| View previous topic :: View next topic |
| Author |
Message |
mohrorless
Mail Order Goat Bride
Joined: 09 Oct 2006
Posts: 11206
Location: NYC
|
 Posted: Fri May 25, 2007 4:50 pm Post subject: SETI is down again!!!! |
 |
|
According to the server status page on the SETI website, most everything is disabled. Does anyone have any idea why?
_________________
Fetch me the Holy Hand Grenade!
Keeper of the Unending keg of PGGBs
Taunter in Training
Campaign Manager for Sir Shrubbery
Plus
 |
|
| Back to top |
|
 |
belenus
Prince
Joined: 08 May 2007
Posts: 1244
Location: Pale blue dot (Aviles, Asturies, Spain, Europe, Eurasia, North of Africa, South of Arctic....)
|
| Posted: Fri May 25, 2007 5:36 pm Post subject: |
|
|
Yeah, they say they're having trouble with the database, and ask to try later.
I'm new to this, so I don't know, but, has SETI always been so chaotic?
_________________
 |
|
| Back to top |
|
|
Elwood
Knight
Joined: 21 Feb 2006
Posts: 84
Location: West Virginia, USA
|
| Posted: Fri May 25, 2007 9:12 pm Post subject: |
|
|
I don't know, it seems to me that a three or four day cache was usually sufficient to get through any outage until about a year ago, but it seems that lately the failures are more catestrophic. It could be popularity or servers and parts aging at about the same rate or any number of things, but I just took a machine that runs SETI more than 90% of the time over to CPDN for the summer. It will take over 3200 hours to run two models and I'm going to keep an eye on the SETI technical news and message boards for a few months and hopefully re-join the efforts with a couple of new-ish machines then.
_________________
 |
|
| Back to top |
|
|
mohrorless
Mail Order Goat Bride
Joined: 09 Oct 2006
Posts: 11206
Location: NYC
|
| Posted: Fri May 25, 2007 9:48 pm Post subject: |
|
|
SETI is usually pretty stable. The last few weeks was from a pretty severe server crash.
They are back up again.
_________________
Fetch me the Holy Hand Grenade!
Keeper of the Unending keg of PGGBs
Taunter in Training
Campaign Manager for Sir Shrubbery
Plus
|
|
| Back to top |
|
|
Bellator
Prince
Joined: 12 Mar 2006
Posts: 741
Location: Provence
|
| Posted: Sat May 26, 2007 12:37 am Post subject: |
|
|
| Elwood wrote: | | I don't know, it seems to me that a three or four day cache was usually sufficient to get through any outage until about a year ago, but it seems that lately the failures are more catestrophic. It could be popularity or servers and parts aging at about the same rate or any number of things, but I just took a machine that runs SETI more than 90% of the time over to CPDN for the summer. It will take over 3200 hours to run two models and I'm going to keep an eye on the SETI technical news and message boards for a few months and hopefully re-join the efforts with a couple of new-ish machines then. |
I hope you have more luck with CPDN than I: for the last five days I get "HTTP internal server error". So there is no contact with the server at all and this problem has happened to me a number of times. It is no fun to detach after having crunched 50 hours or so. SETI has indeed been acting up but after a few days it does come through again. Perhaps they are victims of their own success?!
_________________
Nothing is more futile than the memory of a temptation resisted
ubi bene ibi patria |
|
| Back to top |
|
|
Jimmy P
Knight
Joined: 01 Dec 2005
Posts: 30
|
| Posted: Sat May 26, 2007 1:49 pm Post subject: |
|
|
At last all six of my SETI shrubbers have full caches !!
I guess its now time for ramming speed Mr Cristian.
 |
|
| Back to top |
|
|
Elwood
Knight
Joined: 21 Feb 2006
Posts: 84
Location: West Virginia, USA
|
| Posted: Sat May 26, 2007 1:51 pm Post subject: |
|
|
Yeah, but CPDN models take close to 3000 hours to run. They don't have to contact the server every day, every week or even every month. Yes, it sends trickle-up messages and gives intermittent credit, but so far as I can tell one could run the entire model and upload it at the end to receive full credit which is what, 50000 cobblestones or so?
Now once they did have to send out a universal abort code because there was a problem with the model itself, but mostly it seems to give an absolutely guarenteed shrubbery to work on. Even if every other project went down, odds are that you won't run through 2800-3200 hours of work waiting.
_________________
|
|
| Back to top |
|
|
JerWA
Prince
Joined: 01 Jan 2007
Posts: 1497
Location: WA, USA
|
| Posted: Tue May 29, 2007 1:55 am Post subject: |
|
|
And once again SETI is down. Lots of people talking about being unable to download work, myself included. I have 3 work units showing as pending download, but I've been getting nothing but errors for 8 hours while trying to actually GET those work units. <sigh>
_________________
 
Stats: [BOINC Synergy] - [Free-DC] - [MundayWeb] - [Netsoft] - [All Project Stats] |
|
| Back to top |
|
|
Eaving
Prince
Joined: 23 Feb 2003
Posts: 694
Location: Portland, Or.
|
| Posted: Tue May 29, 2007 3:54 am Post subject: |
|
|
*grumbles under his breath* Getting very frustrated by this. I mean I should probably be happy as Rosetta is getting some good numbers from me and its a more obviously useful project but Ive got a soft spot for Seti and my speediest machine hasnt been able to get any shrubs in weeks now.
