Panic Mode On (34) Server Problems

Message boards : Number crunching : Panic Mode On (34) Server Problems
Message board moderation

To post messages, you must log in.

Previous · 1 . . . 6 · 7 · 8 · 9 · 10 · 11 · Next

AuthorMessage
Profile zoom3+1=4
Volunteer tester
Avatar

Send message
Joined: 30 Nov 03
Posts: 64983
Credit: 55,293,173
RAC: 49
United States
Message 1009105 - Posted: 28 Jun 2010, 4:59:27 UTC - in response to Message 1009102.  

Oh no, starting to see weird connection notices and the boards are on and off...or my internet is going down....

I think It might be Your connection hiamps or at least somewhere between You and Seti@Home, Good Luck.
The T1 Trust, PRR T1 Class 4-4-4-4 #5550, 1 of America's First HST's
ID: 1009105 · Report as offensive
Profile hiamps
Volunteer tester
Avatar

Send message
Joined: 23 May 99
Posts: 4292
Credit: 72,971,319
RAC: 0
United States
Message 1009107 - Posted: 28 Jun 2010, 5:01:07 UTC - in response to Message 1009105.  

Oh no, starting to see weird connection notices and the boards are on and off...or my internet is going down....

I think It might be Your connection hiamps or at least somewhere between You and Seti@Home, Good Luck.

Lets hope....

Official Abuser of Boinc Buttons...
And no good credit hound!
ID: 1009107 · Report as offensive
kittyman Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 9 Jul 00
Posts: 51468
Credit: 1,018,363,574
RAC: 1,004
United States
Message 1009675 - Posted: 29 Jun 2010, 7:24:45 UTC

Ooops....
Just checked the cricket graph, and it looks like outbound traffic has gone away.
"Freedom is just Chaos, with better lighting." Alan Dean Foster

ID: 1009675 · Report as offensive
Profile Uli
Volunteer tester
Avatar

Send message
Joined: 6 Feb 00
Posts: 10923
Credit: 5,996,015
RAC: 1
Germany
Message 1009676 - Posted: 29 Jun 2010, 7:40:44 UTC

Mark go to bed
Pluto will always be a planet to me.

Seti Ambassador
Not to late to order an Anni Shirt
ID: 1009676 · Report as offensive
Terror Australis
Volunteer tester

Send message
Joined: 14 Feb 04
Posts: 1817
Credit: 262,693,308
RAC: 44
Australia
Message 1009685 - Posted: 29 Jun 2010, 10:30:53 UTC

[stamp, pout] Not fair - they went down 20 hours early on the downloads and are now off altogether [/stamp, pout]

:-)

ID: 1009685 · Report as offensive
Profile Miep
Volunteer moderator
Avatar

Send message
Joined: 23 Jul 99
Posts: 2412
Credit: 351,996
RAC: 0
Message 1009699 - Posted: 29 Jun 2010, 11:26:51 UTC - in response to Message 1009689.  

No, they went down exactly at midnight Tuesday morning Berkeley time, if you look at the server graphs. That's not 20 hours early, that's right on schedule.

However they're not off altogether. I can upload work as well as check all web pages


If you look closely outbound went down just a tad before the dateline.
Server status is frozen at 4:30 UTC.

As of 10:30 UTC upload, report and validate was still working (for me).

Looks rather like something in the download pipe went AWOL. In praxi doesn't make a difference - no downloads and unlikely to be fixed and turned on again before the outage.

In any case it's only 8 hours early ;)
Carola
-------
I'm multilingual - I can misunderstand people in several languages!
ID: 1009699 · Report as offensive
Terror Australis
Volunteer tester

Send message
Joined: 14 Feb 04
Posts: 1817
Credit: 262,693,308
RAC: 44
Australia
Message 1009721 - Posted: 29 Jun 2010, 12:50:29 UTC - in response to Message 1009689.  

No, they went down exactly at midnight Tuesday morning Berkeley time, if you look at the server graphs. That's not 20 hours early, that's right on schedule.

So you think someone's in there doing science at midnight. Now that's enthusiasm. Allowing time for a morning coffee, I would have thought somewhere between 8am and 9am Berkeley time would have been the start time. The 20 hours (ok 18 to be precise) is from the time downloads disappeared at around 1400 PST Monday till then.

BTW I'm only being a smart alec, afterall this is a notified planned outage and I have WCG units on standby :-)
ID: 1009721 · Report as offensive
Profile Bill Walker
Avatar

Send message
Joined: 4 Sep 99
Posts: 3868
Credit: 2,697,267
RAC: 0
Canada
Message 1009730 - Posted: 29 Jun 2010, 13:16:30 UTC - in response to Message 1009721.  

So you think someone's in there doing science at midnight. Now that's enthusiasm.


Maybe not someone, but something. Freed up CPU cycles could start on other things, like the big backlog waiting to assimilated, even without a human there.

