Panic Mode On (109) Server Problems?

Message boards : Number crunching : Panic Mode On (109) Server Problems?
Message board moderation

To post messages, you must log in.

Previous · 1 . . . 10 · 11 · 12 · 13 · 14 · 15 · 16 . . . 36 · Next

AuthorMessage
Profile Wiggo
Avatar

Send message
Joined: 24 Jan 00
Posts: 26943
Credit: 261,360,520
RAC: 489
Australia
Message 1909493 - Posted: 29 Dec 2017, 21:14:25 UTC - in response to Message 1909489.  

You are trying to go to Mike's old address. Try http://mikesworld.eu/download.html instead.

Alternate universe it would seem.

Then somebody needs to update the links in the Optimised Applications and Other Binaries - NEW: required for v8: Windows Installer v0.44

You must of missed this thread while playing with your junk. ;-)

Cheers.
ID: 1909493 · Report as offensive
Profile Bernie Vine
Volunteer moderator
Volunteer tester
Avatar

Send message
Joined: 26 May 99
Posts: 9949
Credit: 103,452,613
RAC: 328
United Kingdom
Message 1909495 - Posted: 29 Dec 2017, 21:26:43 UTC
Last modified: 29 Dec 2017, 21:32:02 UTC

I have copied the post from Mikes thread into the "Optimised Applications and Other Binaries - NEW: required for v8: Windows Installer v0.44" thread

So hopefully people will see it now.

PS Anyone could have done that.
ID: 1909495 · Report as offensive
Profile arkayn
Volunteer tester
Avatar

Send message
Joined: 14 May 99
Posts: 4438
Credit: 55,006,323
RAC: 0
United States
Message 1909511 - Posted: 29 Dec 2017, 23:34:03 UTC - in response to Message 1909476.  

Mikeworld is for sale, I downloaded the new 0.45b6 Lunatics file from arkayn's site instead.

Never paid anything there. Anyway DL from Arkayn site is fine too.

I went there, the domain is for sale at Mikeworld, there are no files there, unless the link is bad and I wound up in an alternet universe, so I dled from Arkayn after learning that.

I have not seen the "you don't have a valid version of V8" show since 10:34am, it's now 10:57am, so I think the problem is resolved, I deleted the old Lunatics 0.45b6 zip file from 2016 too.


All three sites have the newest installer now, Mikes, Lunatics and my site.

ID: 1909511 · Report as offensive
Grant (SSSF)
Volunteer tester

Send message
Joined: 19 Aug 99
Posts: 13351
Credit: 208,696,464
RAC: 304
Australia
Message 1909514 - Posted: 29 Dec 2017, 23:46:51 UTC - in response to Message 1909511.  

All three sites have the newest installer now, Mikes, Lunatics and my site.

Looks like you'll need to do a couple of Seti WUs, or you won't be able to start the next panic mode thread.

RAC: 2
Grant
Darwin NT
ID: 1909514 · Report as offensive
Grant (SSSF)
Volunteer tester

Send message
Joined: 19 Aug 99
Posts: 13351
Credit: 208,696,464
RAC: 304
Australia
Message 1909515 - Posted: 29 Dec 2017, 23:48:13 UTC

The Scheduler is back to randomly allocating work, although my cache isn't running down too much before it decides to give out some work at the moment.
Grant
Darwin NT
ID: 1909515 · Report as offensive
Iona
Avatar

Send message
Joined: 12 Jul 07
Posts: 790
Credit: 22,438,118
RAC: 0
United Kingdom
Message 1909547 - Posted: 30 Dec 2017, 6:33:41 UTC

Wanted: Hammer No. 1 Mk 8, qty 1, for application to servers. Alternatively, Boot, Working, Size 9, qty 1, for the same application.
Don't take life too seriously, as you'll never come out of it alive!
ID: 1909547 · Report as offensive
TBar
Volunteer tester

Send message
Joined: 22 May 99
Posts: 5204
Credit: 840,779,836
RAC: 2,768
United States
Message 1909611 - Posted: 30 Dec 2017, 17:11:53 UTC

