• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Seti had another unexpected power-outage...

Page 3 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
Originally posted by: Rattledagger
A slightly different wording in Technical News:
March 5, 2005 - 01:00 UTC
The project is down for the weekend. Although we made some diagnostic progress, the servers are still not able to talk to the UPS's. The power in the building is still not trustworthy. There will probably be a power outage next week so that campus can track this down
Click to expand...
Click to expand...



Oh boy............ My production should be way up this weekend as well... They better get it back up on monday or I will be shooting blanks. 🙁









SHUX
 
Shux, go ahead and redistribute the WUs in my queue(s) (The Wildnet TeAm), if it helps. I have at least 86 WUs available to me (in seti drivers, not a queue), and there's no way I'll get that done before Monday.
 
So I moved my workqueue home, emptied it and I'm dl on my isdn line.

a HUGE THANX goes to soni, Smoke, Shuxclams and Unforgiven for providing me with wu from their queues! I owe you guys bigtime! :beer:😀
 
My queue is still open for anyone that needs it 🙂
Its all setup to move 50wu automatically to a new account, so there should be no problems setting up new users to connect to it 🙂
 
It appears that they shutdown for the weekend. Wonder how long it will take to clear the backlog? I have 241 results to send in.


I was just wondering if this only affects SETI Classic, or is BOINC having the same problem?
 
John, all of SETI is shut down.

I sure hope they get the servers turned on Monday for it is approaching crunch time as far as cached WUs are concerned.

I had about 8,000 stashed away for a rainy day and it appears they have all been lost in one of the system rebuilds. 🙁
 
Originally posted by: JWMiddleton
It appears that they shutdown for the weekend. Wonder how long it will take to clear the backlog? I have 241 results to send in.


I was just wondering if this only affects SETI Classic, or is BOINC having the same problem?
Click to expand...

Yes, BOINC SETI is effected also

Monday is going to suck trying to connect to the servers. Everyone will be slamming them when they turn them on 🙁
Maybe it'll ease up by the time i try to connect later in the day.
 
Originally posted by: JWMiddleton
It appears that they shutdown for the weekend. Wonder how long it will take to clear the backlog? I have 241 results to send in.


I was just wondering if this only affects SETI Classic, or is BOINC having the same problem?
Click to expand...



Results? http://seti.scarieville.com:5517/results.htm


I hate this backlog hoo-haw ~~~~~~ gonna take a while to clear this up.. just like it was all week.







SHUX
 
Originally posted by: Fardringle
I think its the aliens trying to shut down the system before we discover their soap opera broadcasts streaming across deep space. The licensing and royalty costs from that type of network coverage would be staggering! :evil: :laugh:
Click to expand...

LMFAO! 😀 ,they might try to sue earth for illegal sharing! ,I wonder how many trillions of £ 🙂P) that'd be worth 😉

DaFinn
If you'd read the last outage thread then you'd know they do use UPS & they got some more just after the last one 😛 ,that's why they were able to do a 'graceful' shutdown this time 😉 ,also don't forget they are only volunteer funded

Originally posted by: Unforgiven
2 suggestions for the seti folks....

1) invest in ups backups and rtfm.

2) go to a windows operating system because obviously they dont know what the hell they are doing with their equipment and need something a little more basic that they can get a handle on.
Click to expand...

1.They have ,& you might want rtf previous threads😉

2.AFAIK It doesn't seem to be trouble with the OS their having

I think you guys are forgetting that S@H is run by a handful of volunteers ,not to mention they've got their hands full with the transisition to S@H2 (& it is not goverment funded)

I've turned off auto connect on my Q ,will switch it back Tuesday
 
Originally posted by: SmokeI had about 8,000 stashed away for a rainy day and it appears they have all been lost in one of the system rebuilds. 🙁
Click to expand...
I have about 12,000 WU's on DVD that i download a few months ago...
Let me know if you want me to send them over to ya 🙂
 
Originally posted by: Coolkid
Originally posted by: SmokeI had about 8,000 stashed away for a rainy day and it appears they have all been lost in one of the system rebuilds. 🙁
Click to expand...
I have about 12,000 WU's on DVD that i download a few months ago...
Let me know if you want me to send them over to ya 🙂
Click to expand...