ID: 1009730 · Report as offensive
Cosmic_Ocean
Avatar

Send message
Joined: 23 Dec 00
Posts: 3027
Credit: 13,516,867
RAC: 13
United States
Message 1009751 - Posted: 29 Jun 2010, 14:24:48 UTC

Two AP splitters were shut off early Monday, leaving just one. It has made the odds of picking up an AP three times more difficult. I'm definitely going to run out of my AP-only cache now. Would have been fine if the splitters stayed up all of Monday.

[/semi-sarcasm]
Linux laptop:
record uptime: 1511d 20h 19m (ended due to the power brick giving-up)
ID: 1009751 · Report as offensive
Josef W. Segur
Volunteer developer
Volunteer tester

Send message
Joined: 30 Oct 99
Posts: 4504
Credit: 1,414,761
RAC: 0
United States
Message 1009753 - Posted: 29 Jun 2010, 14:30:27 UTC
Last modified: 29 Jun 2010, 14:36:44 UTC

At the time of the last server status page update there were 74709 "Results ready to send" and 4 mb_splitter processes running. The ready to send by itself would have sustained about 25Mb/s download until midnight, letting the splitter processes continue to the end of the channels they were working on wouldn't quite have provided enough additional tasks to give the near 50Mb/s which actually flowed. But with one ap_splitter running maybe there would have been enough. Still, I would estimate that if the rundown of ready to send was intentional they probably didn't inhibit starting on new channels until 10 or 10:30 pm PDT. If it was done that way, running out slightly before midnight makes sense because it wouldn't be possible to predict accurately in advance the total work requests which would come in.

Although there is code in the mb_splitters which is intended to allow starting and stopping within a channel, when it was tried at SETI Beta the new splitter process often didn't get the right starting point and produced redundant WUs from an earlier part of the channel. Since then all mb_splitter runs at Beta have been full channels. The ap_splitters at Beta do produce partial channels, usually about 100 WUs at a time.
                                                             Joe
ID: 1009753 · Report as offensive
Josef W. Segur
Volunteer developer
Volunteer tester

Send message
Joined: 30 Oct 99
Posts: 4504
Credit: 1,414,761
RAC: 0
United States
Message 1010191 - Posted: 1 Jul 2010, 2:53:53 UTC

Forewarned is forearmed, so I took a look at the Arecibo observing schedule for the "tapes" from the 3rd, 4th, and 5th of November 2009. (Those ready to be split as shown on Server status). The P2030 PALFA project was allocated a couple of 3 hour observing blocks, and produces mostly VLARs. But the A2010 project had a couple of 6 hour observing blocks, and produces midrange work exclusively (except possibly during setup or teardown at the ends of the blocks). The number of "tapes" isn't sufficient to cover as much time as was scheduled for those two projects.

I can't really predict what we'll get, those schedules are made up several weeks before the observing time and don't necessarily show what was actually done. For instance, work I got from 03no09aa and 03no09ab was recorded late in that day during a period originally scheduled for x-band radar work. The 03no09ac through 03no09ae "tapes" may all be from the next day.
                                                                 Joe
ID: 1010191 · Report as offensive
Roundel

Send message
Joined: 1 Feb 06
Posts: 21
Credit: 6,850,211
RAC: 0
United States
Message 1010203 - Posted: 1 Jul 2010, 4:27:58 UTC

So... I noticed on the front page that the weekly tuesday outage takes 3 days... was that a slip or have the outage times been extended quite a bit.
I'm unable to upload.... and checking the servers, it seems that everything is disabled. Did I miss something?
ID: 1010203 · Report as offensive
Profile zoom3+1=4
Volunteer tester
Avatar

Send message
Joined: 30 Nov 03
Posts: 64983
Credit: 55,293,173
RAC: 49
United States
Message 1010208 - Posted: 1 Jul 2010, 4:52:32 UTC - in response to Message 1010203.  
Last modified: 1 Jul 2010, 4:55:16 UTC

So... I noticed on the front page that the weekly tuesday outage takes 3 days... was that a slip or have the outage times been extended quite a bit.
I'm unable to upload.... and checking the servers, it seems that everything is disabled. Did I miss something?

It is No slip, This is the new norm(3 days), 3-4 hours(or so) was the old one, Now their analyzing signals, So they need uninterrupted access to the database, So there are No uploads or downloads from Tuesday to Thursday, On Friday upload and downloads should resume, At least the Forums are up most of the time.
The T1 Trust, PRR T1 Class 4-4-4-4 #5550, 1 of America's First HST's
ID: 1010208 · Report as offensive
Profile hiamps
Volunteer tester
Avatar

Send message
Joined: 23 May 99
Posts: 4292
Credit: 72,971,319
RAC: 0
United States
Message 1010210 - Posted: 1 Jul 2010, 4:55:14 UTC - in response to Message 1010203.  

So... I noticed on the front page that the weekly tuesday outage takes 3 days... was that a slip or have the outage times been extended quite a bit.
I'm unable to upload.... and checking the servers, it seems that everything is disabled. Did I miss something?

