Web code update

We (finally!) updated our web code. Don't hesitate to comment on this post in the "News" forum if you find something broken.

Comments

DanNeely
DanNeely
Joined: 4 Sep 05
Posts: 1364
Credit: 3562358667
RAC: 1580

The server status page lots

The server status page lots its layout/formatting and is showing data in a single column of tables.

Yin Gang
Yin Gang
Joined: 23 Feb 05
Posts: 52
Credit: 120187750
RAC: 0

The 'Community - Languages'

The 'Community - Languages' link is broken:

"../languages/compiled/" is not a directory. Please consult the documentation for correctly setting up the translation system.

Welcome To Team China!

Bernd Machenschalk
Bernd Machenschalk
Moderator
Administrator
Joined: 15 Oct 04
Posts: 4265
Credit: 244921706
RAC: 16859

This is a transient problem

Message 98333 in response to message 98331

This is a transient problem that occurs when the DB queries the page makes take longer than the PHP timeout. The pages ends up incomplete in the cache.

This is a problem not introduced by the web update, we'll work on improving the server status page soon anyway.

BM

BM

Bernd Machenschalk
Bernd Machenschalk
Moderator
Administrator
Joined: 15 Oct 04
Posts: 4265
Credit: 244921706
RAC: 16859

RE: "../languages/compiled/

Message 98334 in response to message 98332

Quote:
"../languages/compiled/" is not a directory.

The directory was missing. I crated it. Could you try again?

BM

BM

Gundolf Jahn
Gundolf Jahn
Joined: 1 Mar 05
Posts: 1079
Credit: 341280
RAC: 0

RE: RE: "../languages/com

Message 98335 in response to message 98334

Quote:
Quote:
"../languages/compiled/" is not a directory.

The directory was missing. I crated it. Could you try again?

BM


The link works (now), but there's only one entry to select:
Use browser language setting
and there's a "foo" line beneath "Language selection".

Gruß,
Gundolf

Computer sind nicht alles im Leben. (Kleiner Scherz)

Bernd Machenschalk
Bernd Machenschalk
Moderator
Administrator
Joined: 15 Oct 04
Posts: 4265
Credit: 244921706
RAC: 16859

I ran

I ran "update_translations.php". Does this work now?

BM

BM

DanNeely
DanNeely
Joined: 4 Sep 05
Posts: 1364
Credit: 3562358667
RAC: 1580

Not sure if it's a bug or

Not sure if it's a bug or not; but comment links typically indicate the number of comments people have made.

Gundolf Jahn
Gundolf Jahn
Joined: 1 Mar 05
Posts: 1079
Credit: 341280
RAC: 0

RE: Does this work

Message 98338 in response to message 98336

Quote:
Does this work now?


Yep, it not only shows the list, it even switches languages ;-) (at least with IE6)

Gruß,
Gundolf

Computer sind nicht alles im Leben. (Kleiner Scherz)

Bernd Machenschalk
Bernd Machenschalk
Moderator
Administrator
Joined: 15 Oct 04
Posts: 4265
Credit: 244921706
RAC: 16859

RE: Not sure if it's a bug

Message 98339 in response to message 98337

Quote:
Not sure if it's a bug or not; but comment links typically indicate the number of comments people have made.

Is there any other BOINC project which does show this?

BM

BM

MarkJ
MarkJ
Joined: 28 Feb 08
Posts: 437
Credit: 137324514
RAC: 21360

On the list of computers

On the list of computers under my account there is a column on the right labelled "last contact". This seems to have a link, but when I click on it it tried to open http://einstein.phys.uwm.edu/sched_logs/2010-06-09_10/2010-06-09_10:44.txt which isn't found. There is a similar link on the computers detail page.

On Seti and GPUgrid this column isn't a link but just a date/time, that is you can't click on it.

The RSS feed for the news on the home page seems to be as of the 31st (last item is the running out of space). I'm not sure if is meant to update automatically or not.

Thanks for the server s/w update.

Bernd Machenschalk
Bernd Machenschalk
Moderator
Administrator
Joined: 15 Oct 04
Posts: 4265
Credit: 244921706
RAC: 16859

The "last contact" link is a

The "last contact" link is a speciality of Einstein@home. It used to show the scheduler log of the last contact of the machine. This feature has been disabled due to the high I/O load it caused on the server. We're working on a replacement.

BM

BM

Richard Haselgrove
Richard Haselgrove
Joined: 10 Dec 05
Posts: 2139
Credit: 2752648717
RAC: 1486471

The project preferences page

The project preferences page is suffering from a translation bug which also afflicted CPDN when they updated their web code last week. Seems like it's probably in the shared translation files.

