We are recovering from a database crash


log in

Advanced search

Message boards : News : We are recovering from a database crash

Previous · 1 · 2 · 3 · 4 · 5 · Next
Author Message
Harley
Send message
Joined: 24 Aug 02
Posts: 1
Credit: 5,810,594
RAC: 0
Canada
Message 1302093 - Posted: 4 Nov 2012, 14:42:58 UTC

Thanks Ivan, that worked for me. I was able to upload my results but I'll have to wait a bit to see if I'm able to download more work units. I'm completely out on 1 of my machines.
____________

Dave Offutt
Send message
Joined: 14 Apr 00
Posts: 9
Credit: 1,472,014
RAC: 4,856
United States
Message 1302327 - Posted: 4 Nov 2012, 23:47:59 UTC

I'm empty on my machine. Tried the No New Tasks mentioned below. Now it appears to be unable to communicate as thestatus shows Communications deferred
____________

Profile Cyruss322
Volunteer tester
Send message
Joined: 9 Oct 11
Posts: 1
Credit: 274,350
RAC: 1
United States
Message 1302343 - Posted: 5 Nov 2012, 1:12:42 UTC

Just put my new pc together and cant download any work units! I want to tear it up with this 3770K and gtx 660 TI

Profile Mad Fritz
Avatar
Send message
Joined: 20 Jul 01
Posts: 87
Credit: 11,334,904
RAC: 0
Switzerland
Message 1302345 - Posted: 5 Nov 2012, 1:20:54 UTC

A much better chance to get answers and informations you'll have here
http://setiathome.berkeley.edu/forum_forum.php?id=10
____________

Scott Campbell
Send message
Joined: 27 Jun 99
Posts: 4
Credit: 5,693,664
RAC: 4,640
United States
Message 1302475 - Posted: 5 Nov 2012, 13:53:55 UTC

I finally got some tasks but it appears only the GPU is running a task. Is it something I did wrong?
____________

N9JFE David S
Volunteer tester
Avatar
Send message
Joined: 4 Oct 99
Posts: 10764
Credit: 13,496,662
RAC: 14,930
United States
Message 1302483 - Posted: 5 Nov 2012, 14:40:51 UTC - in response to Message 1302475.

I finally got some tasks but it appears only the GPU is running a task. Is it something I did wrong?

For the information of the several of you who have posted in the last few days about not being able to connect to the servers, it's a combination of AP units being sent again and a shorty storm of MB units. The lab's internet connection is overloaded and data is getting lost.

To answer Scott specifically, assuming your preferences are set to allow both CPU and GPU work, the server will try to give work to your best processor first, which is your GPU. Once the GPU's cache fills up with work, it will then give some to your CPU. However, that probably won't happen until the current network congestion clears up. Be patient.

____________
David
Sitting on my butt while others boldly go,
Waiting for a message from a small furry creature from Alpha Centauri.


Dave
Send message
Joined: 6 Oct 07
Posts: 1
Credit: 15,852,744
RAC: 37,760
United States
Message 1302501 - Posted: 5 Nov 2012, 16:51:40 UTC

David, the best answer to what is going I have read. Thank you, another David

92surf
Avatar
Send message
Joined: 25 Oct 12
Posts: 114
Credit: 62,263
RAC: 0
Australia
Message 1302517 - Posted: 5 Nov 2012, 17:47:54 UTC
Last modified: 5 Nov 2012, 17:49:24 UTC

Approx how long will down time last, will do some running repairs.
____________

rob smith
Volunteer tester
Send message
Joined: 7 Mar 03
Posts: 8149
Credit: 52,867,638
RAC: 76,005
United Kingdom
Message 1302601 - Posted: 5 Nov 2012, 21:05:29 UTC

For the information of the several of you who have posted in the last few days about not being able to connect to the servers, it's a combination of AP units being sent again and a shorty storm of MB units. The lab's internet connection is overloaded and data is getting lost.


Wrong!, nothing to do with the number of AP/shorties, but all to do with some as yet undeclared server problem. The splitters are OFF LINE, and have been for a number of hours, so there are no new work units being generated, or distributed. There is also an abnormally high background amount of server traffic, which has nothing to do with the distribution of WU, as there are NO NEW UNITS of any sort to be distributed.
____________
Bob Smith
Member of Seti PIPPS (Pluto is a Planet Protest Society)
Somewhere in the (un)known Universe?

Michael Deglman
Send message
Joined: 27 Jun 99
Posts: 1
Credit: 2,281,698
RAC: 2,224
United States
Message 1302610 - Posted: 5 Nov 2012, 21:24:31 UTC - in response to Message 1302601.