It's all true, 3 day is the norm from now till ?
Official Abuser of Boinc Buttons...
And no good credit hound!
ID: 1010210 · Report as offensive
Profile zoom3+1=4
Volunteer tester
Avatar

Send message
Joined: 30 Nov 03
Posts: 64983
Credit: 55,293,173
RAC: 49
United States
Message 1010211 - Posted: 1 Jul 2010, 4:56:58 UTC - in response to Message 1010210.  

So... I noticed on the front page that the weekly tuesday outage takes 3 days... was that a slip or have the outage times been extended quite a bit.
I'm unable to upload.... and checking the servers, it seems that everything is disabled. Did I miss something?

It's all true, 3 day is the norm from now till ?

Until someone donates $100,000.00 to Seti@Home, Until then I suppose this will be normal here.
The T1 Trust, PRR T1 Class 4-4-4-4 #5550, 1 of America's First HST's
ID: 1010211 · Report as offensive
Profile hiamps
Volunteer tester
Avatar

Send message
Joined: 23 May 99
Posts: 4292
Credit: 72,971,319
RAC: 0
United States
Message 1010214 - Posted: 1 Jul 2010, 5:00:39 UTC - in response to Message 1010211.  

So... I noticed on the front page that the weekly tuesday outage takes 3 days... was that a slip or have the outage times been extended quite a bit.
I'm unable to upload.... and checking the servers, it seems that everything is disabled. Did I miss something?

It's all true, 3 day is the norm from now till ?

Until someone donates $100,000.00 to Seti@Home, Until then I suppose this will be normal here.

Out of that much it would be interesting to know how much would actually channel down thru Berkeley...
Official Abuser of Boinc Buttons...
And no good credit hound!
ID: 1010214 · Report as offensive
Profile zoom3+1=4
Volunteer tester
Avatar

Send message
Joined: 30 Nov 03
Posts: 64983
Credit: 55,293,173
RAC: 49
United States
Message 1010216 - Posted: 1 Jul 2010, 5:04:54 UTC - in response to Message 1010214.  

So... I noticed on the front page that the weekly tuesday outage takes 3 days... was that a slip or have the outage times been extended quite a bit.
I'm unable to upload.... and checking the servers, it seems that everything is disabled. Did I miss something?

It's all true, 3 day is the norm from now till ?

Until someone donates $100,000.00 to Seti@Home, Until then I suppose this will be normal here.

Out of that much it would be interesting to know how much would actually channel down thru Berkeley...

Yes It would indeed.
The T1 Trust, PRR T1 Class 4-4-4-4 #5550, 1 of America's First HST's
ID: 1010216 · Report as offensive
Josef W. Segur
Volunteer developer
Volunteer tester

Send message
Joined: 30 Oct 99
Posts: 4504
Credit: 1,414,761
RAC: 0
United States
Message 1010224 - Posted: 1 Jul 2010, 6:03:12 UTC - in response to Message 1010214.  

So... I noticed on the front page that the weekly tuesday outage takes 3 days... was that a slip or have the outage times been extended quite a bit.
I'm unable to upload.... and checking the servers, it seems that everything is disabled. Did I miss something?

It's all true, 3 day is the norm from now till ?

Until someone donates $100,000.00 to Seti@Home, Until then I suppose this will be normal here.

Out of that much it would be interesting to know how much would actually channel down thru Berkeley...

I have no reason to doubt the FAQ on the Donate to SETI@home page:
Q: How much of my donation actually goes to you?
A: Over 89% of your donation goes directly to SETI@home; the rest goes to the University of California, and pays for our office space, server room, and electricity.

I'm really glad they have electricity to run those servers, any of you California residents care to estimate the monthly cost? Space for the servers and someplace for the staff to sit is nice too.
                                                                Joe

ID: 1010224 · Report as offensive
FiveHamlet
Avatar

Send message
Joined: 5 Oct 99
Posts: 783
Credit: 32,638,578
RAC: 0
United Kingdom
Message 1010232 - Posted: 1 Jul 2010, 6:34:41 UTC

Rig on a Bench has no work left and has earned a rest.
Maybe this room will be cooler for a while.

Dave
ID: 1010232 · Report as offensive
Roundel

Send message
Joined: 1 Feb 06
Posts: 21
Credit: 6,850,211
RAC: 0
United States
Message 1010364 - Posted: 1 Jul 2010, 17:33:21 UTC

I hate to complain... I don't like complainers... But I see this as a major issue. It would seem that there maybe alot of trouble actually getting uploads and downloads during the times the servers are available... almost like what happens after an extended outage, the servers are bombarded and it takes alot of time for things to calm down. It would seem now that every weekend is now a race to get as much resources exchanged as possible...
ID: 1010364 · Report as offensive
Previous · 1 . . . 6 · 7 · 8 · 9 · 10 · 11 · Next

Message boards : Number crunching : Panic Mode On (34) Server Problems


 
©2022 University of California
 
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.