If the Servers get up and running tomorrow for a reasonable period of time I think we will come out of this in fine shape. I'm working with Mucman to get a little additional redundancy in our Q fleet. I've parceled out all of the WUs I have leaving about a two day supply for "The Enterprise". Each client should have at least a week so we have a little breathing room left.

If the Seti Servers don't give us a change to catch up by Wednesday, I'll get in touch with you Coolkid about uploading those WUs to some site where I can get access. 😉

DaFinn, we certainly appreciate all of your contributions. Without guys like you we would not have achieved the lofty level we find ourselves today. :thumbsup:
 
Originally posted by: Spacehead
Originally posted by: JWMiddleton
It appears that they shutdown for the weekend. Wonder how long it will take to clear the backlog? I have 241 results to send in.


I was just wondering if this only affects SETI Classic, or is BOINC having the same problem?
Click to expand...

Yes, BOINC SETI is effected also

Monday is going to suck trying to connect to the servers. Everyone will be slamming them when they turn them on 🙁
Maybe it'll ease up by the time i try to connect later in the day.
Click to expand...

It's going to be a joke all week long. I was uploading wu's in the 11th hour during the last shutdown. Hate to say it, but I think I'm going to start moving some crunchers over to FaD next week.

I don't understand why none of the Berk team is committed enough to this project to come in and fire up the servers during the day Sat and Sun. I mean, at least just a few hours of baby sitting..big deal. I know if it was my project I would, especially considering the number of users involved in SETI.

 
I just had a brilliant idea that I tried to post on the S@H message boards, but couldn't. Either the server for that is down, or I didn't have the right password.

Anyway, if someone else who knows how can post it, I'm sure many people would appreciate it:
Subject: How to automate server shutdown

I just had a brilliant idea. The problem you (the admins) are having is detecting, from the UPS, when the power cuts out, right? Well, why do you need the UPS to tell you this?

First, get an old desktop system (CPU doesn't matter; all that matters is the network card). Set it up with a basic Linux system - the simpler the better, as long as it has basic networking. Plug it into the building power but not the UPS, and let it sit there (crunching SETI, perhaps).

The reason I suggest a Linux system is it's less likely to crash on its own than *some other* OSes. The reason I suggest a basic system is so if it does crash, or if there's a short outage the UPS can handle, it can come back up and accept pings before all the nested pings have been used up. But you could even use someone else's desktop in the building if you want to.

Next, you need to write a batch script on one of the systems that is on the UPS. Have it ping the desktop every so often (e.g. once a minute). If there's an error, drop into a nested if and ping it again just to make sure. Once enough ping errors have built up (say, three), and you're pretty sure the building power is down, launch the automatic shutdown procedure from the batch script.

I hope this helps!

Ken
Click to expand...
 
Ken, yes the message boards don't seem to work when the servers are shut off. Even if they did work it doesn't appear that anyone from S@H reads them anymore or very, very rarely. Maybe they read the BOINC SETI forums.

Maybe you could try contacting Dave Anderson or one of the others directly.
S@H Personel
 
Good idea, thanks! 🙂 Mail sent to Court Cannick, Systems Administrator. :thumbsup:
 
The message-boards relies on the database, so someone can't just start spamming the boards but have to log-in, and is therefore off when the database is down. This is the same for both projects.

Both Matt Lebofsky and Rom Walton checks and posts to the BOINC-boards ocassionally, but the best is to use one of the BOINC mailing-lists or emailing directly.


While there has been server-upgrades for BOINC on Sundays, it's not certain anyone had the opportunity this weekend to go to the lab and start the computers for some hours.

Besides, a couple days outage isn't really a problem, in "classic" just use public and/or personal queues.
As for BOINC, since it's multi-project an outage or not enough wu in one project just means you'll get more work from the other projects you've attached to. You can get work from Predictor@home, Einstein@home and CPDN, and if already signed-up you can also get work from LHC.
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top