Thanks for the information. I will just be patient.
____________
Michael Deglman

N9JFE David S
Volunteer tester
Avatar
Send message
Joined: 4 Oct 99
Posts: 10764
Credit: 13,496,662
RAC: 14,930
United States
Message 1302617 - Posted: 5 Nov 2012, 21:40:49 UTC - in response to Message 1302601.

For the information of the several of you who have posted in the last few days about not being able to connect to the servers, it's a combination of AP units being sent again and a shorty storm of MB units. The lab's internet connection is overloaded and data is getting lost.

Wrong!, nothing to do with the number of AP/shorties, but all to do with some as yet undeclared server problem. The splitters are OFF LINE, and have been for a number of hours, so there are no new work units being generated, or distributed. There is also an abnormally high background amount of server traffic, which has nothing to do with the distribution of WU, as there are NO NEW UNITS of any sort to be distributed.

Yes, I wasn't aware of this when I posted earlier. What I said is usually the cause of such problems, but not in this case. (I think it was the cause but it got so bad they shut off the splitters until everything can settle down and they see what went wrong.)

____________
David
Sitting on my butt while others boldly go,
Waiting for a message from a small furry creature from Alpha Centauri.


Tom
Send message
Joined: 12 Aug 11
Posts: 114
Credit: 4,566,097
RAC: 0
United States
Message 1302621 - Posted: 5 Nov 2012, 21:43:16 UTC - in response to Message 1302601.
Last modified: 5 Nov 2012, 21:44:51 UTC

I think before we can reliably poo poo excellent reasons for the problem
we need to have a statistic in the server status that shows the number of ghost
processes that the system thinks are out in the field but in no way are downloaded currently by the clients.

If we have that stat somewhere we need to see it, before we can dismiss
AP's and Shorties as the cause of the recent total outage as far as most
clients are concerned.

PS - I could not get anything to download and my Ghosts were increasing
to several hundred with the basic default 7.0.28 client until I chose to use
HotSpot Shield as a Proxy/concentrator. then I was able to download all my Ghosts in a timely fashion.

edit - Prior to any server changes

Profile joost
Send message
Joined: 15 Jan 09
Posts: 1
Credit: 1,674,064
RAC: 1,255
Netherlands
Message 1303265 - Posted: 7 Nov 2012, 20:23:02 UTC

Well, i am glad this forum/topic reveals at least sómething of what is happening, tho the details are unclear 2 me.
As i type this, at this very moment, i see new work units from SETI coming in, while i was waiting for days for new downloads.
The suggestion to set the project to -no new tasks- , up date, then to - new tasks enabled- and update again apparently worked.
I do hope the situation returns to normal soon again,
greetz from holland
joost
____________

Profile S@NL Blue Angel
Avatar
Send message
Joined: 11 May 03
Posts: 224
Credit: 4,544,373
RAC: 2
Netherlands
Message 1304357 - Posted: 10 Nov 2012, 9:05:29 UTC - in response to Message 1298434.

Talking about data base, I have this on my account
es (1895) · In uitvoering (1612) · Validation pending (215)
point is, I have less then 200 works to chrunch and the figures
shows I should have 1612? I wish I had them, but anyway also
215 works waiting for validition witch I din't get yet.
I hope my work don't get lost?

Greetings Blue Angel
____________

Profile Dannis
Send message
Joined: 29 Jan 06
Posts: 24
Credit: 5,248,145
RAC: 3,090
United States
Message 1305605 - Posted: 12 Nov 2012, 23:04:57 UTC

This issue, need I say so has been a server issue for months. It gets a band aid every now and then. But it continues to occur in one form or another. Either no one knows how to fix it, or they just don't have the time. Is is now costing us members. I know that other projects do not have constant problems keeping up and running. What will Seti do when all the members are gone and they have to pay for Computer time to analyze results. IF they don't put some emphasis on getting this under control, that's exactly where we are heading. My own team has seen a decline in participation over the last 8 months. When I ask why, these issues are given as the reason why members move on to more stable projects.

Profile KWSN THE Holy Hand Grenade!
Volunteer tester
Avatar
Send message
Joined: 20 Dec 05
Posts: 1902
Credit: 9,223,054
RAC: 13,685
United States
Message 1306410 - Posted: 15 Nov 2012, 11:47:49 UTC
Last modified: 15 Nov 2012, 12:01:38 UTC

Every time your computer comes in with a "requesting tasks" scheduler request a certain number are assigned - but if your request times out, then you never download them. (your computer never finds out about them…) They'll be downloaded as "resent lost tasks" (…usually 20 at a time…) when a "requesting tasks" scheduler request eventually makes it through (…if one ever does…) – or when they run up against the task's deadline (I've got a huge number that will expire on 11/17/12, for example) they will get assigned to another computer. The way to stop this nonsense is for the SETI crew to determine what is wrong with scheduler requests - I haven't gotten a good reply to a "reporting xx tasks, requesting tasks" request for almost a week. (The few times I've not gotten a failure are when I just report tasks or request them.)

