Is SETI server down?

[Alex99]

I got intot the office this morning and my seti spy is yellow. When I do a manual transmit I get this. This is the 3rd time it happens, I losing some WUs because of this since I'm not here to baby sit the system on weekends. This sucks.

[Yodums]

It was down for me last night. Before you leave work Friday put the cache to whatever that fills 3 days.

[Alex99]

Originally posted by Yodums
It was down for me last night. Before you leave work Friday put the cache to whatever that fills 3 days.

Oh yeah, great idea. I'll set it to 5. Thanks Yodums.

[killem1x1]

I'm having the same problem :( I've got 5 mins before I go out of town, and I had another machine that would have been crunching all week. Man, this is getting so old:mad:

[Yodums]

I used this about a year ago I never had this problem :rolleyes:

[Morpheus]

Saw this for the first time myself... a couple days ago... when adjusting cache size of a previously working SETIspy & Driver setup...

New problem, to me... anybody else?

[rogerdugans]

I can download (SetiQ rocks!) but I have 34 wu ready for upload!

Hopefully they'll go in tonight.........

[hallen]

I think is a complete shut down now. I been trying hard to get in and get nothing.

You would think they would do a little PR work and explain. It's really wasting a lot of peoples time.

[Yodums]

Originally posted by hallen
I think is a complete shut down now. I been trying hard to get in and get nothing.

You would think they would do a little PR work and explain. It's really wasting a lot of peoples time.

Yea, people get impatient and just shut off Seti. Its getting all of frustrated!

[TC]

Originally posted by hallen
I think is a complete shut down now. I been trying hard to get in and get nothing.

You would think they would do a little PR work and explain. It's really wasting a lot of peoples time. This is not the first time they've had problems like this. There was a considerable problem for a long time whereby completed work units were not being credited. They hardly acknowledged it for over a month if I remember correctly. These problems arise from time to time, but the overall reliability of the project is far better than folding imo.

[hallen]

Got one wu off and new one coming in.

[hallen]

This is the official word-

The goal is for SETI@home users to get as good a data rate as possible while at the same time not adversely affecting the rest of campus. The current (new) policy is for SETI@home to get all excess capacity in outbound bandwidth (inbound is not an issue). This is being done by placing our traffic at a lower priority than that of other users, but applying no rate limit other than the campus wide cap. Placing us at a lower priority is necessary so that interactive Internet users do not experiencYe painfully high latency.This policy translates into enormous capacity at (local) night, but a tight squeeze during peak hours.

But they don't tell you when this enormous capacity begins.

edit- I got a total of one in one out. No enormous capacity as of 7:53 PST.

[Yodums]

Its 10:55est here... At about sometime in the morning it should be sending its WU's out I guess.

PS : Check PM

[Alex99]

Originally posted by hallen
This is the official word-

The goal is for SETI@home users to get as good a data rate as possible while at the same time not adversely affecting the rest of campus. The current (new) policy is for SETI@home to get all excess capacity in outbound bandwidth (inbound is not an issue). This is being done by placing our traffic at a lower priority than that of other users, but applying no rate limit other than the campus wide cap. Placing us at a lower priority is necessary so that interactive Internet users do not experiencYe painfully high latency.This policy translates into enormous capacity at (local) night, but a tight squeeze during peak hours.

But they don't tell you when this enormous capacity begins.

edit- I got a total of one in one out. No enormous capacity as of 7:53 PST.

Thanks for the info. That explains it. Guess it's time to fatten up the SetiSpy cache.

[hallen]

The way things are going it might be best to fatten up enough to last until spring break. I wonder what the future of America is doing hogging all that bandwidth. Science? or is some Judge about to shut down KaZaA?

Good luck

Keep crunching. Air if you have to.

Harvey

[hallen]

Originally posted by TC
This is not the first time they've had problems like this. There was a considerable problem for a long time whereby completed work units were not being credited. They hardly acknowledged it for over a month if I remember correctly. These problems arise from time to time, but the overall reliability of the project is far better than folding imo.

