a day-to-day basis.

My best offer is to suggest a setting of 4 days for your queue size and to try to get into cp.net right now, and on the 10th try to get into Predictor@Home so that you have multiple projects to distribute the load and to try to keep your computer busy. Idle computers are the devil's playground ...

---

> This is the essence of the "Connect every 'x' days" setting. We are in an
> "awkward" time in the life of BOINC where we are still in the begining. As an
> artifact of this point in time is that it is difficult to truly see BOINC
> working as advertised. There was a point last summer where it was easier to
> see the flexibility and robustness of the BOINC System. I am of course talking
> of the time when there were more than two projects live and serving work.

It might also be good to point out that BOINC is __designed__ to tolerate outages well. It buffers work units for future processing, and it holds results until they can be uploaded and reported.

There is no reason to panic if the scheduler is down.

---