If the SETI crew would shorten the 5 minute delay after a good Scheduler request, we (the SETI community) could at least do the NNT, report, Request new tasks routine in a more timely fashion... (and some of us have to do this on multiple computers!) (could someone explain why the delay is 5 minutes? I know that a delay is now required - but what's the magic of 5 minutes?)
____________
.

Harry62
Send message
Joined: 14 May 09
Posts: 6
Credit: 964,192
RAC: 0
Message 1306435 - Posted: 15 Nov 2012, 14:06:29 UTC - in response to Message 1305605.

Why I’m leaving Seti@Home

I first joined Seti@Home in 1999. I believed in the project, and was glad to let Seti@Home use my computer. All was good in the world. I was using the OLD (original ) Seti interface, and I downloaded several User Certificates, the Last being for 750 work Units. THEN . . . . the did away with the old interface. They said it wasn’t providing them with the data they needed, and they introduced a NEW SETI@ HOME INTERFACE, the BOINC. I hated it, it never ( at least for me ) worked correctly, would constantly hang, not upload completed tasks, not download any new tasks. ( Sound familiar anyone?)

I and a whole lot of other original users complained, and we were told sorry but this gives us mush more useful data and is here to stay. So I left. No more SET@HOME for me, in fact NO more any program that used BOINC.

In 2009 and friend told me he had discovered a new site called SETi@Home . . . and when I queried him about it he told me that the interface worked fine. So I again joined, tried it for a while and discovered it was the same intermittent sometimes it works, sometimes it doesn’t, interface. So then I left again.
2012 and again a friend told me that most of the bugs were gone and the interface was reliable. So for the LAST time, I dusted off my user account, and started processing Seti data again. At first I will admit I was pleasantly surprised. Yea it would hang or crash but not nearly as often. But then it started all over again, in the past few weeks, I have seen the old problems return. First I went 7 days with only getting GPU data to process, then now for the past three days, I’ve been trying to upload 90 Plus completed work files ( with NO LUCK, nor have I been able to down load an new work files.)

OK, guys, you got me . . . you actually made me comeback to the same problems with this interface 3 times. SHAME ON ME ! ! there will be NO forth time ! you can take my 873823.19 work units and my 15,491 average work units, and do with them what you will. ( yea I know, I’m NOT a power user, and those numbers aren’t the greatest, but I was letting you use my computer, when I could )

“ET . . . could you please hold the line, our data base crashed”

PS: I am a US Citizen, the senior Network administrator for a medium size company here in Germany, and I’ve had enough of SETI and BOINC !

Profile CLYDE
Volunteer tester
Avatar
Send message
Joined: 9 Aug 99
Posts: 1503
Credit: 21,134,779
RAC: 35,280
United States
Message 1306444 - Posted: 15 Nov 2012, 15:20:19 UTC - in response to Message 1306435.

Why I’m leaving Seti@Home

I first joined Seti@Home in 1999. I believed in the project, and was glad to let Seti@Home use my computer. All was good in the world. I was using the OLD (original ) Seti interface, and I downloaded several User Certificates, the Last being for 750 work Units. THEN . . . . the did away with the old interface. They said it wasn’t providing them with the data they needed, and they introduced a NEW SETI@ HOME INTERFACE, the BOINC. I hated it, it never ( at least for me ) worked correctly, would constantly hang, not upload completed tasks, not download any new tasks. ( Sound familiar anyone?)

I and a whole lot of other original users complained, and we were told sorry but this gives us mush more useful data and is here to stay. So I left. No more SET@HOME for me, in fact NO more any program that used BOINC.

In 2009 and friend told me he had discovered a new site called SETi@Home . . . and when I queried him about it he told me that the interface worked fine. So I again joined, tried it for a while and discovered it was the same intermittent sometimes it works, sometimes it doesn’t, interface. So then I left again.
2012 and again a friend told me that most of the bugs were gone and the interface was reliable. So for the LAST time, I dusted off my user account, and started processing Seti data again. At first I will admit I was pleasantly surprised. Yea it would hang or crash but not nearly as often. But then it started all over again, in the past few weeks, I have seen the old problems return. First I went 7 days with only getting GPU data to process, then now for the past three days, I’ve been trying to upload 90 Plus completed work files ( with NO LUCK, nor have I been able to down load an new work files.)