Seems to be particularly Bad at the moment. All three Linux machines are being refused work, something about 'Project has no tasks available' even though RTS is hovering around 600k. I suspect that is a Red Herring and only appears due to asking for AP work. If you ask for just MB work, as with my Mac, the Server can't come up with a reason for Not sending work and merely says, No Tasks Sent. Fortunately, the Mac is still being sent work suggesting prejudice, but alas, even Windows machines are denied work on occasion. Strange how it always seems to start with the Linux machines though. One machine is right at 100 tasks in the red, will it recover, or will the Server run it out of work?
ID: 1909611 · Report as offensive
Profile zoom3+1=4
Volunteer tester
Avatar

Send message
Joined: 30 Nov 03
Posts: 64912
Credit: 55,293,173
RAC: 49
United States
Message 1909671 - Posted: 30 Dec 2017, 21:58:06 UTC - in response to Message 1909547.  

Wanted: Hammer No. 1 Mk 8, qty 1, for application to servers. Alternatively, Boot, Working, Size 9, qty 1, for the same application.

How's this Iona? Oh and a belated Merry Xmas and Happy New Year to you, or as happy as circumstances permit at least.
ID: 1909671 · Report as offensive
TBar
Volunteer tester

Send message
Joined: 22 May 99
Posts: 5204
Credit: 840,779,836
RAC: 2,768
United States
Message 1909702 - Posted: 30 Dec 2017, 23:46:29 UTC

The Server has decided to continue to Strangle 2 of my Linux machines. The third machine recovered to a full cache, for a while, and then continued to be sent less tasks than was being reported. The other two are getting very low on tasks and it appears the server would be content to run them out of GPU work. Both machines have 3 GPUs and should have around 330 tasks in the 1 day cache;
State: All (2474) · In progress (97) · Validation pending (1092)
State: All (1898) · In progress (98) · Validation pending (785)
Strange how the Server sends them about the same number of tasks instead of the number the Host is Requesting. What are the chances that after hours one is at 97 and the other 98?
I suppose I'm going to have to intervene or else watch the Server Starve my machines to death. Now down to 94 & 95.
ID: 1909702 · Report as offensive
Grant (SSSF)
Volunteer tester

Send message
Joined: 19 Aug 99
Posts: 13351
Credit: 208,696,464
RAC: 304
Australia
Message 1909704 - Posted: 30 Dec 2017, 23:54:31 UTC - in response to Message 1909702.  
Last modified: 31 Dec 2017, 0:21:45 UTC

The Server has decided to continue to Strangle 2 of my Linux machines.

It's not just Linux systems.
I've had to triple update my Win10 i7 several times this morning to keep the work coming.

Edit- even my Vista C2D has required a bump or 2.

Edit- and it's not surprising. Very little Arecibo work about, and no AP. When there is Arecibo & AP work available, then you can get MB without extra effort being needed.
Grant
Darwin NT
ID: 1909704 · Report as offensive
Grant (SSSF)
Volunteer tester

Send message
Joined: 19 Aug 99
Posts: 13351
Credit: 208,696,464
RAC: 304
Australia
Message 1909716 - Posted: 31 Dec 2017, 1:07:10 UTC - in response to Message 1909704.  
Last modified: 31 Dec 2017, 1:07:22 UTC

Edit- even my Vista C2D has required a bump or 2.

Make that several bumps.
Struggling to get work on either system.
Grant
Darwin NT
ID: 1909716 · Report as offensive
Profile Keith Myers Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 29 Apr 01
Posts: 13141
Credit: 1,160,866,277
RAC: 1,873
United States
Message 1909719 - Posted: 31 Dec 2017, 1:10:08 UTC

Only the linux machine is massively down. It is down about 200 tasks. It is getting about 11 tasks per request on average for the past couple of hours. Nowhere near enough to replenish tasks after task retirement per request. Triple Update isn't doing much for it.
Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)
ID: 1909719 · Report as offensive
juan BFP Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 16 Mar 07
Posts: 9786
Credit: 572,710,851
RAC: 3,799
Panama
Message 1909720 - Posted: 31 Dec 2017, 1:13:44 UTC
Last modified: 31 Dec 2017, 1:41:31 UTC

My host get a lot of WU but all are Arecibo Vlars, maybe that is why your host did not get any new GPU work.
ID: 1909720 · Report as offensive
Profile Keith Myers Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 29 Apr 01
Posts: 13141
Credit: 1,160,866,277
RAC: 1,873
United States
Message 1909723 - Posted: 31 Dec 2017, 1:22:27 UTC - in response to Message 1909720.  