_________________
  |
|
| Back to top |
|
|
Lloyd M.
Prince
Joined: 02 Mar 2007
Posts: 521
|
| Posted: Thu May 31, 2007 10:20 pm Post subject: |
|
|
| Eaving wrote: | | *grumbles under his breath* Getting very frustrated by this. I mean I should probably be happy as Rosetta is getting some good numbers from me and its a more obviously useful project but Ive got a soft spot for Seti and my speediest machine hasnt been able to get any shrubs in weeks now. |
SETI is usually pretty good. They had a database server crash, got a replacement either free or very cheap from Sun, but it was unsupported. So Sun donated a brand new one.
My problem is some major phantom WU's (SETI thinks they have sent them to me, but I don't have them). This resulted in my fastest machine not getting any work for quite a while, because SETI thought I already had plenty.
I've heard the way to take care of that is to go to no new tasks, clear out your cache, then detach and reattach from the project. I don't feel like fooling with that, so I'm just going to let the phantom WU's time out. I did finally get a few on my fastest machine.
I've been concentrating on Rosetta, and Einstein has dumped a nearly impossible load of WU's on me as well.
Once I get my cache cleared out, I'm going to try a much shorter network connect interval. As SETI started crashing, I had turned it up in a vain attempt to try to get enough work from them to carry over through their outages.
_________________
 |
|
| Back to top |
|
|
Eaving
Prince
Joined: 23 Feb 2003
Posts: 694
Location: Portland, Or.
|
| Posted: Fri Jun 01, 2007 5:19 am Post subject: |
|
|
I did the detatch/reattach thing and eventually started getting work units again, but it was more than slightly frustrating the amount of time it took between when they got back up and when I started receiving shrubs. I've no problem during their down time, the server is gone so thats understandable, was the later issues that annoyed. I really don't want to cut down on teh check with network time because I like having a large buffer for when they have minor issues. Had there not been issues wiht sending out new units after they got up and running 2 of my 3 machines would have actually had a tiny bit of work left on hand, as it was I was out of the running for awhile before things reconnected. 2 of my three are recieving shrubs at the moment and the third is down due to me not them for once. (Rearranging the house and awaiting a new wireless card to get net access to that machine again.)
_________________
|
|
| Back to top |
|
|
JerWA
Prince
Joined: 01 Jan 2007
Posts: 1497
Location: WA, USA
|
| Posted: Fri Jun 01, 2007 6:13 am Post subject: |
|
|
It appears to all be working happily again. The 2nd outage wasn't bad, it just came at a bad time when everyone was already bent out of shape about the huge outage from the crash.
_________________
Stats: [BOINC Synergy] - [Free-DC] - [MundayWeb] - [Netsoft] - [All Project Stats] |
|
| Back to top |
|
|
Pooh Bear 27
Prince
Joined: 28 Jan 2005
Posts: 1358
Location: Fond du Lac, WI
|
| Posted: Fri Jun 01, 2007 9:53 pm Post subject: |
|
|
Weekend of June 8-10 will be another outage, because Berkeley itself is having power work done.
** New notice shows this will be in July sometime.
_________________
Watch the movie about me The 4-1-4s: The Original Teenage Hackers (It is only about 12 minutes long.)
My lucky numbers are 121*2^4553899-1 and 3756801695685*2^666669±1
Last edited by Pooh Bear 27 on Sat Jun 02, 2007 4:42 am; edited 1 time in total |
|
| Back to top |
|
|
Lloyd M.
Prince
Joined: 02 Mar 2007
Posts: 521
|
| Posted: Fri Jun 01, 2007 10:48 pm Post subject: |
|
|
| JerWA wrote: | | It appears to all be working happily again. The 2nd outage wasn't bad, it just came at a bad time when everyone was already bent out of shape about the huge outage from the crash. |
I agree. Things were flaky on and off for a while, then they cratered. The thing is, I'm sure I'm not the only one that had a lot of phantom WUs, so it took a while for my best machine to get any more, because SETI thought I already had plenty.
A lot of people seemed to take it personally, particularly those that didn't understand that there are plenty of worthy projects to shrub for.
I think that's a definite advantage to being on a team. I got back with SETI, because I had done the classic version (lo these many years ago), then I added other projects as teammates suggested them for one reason or another (e.g. improving our team standing for that project).