OK, guys, you got me . . . you actually made me comeback to the same problems with this interface 3 times. SHAME ON ME ! ! there will be NO forth time ! you can take my 873823.19 work units and my 15,491 average work units, and do with them what you will. ( yea I know, I’m NOT a power user, and those numbers aren’t the greatest, but I was letting you use my computer, when I could )

“ET . . . could you please hold the line, our data base crashed”

PS: I am a US Citizen, the senior Network administrator for a medium size company here in Germany, and I’ve had enough of SETI and BOINC !


I also started in the pioneer year of 1999, and I am starting to agree.

The problem, as I see it, is not the problem(s), but LACK of communication from the people running this. THERE IS NONE!!!

I hear "Matt is on vacation" and we have to await his return to fix whatever the problem is! This project is over 13 years old, employing thousands of volunteers, utilizing many thousands of computers, large electrical usage, accepting our donations, and it depends on one person? Berkeley should be ashamed.

I am beginning to smell a scam. If it is not, perhaps we should think of this effort as nothing more than a glorified High School Project.

I hope my many years of effort, not to mention money, have not gone to waste.
____________

Profile Bernie Vine
Volunteer moderator
Volunteer tester
Avatar
Send message
Joined: 26 May 99
Posts: 6824
Credit: 24,650,627
RAC: 26,817
United Kingdom
Message 1306456 - Posted: 15 Nov 2012, 16:44:21 UTC - in response to Message 1306435.
Last modified: 15 Nov 2012, 16:45:54 UTC

Why I’m leaving Seti@Home

I first joined Seti@Home in 1999. I believed in the project, and was glad to let Seti@Home use my computer. All was good in the world. I was using the OLD (original ) Seti interface, and I downloaded several User Certificates, the Last being for 750 work Units. THEN . . . . the did away with the old interface. They said it wasn’t providing them with the data they needed, and they introduced a NEW SETI@ HOME INTERFACE, the BOINC. I hated it, it never ( at least for me ) worked correctly, would constantly hang, not upload completed tasks, not download any new tasks. ( Sound familiar anyone?)

I and a whole lot of other original users complained, and we were told sorry but this gives us mush more useful data and is here to stay. So I left. No more SET@HOME for me, in fact NO more any program that used BOINC.

In 2009 and friend told me he had discovered a new site called SETi@Home . . . and when I queried him about it he told me that the interface worked fine. So I again joined, tried it for a while and discovered it was the same intermittent sometimes it works, sometimes it doesn’t, interface. So then I left again.
2012 and again a friend told me that most of the bugs were gone and the interface was reliable. So for the LAST time, I dusted off my user account, and started processing Seti data again. At first I will admit I was pleasantly surprised. Yea it would hang or crash but not nearly as often. But then it started all over again, in the past few weeks, I have seen the old problems return. First I went 7 days with only getting GPU data to process, then now for the past three days, I’ve been trying to upload 90 Plus completed work files ( with NO LUCK, nor have I been able to down load an new work files.)

OK, guys, you got me . . . you actually made me comeback to the same problems with this interface 3 times. SHAME ON ME ! ! there will be NO forth time ! you can take my 873823.19 work units and my 15,491 average work units, and do with them what you will. ( yea I know, I’m NOT a power user, and those numbers aren’t the greatest, but I was letting you use my computer, when I could )

“ET . . . could you please hold the line, our data base crashed”

PS: I am a US Citizen, the senior Network administrator for a medium size company here in Germany, and I’ve had enough of SETI and BOINC !


Actually it is nothing to do with "the interface" as you call it BOINC works fine, does for me. It is actually SETI@Home that is having problems currently.

I have never had BOINC "hang or crash" on any of my 6 computers.

Other projects using BOINC are not having these problems.
____________


Today is life, the only life we're sure of. Make the most of today.

Alec
Send message
Joined: 3 Jun 06
Posts: 10
Credit: 816,773
RAC: 0
United States
Message 1306481 - Posted: 15 Nov 2012, 17:50:23 UTC - in response to Message 1306435.

SETI may have the biggest "cool factor" for most of us, but There are many other BOINC projects that are worthy of your computing time; rather than throwing in the towel completely, why not attach to another project or two and crunch those during the times that SETI is unavailable?

For example, Einstein@Home uses the CPU to (try to) detect gravitational waves, and GPU to look for previously undiscovered binary pulsars, with nearly 50 new pulsars discovered so far. That's pretty neat stuff. There are two projects modeling protein chains to advance our ability to cure diseases. There's one doing climate modeling, and a few doing mathematics testing. I'm sure you can find something that meets your particular interests while you wait for things to clear at SETI.


____________

Previous · 1 · 2 · 3 · 4 · 5 · Next

Message boards : News : We are recovering from a database crash

Copyright © 2014 University of California