TC- I was thinking that this was such a great measure of systems performance I should be paying them. Now I've changed my mind and think they should be paying us. But that will pass and things will get rolling again I'm sure.

Harvey

[rogerdugans]

In the last 6 hours I have completed 4 and uploaded one. Not completely down but VERY hard to connect.

[Wookie8662]

I just raised the cache on my 3 PC's from 4 days to 7 days.

You never know if they will shutdown to do repairs/maintenance.
Better to be safe, then not have any WU to crunch on.

PS. Its 10:10PM here in the same state as SETI, and can't connect right now.

[JAWS]

Around 12:30, I sent two and downloaded two.

After that, there was no hope! :(

[Basher]

I also bumped my cache up to 1 week :) I have 4 to send right now and I only do 6 a day with a constant internet connection :(

[dru]

Oddest part seems to be that I am seeing the direct opposite of what they are advising (and the same as many of you are reporting.) I can download new WUs with little problem, but can't seem to get any uploads. Since they say the limits are on the outbound traffic from their side, this doesn't seem quite right.

I've got 13 results qued to be sent in, and none of them seem to want to leave:(

I've set my SetiQue Server to not connect till later in the night/morning, so maybe that'll help. I've also increased the time between downloads of incoming WUs, to try to open some bandwidth for others. I'm also dialing up the level of both the minimum and maximum level of qued WUs to maintain here, which is going to put more load on their connect short term, but I'd hate to see my machines putting out all this heat and having nothing to crunch!

Hope they get this together soon... it's a great project, and I just joined up with you folks, and got people to pass:D

[Wookie8662]

Originally posted by dru

Hope they get this together soon... it's a great project, and I just joined up with you folks, and got people to pass:D

Hey welcome to the Team!

[Basher]

Originally posted by dru


Hope they get this together soon... it's a great project, and I just joined up with you folks, and got people to pass:D


Welcome to the Team!!!!!!!!

[LandShark]

as of now, near 3am EST, just sent 3 out, but there's still 'bout 30+ waiting to upload tho...........:eh?:

also, up my machines' cache from 4days to 7 days now!! just in case!!

[hallen]

Bandwidth flying this AM. 10:44est. Looks like the Stamies are late risers.
Off to load up.:cool:

[Yodums]

Yea I noticed it was back to normal this morning too. Let's just see how long it'll last ;)

Yodums

[JAWS]

Originally posted by hallen
Bandwidth flying this AM. 10:44est. Looks like the Stamies are late risers.
Off to load up.:cool:

I second that! Working great! :)

[e_storm]

I was having probs both uploading finished WUs and downloading new ones. I tried dumping my cache of 6 last night a few times and got nothing. I bumped the cache up to 10 so it would be crunching while the server probs were sorted out, but couldn't even download new ones! I hope my rig isn't idle when I get home :mad:

[rogerdugans]

I haven't been able to upload a wu since yesterday- about 5 total I think!

I have upped my cache some more, since I have no trouble at all getting units, I just can't return them!

Two days worth ready to go.........

[bobt17]

i also uped my cache just in case but for now it seems u can transmit/download work units REALLY late at night or REALLY early in the morning

[rogerdugans]

I am set up for automatically connecting, so it tries 24 hours a day! It was working great until the server problems started, but now I can download and not upload!!!!!

I'm up to 57 wu now, still receiving wus and not sending:(

[Morpheus]

I am uploading & downloading just fine... the download is VERY, VERY, slow though...

Are you, perhaps, "timing out" because the Berkeley server is so slow?

[e_storm]

Ok, I'm ok now...sending and receiving.

[rogerdugans]

Just started connecting for uploads finally- but STILL bad news: SetiQ shows 17 units uploaded, but SetiTeam only shows 11 more units credited!