I think I have a good mix of projects now. I was even able to keep my shrubbers busy when both SETI and PrimeGrid were having problems. Recently, when Rosetta couldn't get it together to download WUs to my son's very intermittently connected Sempron box, I just attached Einstein, and got plenty of work to keep it shrubbing for a while. Rosetta eventually sent some WUs, too.
_________________
|
|
| Back to top |
|
|
JerWA
Prince
Joined: 01 Jan 2007
Posts: 1497
Location: WA, USA
|
| Posted: Sat Jun 02, 2007 1:47 am Post subject: |
|
|
Yeah, one good thing that came about because of all of this is the BOINC programming team saw the angst, and did something that's so simple it's brilliant: They've included a project list in the next BOINC client. I.e., you can just click a button and add a project, rather than have to go look them up. BoincView already does this, and it makes it super easy for me to shuffle projects around as needed. Having it built into the default client will make it much easier for people to find additional projects when their main project is down.
_________________
Stats: [BOINC Synergy] - [Free-DC] - [MundayWeb] - [Netsoft] - [All Project Stats] |
|
| Back to top |
|
|
Lloyd M.
Prince
Joined: 02 Mar 2007
Posts: 521
|
| Posted: Sat Jun 02, 2007 11:12 pm Post subject: |
|
|
| JerWA wrote: | | Yeah, one good thing that came about because of all of this is the BOINC programming team saw the angst, and did something that's so simple it's brilliant: They've included a project list in the next BOINC client. I.e., you can just click a button and add a project, rather than have to go look them up. BoincView already does this, and it makes it super easy for me to shuffle projects around as needed. Having it built into the default client will make it much easier for people to find additional projects when their main project is down. |
Cool. Has this been released?
_________________
|
|
| Back to top |
|
|
JerWA
Prince
Joined: 01 Jan 2007
Posts: 1497
Location: WA, USA
|
| Posted: Sat Jun 02, 2007 11:21 pm Post subject: |
|
|
Not yet, I think it will be in the 5.10 client which is due soon (the servers have already started adding support for it). Depends on if they added it to 5.10 or if it will be in the release after that.
_________________
Stats: [BOINC Synergy] - [Free-DC] - [MundayWeb] - [Netsoft] - [All Project Stats] |
|
| Back to top |
|
|
Lloyd M.
Prince
Joined: 02 Mar 2007
Posts: 521
|
| Posted: Sat Jun 02, 2007 11:30 pm Post subject: |
|
|
| JerWA wrote: | | Not yet, I think it will be in the 5.10 client which is due soon (the servers have already started adding support for it). Depends on if they added it to 5.10 or if it will be in the release after that. |
Hmmm... OK.
While I'm thinking about it, didn't I see a message thread hereabout concerning the best way to switch clients? I might just go with whatever is stable now, especially since the Einstein folks say their app works a lot better (read: fewer errored-out WUs) with a newer app, though I forgot the exact version. 5.9?
_________________
|
|
| Back to top |
|
|
JerWA
Prince
Joined: 01 Jan 2007
Posts: 1497
Location: WA, USA
|
| Posted: Sun Jun 03, 2007 12:39 am Post subject: |
|
|
The BOINC version and the app for the project are separate. The newest BOINC stable release is 5.8.16 still. The most recent development version is 5.10.0 so they just moved up out of the 5.9 tree (Maybe we'll get a 5.9 stable, but it looks like they're skipping straight to 5.10).
_________________
Stats: [BOINC Synergy] - [Free-DC] - [MundayWeb] - [Netsoft] - [All Project Stats] |
|
| Back to top |
|
|
Lloyd M.
Prince
Joined: 02 Mar 2007
Posts: 521
|
| Posted: Sun Jun 03, 2007 6:46 pm Post subject: |
|
|
| JerWA wrote: | | The BOINC version and the app for the project are separate. The newest BOINC stable release is 5.8.16 still. The most recent development version is 5.10.0 so they just moved up out of the 5.9 tree (Maybe we'll get a 5.9 stable, but it looks like they're skipping straight to 5.10). |
On Einstein I'm pretty sure that they were recommending that people use 5.9.something client, because their app was happier with it. They've had some real problems with their app (particularly with AMD CPUs), and they claim that their app is better behaved with some particular version of the client.
On a related note, my client cratered completely today. I had had to do a hard reset on Windoze, and all of my tasks were gone when I could get the BOINC client going again (which took two tries).
Figuring that I had nothing to lose; I did the detach/reattach trick with Einstein, and when I merged the old instance of this computer on Einstein with the new one it had just created, the Einstein server immediately re-sent the WU's I had lost when the BOINC client crashed!
This makes me want to try it with SETI, once I get the cache cleared (I had let it download some SETI WU's when I got the client restarted).
Anyway, the good news is I don't think I lost too much processing when the client crashed.
_________________
|
|
| Back to top |
|
|
|
FAQ
Search
Memberlist
Usergroups
Register
Profile
Log in to check your private messages
Log in