This is the same (second) iten on the Einstein preference page, in German and then in English:

Quote:

Ausführung auf Grafikkarte anhalten wenn der Computer benutzt wird?
Unterstützt ab Version 6.7 aufwärts ja

Use CPU
Enforced by version 6.10+ yes

Bernd Machenschalk
Bernd Machenschalk
Moderator
Administrator
Joined: 15 Oct 04
Posts: 4265
Credit: 244921706
RAC: 16859

Hm. I fixed the German

Message 98343 in response to message 98342

Hm. I fixed the German translation file (locally) as much as I could, should be better now. But I'm not familiar enough with the other languages, and with the translation service in general to fix this in BOINC.

BM

BM

Stranger7777
Stranger7777
Joined: 17 Mar 05
Posts: 436
Credit: 417491705
RAC: 33499

I'm ready to translate

I'm ready to translate anything into Russian. One of my specialities is a translator of technical papers. So, I'm ready help :)

astro-marwil
astro-marwil
Joined: 28 May 05
Posts: 511
Credit: 402300833
RAC: 1070889

At me ist works fine, as far

At me ist works fine, as far I can see now.
But I was hoping the page "All Tasks for "user"" become further improved, so that one can sort the lines in regard to the different collums, as it can be done in BOINC Statistics. Until now they are fixed sorted in regard to the Wokunit ID. That would be helpful for my own statistics.
For the "Server Status Page" I like to suggest, to sort the fields in the second collumn by actuallity. At now the "S5GC1 search progress" should become the top field, and "Previous searches" the lowest one.

Kind regards
Martin

Richard Haselgrove
Richard Haselgrove
Joined: 10 Dec 05
Posts: 2139
Credit: 2752648717
RAC: 1486471

Most of the BOINC projects

Most of the BOINC projects which have updated their web pages recently have a standard page footer:

[pre]Home | My Account | Message Boards[/pre]
which is a useful expansion on the "Return to Einstein@Home main page" which we've inherited here.

Milwaukee Institute
Milwaukee Institute
Joined: 1 Dec 08
Posts: 1
Credit: 61183141
RAC: 0

Where did the top

Where did the top participants link go?

Thanks.

Jon

Jord
Joined: 26 Jan 05
Posts: 2952
Credit: 5779100
RAC: 0

RE: Where did the top

Message 98348 in response to message 98347

Quote:
Where did the top participants link go?


It's under Statistics, which is on the front page.

mo.v
mo.v
Joined: 9 Jun 10
Posts: 13
Credit: 132471785
RAC: 0

Bernd, thank you for

Bernd, thank you for correcting the errors in the German translation of the Computing and Project preferences of members' accounts.

CPDN upgraded its Boinc version last week and has the same problem with three items in the German translation. I have listed your changes and asked Milo to change the same items for CPDN using your translations.

I've also posted on the boinc_loc email list to say that the Boinc generic German translation file needs to be updated by a volunteer.

Bernd Machenschalk
Bernd Machenschalk
Moderator
Administrator
Joined: 15 Oct 04
Posts: 4265
Credit: 244921706
RAC: 16859

RE: But I was hoping the

Message 98350 in response to message 98345

Quote:
But I was hoping the page "All Tasks for "user"" become further improved, so that one can sort the lines in regard to the different collums, as it can be done in BOINC Statistics.

Well, we had enough to do bringing the web code up to date with current BOINC. We don't have much time to drive the development on the web code further that what BOINC currently offers.

There's one feature I added, though, that's currently hidden. If you add "&show_hosts=1" to the URL of the "Tasks" page, you'll see the hostids (with links), too. For me and probably others with a lot of hosts it's useful to see where e.g. the client errors do happen, but there might be a good reason why this is not BOINC standard. So for now I leave it that way.

BM

BM

Jord
Joined: 26 Jan 05
Posts: 2952
Credit: 5779100
RAC: 0

What I miss and what I find

What I miss and what I find very useful at other projects is the direct links for "Home | My Account | Message Boards" at the bottom of each page.

(Some have "Home | Project | My Account | Message Boards", but they use a Drupal front).

Bernd Machenschalk
Bernd Machenschalk
Moderator
Administrator
Joined: 15 Oct 04
Posts: 4265
Credit: 244921706
RAC: 16859

Like this? BM

Message 98352 in response to message 98351

Like this?

BM

BM

Richard Haselgrove
Richard Haselgrove
Joined: 10 Dec 05
Posts: 2139
Credit: 2752648717
RAC: 1486471

RE: Like

Message 98353 in response to message 98352

Quote:

Like this?

BM


That's the one! Thumbs up!

Major
Major
Joined: 22 May 06
Posts: 6
Credit: 111637245
RAC: 0

RE: Like this? BM yeah

Quote:

Like this?

BM


yeah ... but ...

Quote:
zu Hause · Ihr Konto · Diskussionsforen


hmmm ... the translation is funny ;-)

Jord
Joined: 26 Jan 05
Posts: 2952
Credit: 5779100
RAC: 0

RE: Like this? That's

Message 98355 in response to message 98352

Quote:
Like this?