Hopefully its just a little glitch and the missing 6 will show up, but.......:(

I think I'll try just SetiDriver on the Win clients and use SetiQ for linux!

[LandShark]

it seems like in the morning (EST) is the best time. i have 'bout 40WUs from yesterday that couldn't upload nor download any new one. this morning when i woke up, the cache was all full again! and so does early afternoon today. no problem. guess the worst time is after 6-7pm EST since west coast it's bout 3-4pm which most students is done for school and start surfing/fragging.....

[JigPu]

Well, I just uploaded and downloaded some more WUs. It didn't seem to take any longer than normal for me...

The joys of only having to upload only 4 WUs at a time :D
JigPu

[rogerdugans]

Definitely is something with SetiQ.......I have up- and downloaded two wus with SetiDriver and still not another one with SetiQ;

I'm holding on to it in case I get stuck for wus, but I don't need the cache on linux enough to lose a third of my units!

I wish the guy who wrote it had a link for bug reporting.

[eobard]

My K6-2 just uploaded 5 wu's and downloaded 5 more. It took me no extra time at all. Hopefully things are getting better.

[RED Hot Machine]

The server is still slow for me, not been too good the last few days what ever time it is :(

I've got enough cached, would be nice to upload though.

[Arkamedies]

Being new to using the SETI command client, is the message below normal when you can't connect? And if so is there any way to have it check more than once an hour to transmit?

I am completing WU's in about 3.47 hours a WU, so for a day I guess I should set my cache to about 7 or 8, right?

[rogerdugans]

I've got my caches set to a week of wus.

I know of no way to change clc's 1 hour time out- maybe a hidden command-line switch. You can set it to manual and toggle back and forth though!

[vodka314]

Just tried to send in a wu, but its not workin, its still only about 10:30 berkley time, so... yeah.. they should get more bandwidth.

[vodka314]

Just tried again, got the wu off, along with one that just finished, but I cant get any, weird...

[hallen]

From status page. I found 4am est to be when things were clearing up

January 29, 2002

The campus wide outbound bandwidth cap is set at 70Mb/s. We are getting only around 10Mb/s of this during the day but are getting all we need at night. The best time for connecting varies but is roughly for the 2 hours on either side of 1:00 PST, or 9:00 UT.

http://setiathome.ssl.berkeley.edu/tech_news.html



I also picked up this tidbit. When chrunch comes to shove they just add more processing to the client to keep our hungry PC's from grabbing more wu's. Prepare for client upgrade!!

Another important resource to the project is our pipe to the Internet. UC Berkeley has been quite generous in allowing us to use the campus ISP, given that SETI@home accounts for 30% of all UC Berkeley Internet traffic! This is an expensive resource ($18,000/month). It has gotten to the point where campus has had to limit our bandwidth usage. We are limited to 30Mbits/s outbound (most of our traffic is outbound). This in turn is resulting in connections getting dropped and generally sluggish server response. We need to back off our bandwidth consumption.

Fortunately, we can do this not by an artificial slowdown but by gaining more science. So, yes, you will see the client slow down, but the server response should pick up. More importantly, you will be doing more science.

http://setiathome.ssl.berkeley.edu/version303.html

[hallen]

Server has been fast all day today (Saturday) trying to load enough wu's to last till mid week then struggle through to next weekend.

[vodka314]

Agreed, the server is working great for me last night (friday) and today. We will see how next week goes.... hopefully all the berkley students got their porn fix for a while last week...

[Basher]

Originally posted by vodka314
Agreed, the server is working great for me last night (friday) and today. We will see how next week goes.... hopefully all the berkley students got their porn fix for a while last week...
I would still would cache a weeks worth, I bet it will be the same this week as last but you never know.

[vodka314]

Oh, Im still caching lots, I got 30 cached. It cant hurt to have that many cached.... 10 mb on a 40 gig hdd isnt really going to matter.