Failing downloads.

Wiggo
Wiggo
Joined: 29 Oct 10
Posts: 10
Credit: 243184304
RAC: 0
Topic 221468

I've just connected this rig, https://einsteinathome.org/host/12791512, and while everything else downloaded fine the actual work units themselves are failing with a "permanent HTTP error".

 

Is the problem mine or is it the servers?

 

Cheers.

Wiggo
Wiggo
Joined: 29 Oct 10
Posts: 10
Credit: 243184304
RAC: 0

Well it looks like the

Well it looks like the problem wasn't mine as they're downloading now.

 

Now I'll see how these first 2 GPU units go.

 

Cheers.

Gary Roberts
Gary Roberts
Moderator
Joined: 9 Feb 05
Posts: 5851
Credit: 110737191884
RAC: 32494716

Wiggo wrote:Is the problem

Wiggo wrote:
Is the problem mine or is it the servers?

By the look of things, it might be some sort of (hopefully temporary) server glitch.

The GW tasks you are doing require the downloading of a fairly large set of data files.  Your tasks list shows 4 separate attempts to download the data files for 4 separate pairs of tasks.

The first 3 pairs gave download errors but the 4th is still "in progress".

So, to me it looks like there may be a server-side problem preventing the downloading of the necessary large data files for each of the pairs of tasks previously sent.  Can you please open BOINC Manager (Advanced view) and go to the transfers tab and check if the large data files show progress in downloading?  There should be lots of these files, downloading in pairs and there should be a downloading rate and % progress that are continually updating.  Please report the sorts of values you are seeing.  If no progress is being made, please check if somehow your end is blocking.

EDIT:  Hadn't seen your 2nd post - it must have arrived whilst I was responding to your first one.  So yes, some sort of server glitch that has now corrected itself.  Things are probably under a bit of pressure with all the new arrivals :-).

Cheers,
Gary.

Gary Roberts
Gary Roberts
Moderator
Joined: 9 Feb 05
Posts: 5851
Credit: 110737191884
RAC: 32494716

Wiggo wrote:Now I'll see how

Wiggo wrote:
Now I'll see how these first 2 GPU units go.

The biggest problem will be the poor initial values for estimated crunch time.  Don't be surprised to see 5 times or more as compared to the initial estimates.  The very first task to finish will cause (via DCF - duration correction factor) the estimates for further tasks to be corrected.

Part of the reason for this is probably that not all of the calculations use the GPU.  Some stuff still has to run on a CPU core.  If you have lots of CPU cores, you can gain significant throughput by running concurrent tasks on each GPU.  Since you only have 4 cores, running 2 tasks per GPU may not give as much benefit as you might expect, since all your cores would be somewhat over-utilized.  It's probably worth a test though :-).

Cheers,
Gary.

Falconet
Falconet
Joined: 9 Mar 09
Posts: 47
Credit: 15265451
RAC: 1045

Just got this problem. A

Just got this problem. A bunch of tasks failing to download.

 

30/03/2020 22:04:18 | Einstein@Home | Giving up on download of h1_1488.25_O2C02Cl4In0.GWV0: permanent HTTP error

perlind
perlind
Joined: 24 Mar 20
Posts: 3
Credit: 13036839
RAC: 0

Add me to the list.Could

Add me to the list.

Could the change to Daylight savings past weekend, have something to do with it?

Cheers

 

Boinc Stats

Maike-Sophie Mittelstädt
Maike-Sophie Mi...
Joined: 22 Nov 19
Posts: 2
Credit: 21484313
RAC: 26257

Yeah same Problem

Yeah same Problem here 

 

31.03.2020 06:06:45 | Einstein@Home | Started download of h1_1509.55_O2C02Cl4In0.aiDz
31.03.2020 06:06:45 | Einstein@Home | Started download of l1_1509.55_O2C02Cl4In0.aiDz
31.03.2020 06:06:48 | Einstein@Home | Giving up on download of h1_1509.55_O2C02Cl4In0.aiDz: permanent HTTP error
31.03.2020 06:06:48 | Einstein@Home | Giving up on download of l1_1509.55_O2C02Cl4In0.aiDz: permanent HTTP error
31.03.2020 06:06:49 | Einstein@Home | Started download of h1_1509.60_O2C02Cl4In0.nrF7
31.03.2020 06:06:49 | Einstein@Home | Started download of l1_1509.60_O2C02Cl4In0.nrF7
31.03.2020 06:06:52 | Einstein@Home | Giving up on download of h1_1509.60_O2C02Cl4In0.nrF7: permanent HTTP error
31.03.2020 06:06:52 | Einstein@Home | Giving up on download of l1_1509.60_O2C02Cl4In0.nrF7: permanent HTTP error
31.03.2020 06:06:53 | Einstein@Home | Started download of h1_1509.65_O2C02Cl4In0.2Gk-
31.03.2020 06:06:53 | Einstein@Home | Started download of l1_1509.65_O2C02Cl4In0.2Gk-
31.03.2020 06:06:55 | Einstein@Home | Giving up on download of h1_1509.65_O2C02Cl4In0.2Gk-: permanent HTTP error
31.03.2020 06:06:55 | Einstein@Home | Giving up on download of l1_1509.65_O2C02Cl4In0.2Gk-: permanent HTTP error
31.03.2020 06:06:56 | Einstein@Home | Started download of h1_1509.70_O2C02Cl4In0.TNoZ
31.03.2020 06:06:56 | Einstein@Home | Started download of l1_1509.70_O2C02Cl4In0.TNoZ
31.03.2020 06:06:57 | Einstein@Home | Giving up on download of h1_1509.70_O2C02Cl4In0.TNoZ: permanent HTTP error
31.03.2020 06:06:57 | Einstein@Home | Giving up on download of l1_1509.70_O2C02Cl4In0.TNoZ: permanent HTTP error

Gary Roberts
Gary Roberts
Moderator
Joined: 9 Feb 05
Posts: 5851
Credit: 110737191884
RAC: 32494716

Apparently the problem was

Apparently the problem was caused by some files being missing from the download mirrors.  I've been informed that this has been rectified so there should be no further problems giving "permanent HTTP error" messages of this type.

Please report here if you see any further examples of this particular message.

Cheers,
Gary.

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.