That's the spirit. Thanks Bernd. :-)

astro-marwil
astro-marwil
Joined: 28 May 05
Posts: 511
Credit: 402300833
RAC: 1070889

Conc.: ABPS Search Progress

Conc.: ABPS Search Progress Page

Up to the Jul 27th one could calculate the "Fraction completet" by (Time pocessed)/(Total survey observation time). But yesterday morning it jumped up to nearly dubble that value, from 23.74% to 47.20%. What´s the new reference?

There are two more breaks concerning "Average data taking rate" and "Percentage of data taking rate". "Progress rate today" is ok. I hope to find time later today to discribe this, as it is a bit more complicated.

Martin

Benjamin Knispel
Benjamin Knispel
Joined: 1 Jun 06
Posts: 148
Credit: 4981579
RAC: 0

Hi Martin, until recently

Message 98357 in response to message 98356

Hi Martin,

until recently we used an imprecise estimate of how much work was done for the ABP search. For this I pulled the information of how much telescope time was used for the survey with the Arecibo telescope from an official website. I compared this to the amount of observation time equivalent the Einstein@Home clients had already processed. Dividing the latter by the first gave then an estimate of how much was processed.

This however did not take into account how much of the total telescope time was spent in observing mode. The telescope has to slew from sky position to sky position and there's also calibration time at the beginning of the observation.

The new reference now fixes that. We now have a complete list of all observations done that qualify for processing by Einstein@Home. Each sky position pointing with Arecibo yields a set of seven "beams" (because the receiver / radio camera has seven feed horns: it's a seven-pixel camera). Now, the number of beams processed is compared to the total number of beams available. Accordingly the stats in the table for the daily amount of work done are computed in "beams/day".

"Average data taking rate" is now also in beams per day; it's computed by dividing the number of beams collected in total by the stretch of time over which they were acquired. And since the old estimate didn't take into account slew/calibration time, the "Percentage of data taking rate" also goes up.

Hope that helps,
Ben

 

Einstein@Home Project

astro-marwil
astro-marwil
Joined: 28 May 05
Posts: 511
Credit: 402300833
RAC: 1070889

Hallo Benjamin! Thanks for

Message 98358 in response to message 98357

Hallo Benjamin!
Thanks for your answer!
My question is: Is it possible to link the data published until yesterday morning with the data since yesterday midday?
Yesterday(UTC)____________9:40______ ________12:45
Average data taking rate ____44,13[min/day] ______48,80[beams/day]
Progress rate today ________41,78[min/day] ______65,36[beams/day]
( The data for ABPS become once a day in the morning updated. Unfortunately there is no timestamp, from when the date are taken.)

If you take (Progress rate today) 65,36/41,78 = 1,5644[beams/min]. This is very accurate the value one can derive from the data published until yesterday morning: 1,5645[beams/min] = 38,350[sec/beam]. Is this just luck ???

The factor for "Average data taking rate" is clearly smaller 48,80/44,13 = 1,1058[beams/min]. This explains, why the "Percentage of data taking rate" jumped from 94,67% up to 133,93%.
Are this factors further and/or backward valid?

Martin

Benjamin Knispel
Benjamin Knispel
Joined: 1 Jun 06
Posts: 148
Credit: 4981579
RAC: 0

RE: If you take (Progress

Message 98359 in response to message 98358

Quote:
If you take (Progress rate today) 65,36/41,78 = 1,5644[beams/min]. This is very accurate the value one can derive from the data published until yesterday morning: 1,5645[beams/min] = 38,350[sec/beam]. Is this just luck ???

Nope, this is related. One beam is 268.5 seconds long.

Quote:

The factor for "Average data taking rate" is clearly smaller 48,80/44,13 = 1,1058[beams/min]. This explains, why the "Percentage of data taking rate" jumped from 94,67% up to 133,93%.
Are this factors further and/or backward valid?

For data taking rate there are some other factors that have to be taken into account, so there's no backwards compatibility here. But our webpage's history is exact all the way back. I will update the webpage soon to show the progress on the anti-center search correctly.

Cheers, Ben

 

Einstein@Home Project

Benjamin Knispel
Benjamin Knispel
Joined: 1 Jun 06
Posts: 148
Credit: 4981579
RAC: 0

Hi all, I've updated our

Message 98360 in response to message 98359

Hi all,

I've updated our progress page. This now also includes the progress from the search on "anti-center" data. These are pointings with the Arecibo telescope that aim away from the Galactic center in contrast to all the observations we've processed so far. Therefore, there's also a new set of plots showing the sky region covered by the anti-center search and our progress in color code.

While I was updating the page I've also made all the numbers on the progress table based on the number of beams processed, getting rid of all time equivalents. One Arecibo pointing consists of seven beams (the seven feed horns of the receiver / radio camera). Each grey point in the sky maps is one pointing (a set of these seven beams).

Hope you like it & happy crunching!
Benjamin

 

Einstein@Home Project