The two computers that got any tasks of any amount (36) were Arecibo shorties and BLC. Need a couple of hundred each machine. Could be a Arecibo VLAR storm be the cause again?
Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)
ID: 1909723 · Report as offensive
Profile Brent Norman Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester

Send message
Joined: 1 Dec 99
Posts: 2786
Credit: 685,657,289
RAC: 835
Canada
Message 1909727 - Posted: 31 Dec 2017, 1:42:34 UTC - in response to Message 1909723.  

Nah, it's not a VLAR Storm. I have 3 computers that are down on CPU tasks too. Nothing is coming out the pipe ... or at least, not much.
ID: 1909727 · Report as offensive
Profile Keith Myers Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 29 Apr 01
Posts: 13141
Credit: 1,160,866,277
RAC: 1,873
United States
Message 1909730 - Posted: 31 Dec 2017, 1:56:40 UTC - in response to Message 1909727.  
Last modified: 31 Dec 2017, 1:58:25 UTC

Yes, I have to agree. The pattern is nothing coming out of the servers. Down on cpu tasks all machines too. So, no handicap there on VLAR's. The fastest cpu machines, the Ryzens can't get enough cpu work either. Lots of no tasks to send messages.

[Edit] The Haveland graphs support that assessment. Number of tasks in progress is dropping over the last couple of hours.
Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)
ID: 1909730 · Report as offensive
TBar
Volunteer tester

Send message
Joined: 22 May 99
Posts: 5204
Credit: 840,779,836
RAC: 2,768
United States
Message 1909731 - Posted: 31 Dec 2017, 1:57:25 UTC
Last modified: 31 Dec 2017, 1:57:49 UTC

In Progress seems to be dropping. I suppose that means the Server is sending less tasks. I thought a VLAR Storm was alleged to be impossible with BLC tasks running on GPUs.



A couple machines just got topped off, another is still down by over 200 tasks.
ID: 1909731 · Report as offensive
Grant (SSSF)
Volunteer tester

Send message
Joined: 19 Aug 99
Posts: 13351
Credit: 208,696,464
RAC: 304
Australia
Message 1909733 - Posted: 31 Dec 2017, 2:04:54 UTC - in response to Message 1909731.  

I thought a VLAR Storm was alleged to be impossible with BLC tasks running on GPUs.

We had one for several hours last week (or was it the week before?)
But this isn't that. It's just the problem with the Scheduler- little or no AP or Arecibo WUs means not even GBT work will be allocated when it's requested.
When the Arecibo and/or AP work becomes available again, then we'll be able to get GBT WUs again.
It's been this way for 12 months.
Grant
Darwin NT
ID: 1909733 · Report as offensive
Profile Brent Norman Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester

Send message
Joined: 1 Dec 99
Posts: 2786
Credit: 685,657,289
RAC: 835
Canada
Message 1909737 - Posted: 31 Dec 2017, 2:25:28 UTC - in response to Message 1909733.  

My read of the way the GBT splitters act is they run fine until the server cache reaches 600k, then they shut off and don't restart (or just dribble a little) and the Arecibo splitters take over.

When the server does it's Expired/TimedOut task check, the GBT splitters restart. I always see a blast of resends on a least 1 computer just before the GBT tasks reappear. They run for approx 1/2 an hour and shut down again.

So until the server does it's check, it is possible to get stuck in a VLAR storm with no GBT being split.
ID: 1909737 · Report as offensive
Profile Keith Myers Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 29 Apr 01
Posts: 13141
Credit: 1,160,866,277
RAC: 1,873
United States
Message 1909742 - Posted: 31 Dec 2017, 3:20:49 UTC

I'm down to my last 2 gpu tasks on the Linux cruncher. Guess it's going to do Einstein for the rest of the evening.
Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)
ID: 1909742 · Report as offensive
Previous · 1 . . . 10 · 11 · 12 · 13 · 14 · 15 · 16 . . . 36 · Next

Message boards : Number crunching : Panic Mode On (109) 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.