No schedulers responded

Contrablue
Contrablue
Joined: 29 Apr 05
Posts: 2
Credit: 3004
RAC: 0

I have a number of machines

I have a number of machines that have "constipation" from Einstein@home failing to respond to report uploads. Looking over my message logs, the problem first developed at 11:28am PST (I'm in Los Angeles) April 29 (7:28 pm GMT).

I checked the project website, and somewhat unsurprisingly, there appears to have been an attempt to bring up another Einstein@home server in Glasgow at about the time I and other volunteers started reporting server communication problems.

Could the "cutover" to the Glasgow site have been botched somehow? A mind (or should I say, a teraflop or so of computer capacity) is a terrible thing to waste...

Seriously, can somebody let the people all over the world who are volunteering on this know what the heck is going on???

Contrablue

Contrablue
Contrablue
Joined: 29 Apr 05
Posts: 2
Credit: 3004
RAC: 0

Oh yea, If we "little

Oh yea,

If we "little people" need to do something to our machines to get up and running again, could you post the solution on the PROJECT WEBSITE in the NEWS/ANNOUNCEMENT section? This is where, I think, most people will look for information -- and it's where I spotted the item concerning the Glasgow cutover.

Thanks!

Gary Roberts
Gary Roberts
Moderator
Joined: 9 Feb 05
Posts: 5872
Credit: 117640426160
RAC: 35131833

> If we "little people" need

Message 11221 in response to message 11220

> If we "little people" need to do something to our machines to get up and
> running again, could you post the solution on the PROJECT WEBSITE......

I'm sure they will once someone actually realises that there has been a problem. They must all be spending the weekend in the Bahamas or somewhere. Unfortunately, for now, the workaround is probably only known by those dedicated enough to keep a good eye on the message boards.

I've been very pleasantly surprised to see how many first time posters there have been in the last few hours or so :). C'mon guys, keep it up and start contributing some fresh ideas to these lists :).

> the item concerning the Glasgow cutover....

I think that relates to the big data downloads which come from the daemon that distributes the occassional 12 - 14 MB data files. The scheduler is a different daemon which distributes the "intructions" for slicing up the large data files. It's those instructions we're not getting because the scheduler seems hellbent on ignoring us until "prompted" by a preferences update. However it is a big coincidence so you may well be right....

Cheers,
Gary.

PClop
PClop
Joined: 18 Jan 05
Posts: 2
Credit: 194599
RAC: 0

> > 10:40 in Ohio and still

Message 11222 in response to message 11200

> > 10:40 in Ohio and still unable to connect.
> > I've been trying all day and though I was able to upload results the
> update
> > only results in repeated retry scheduling.
>
> Update your preferences on the website first!!!!
>

Thanks it did it ;-)

Peter Koek
Peter Koek
Joined: 4 Mar 05
Posts: 21
Credit: 1067721
RAC: 0

My computer has tried to

My computer has tried to connect for the past 12 hours, with no luck. Someone on this board suggested to updated the preferences, so I did. The next time boinc tried to connect, it succeeded right away. Very strange indeed.

Saenger
Saenger
Joined: 15 Feb 05
Posts: 403
Credit: 33009522
RAC: 0

A toast to

A toast to Rockenstein!

Did the preferences update trick, and everything's fine as it should.

Let's wait 'til monday for explanations on the news site ;)

Grüße vom Sänger

Gary Roberts
Gary Roberts
Moderator
Joined: 9 Feb 05
Posts: 5872
Credit: 117640426160
RAC: 35131833

> Let's wait 'til monday for

Message 11225 in response to message 11224

> Let's wait 'til monday for explanations on the news site ;)

Remember, it is a "once off" fix. If you have a fast turnaround you've gotta keep doing it, particularly if you have a small cache size. Read the whole thread carefully for the full details.

Cheers,
Gary.

Magenta
Magenta
Joined: 8 Mar 05
Posts: 8
Credit: 470619
RAC: 0

> Remember, it is a "once

Message 11226 in response to message 11225

> Remember, it is a "once off" fix. If you have a fast turnaround you've gotta
> keep doing it, particularly if you have a small cache size. Read the whole
> thread carefully for the full details.

Yep, just clicked about that when I did the change, got the file uploaded, then the BOINC manager wouldn't update. Looks like I need to do the change between every connection, dl, ul, and update.

I'm just toggling my disk space limit back and forth.

PClop
PClop
Joined: 18 Jan 05
Posts: 2
Credit: 194599
RAC: 0

> > > 10:40 in Ohio and still

Message 11227 in response to message 11222

> > > 10:40 in Ohio and still unable to connect.
> > > I've been trying all day and though I was able to upload results
> the
> > update
> > > only results in repeated retry scheduling.
> >
> > Update your preferences on the website first!!!!
> >
>

Thanks it did it ;-) A toast to Rockenstein !

Dodge
Dodge
Joined: 26 Apr 05
Posts: 3
Credit: 45916
RAC: 0

Well, I was going to post a

Well, I was going to post a question how to get my BOINC working again, but that doesn't seem necessary anymore. Thank you Rockenstein! I'll be sure to mention this to the members of my team :).

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.