Milkyway is back up.

Looks like there was a problem with one of the hard drives. We've replaced it and it looks like the database is back into RAM, so hopefully we'll have some smooth sailing from here on out.

There are significant problems on POEM

The system has been changed but is far from working.

Firstly it is impossible to tell POEM about the problems as I cannot logon to their system as I am told the URL is wrong.

Every work unit has the same characteristics 347 gflops. On my system this would take two minutes, so far it has run for nearly two hours.

I have subsequently aborted another 20 units as they all say 347 Gflops and 49 secs.

Every work unit does not have a proper time each one say 49 secs, as you process time remaining goes up not DOWN.

This really is a very bad implementation of a system change .

Considering we "donate" computer time this change seems to be a waste of resources.

This thread is for reporting project outages, the POEM project isn't really down.
If you want to rant about them, or ask for help, there's a whole forums outside this thread where you can start a new thread. :-)

In the mean time, to log in on POEM, follow http://boinc.fzk.de/poem/get_passwd.php, fill in your email address, click the OK button under that, go to your email account, open the email from the POEM project, click the temporary log in link in it. This link allows you to log in for 24 hours. Use it to change or redo your password. The old password isn't needed to change to a new password.

After that you are logged in again. Then don't dilly dally, but go to this thread and tell Timo what you see.

It looks like the server at ABC@home has gone down. I can't upload anything, make any scheduler requests, or connect to its website.

It looks like ABC@home is back up again.

CPDN main project

CPDN will be taken offline at 9.00 am GMT on Thursday 10 November 2011.

This is to facilitate the relocation of our project servers. It is anticipated that the downtime will be for no longer than 48 hours, but the project should be considered 'at-risk' until noon on 14 November 2011 GMT.

Jonathan

_________________
Jonathan Miller
CPDN SysAdmin

It will still be possible to register and post on the independent forum:

http://climateprediction.net/board/index.php

FreeHAL is MIA. Or AWOL. Can't decide.

From what BOINC and FireFox will tell me the server(s) are down. Odd that all my machines were able to upload their work, but I can't get any more. :|

CPDN remains offline (5PM GMT 14 November). Not really a surprise, it seems the nature of CPDN servers and environment in general lends to '2x' downtime cycles when doing planned maintenance.

Hopefully, we will see some form of update somewhere from one of the folks from CPDN to get an updated 'return to life'.

CPDN main project

CPDN will be taken offline at 9.00 am GMT on Thursday 10 November 2011.

This is to facilitate the relocation of our project servers. It is anticipated that the downtime will be for no longer than 48 hours, but the project should be considered 'at-risk' until noon on 14 November 2011 GMT.

Jonathan

_________________
Jonathan Miller
CPDN SysAdmin

It will still be possible to register and post on the independent forum:

http://climateprediction.net/board/index.php

Both CPDN and its Beta project are up and running again.

Jonathan worked on the servers through the weekend but was faced with a network problem which couldn't be sorted out until the non-CPDN Oxford Uni IT staff were back at work on Monday.

Some hadam3p regional model files are still unable to upload; this problem will probably be fixed on Wednesday UK time but it will take some time for the large backlog of waiting files to upload.

Both CPDN and its Beta project are up and running again.

Jonathan worked on the servers through the weekend but was faced with a network problem which couldn't be sorted out until the non-CPDN Oxford Uni IT staff were back at work on Monday.

Some hadam3p regional model files are still unable to upload; this problem will probably be fixed on Wednesday UK time but it will take some time for the large backlog of waiting files to upload.

...and, from...

http://climateprediction.net/board/viewtopic.php?f=36&t=5927&start=150#p97158 ...

Re: News

Unread postby Les Bayliss » Fri Nov 18, 2011 1:12 am
There's still a problem for a few uploads, as the university's IT people haven't yet updated the relevant DNS records for the servers that were moved.
So more thumb twiddling. :roll:

(Can't post to that thread...)

Hi Folks,

FYI: As of a.m. Nov 19 my CPDN Transfers queue is populated with 23 items totaling over 500MB of .zip files, and my CPDN Tasks queue is populated with 3 completed HADAM3P's, all waiting for upload. The pile continues to grow since the Nov 10th "offline". This is to be expected?

Thankfully my service provider doesn't cap my uploads... :)

Peace,
Jimmy G

The Albert@Home project (Einstein's test project) will not have any downloads due to server trouble this weekend. Nothing yet said about Einstein themselves.

The Albert@Home project (Einstein's test project) will not have any downloads due to server trouble this weekend. Nothing yet said about Einstein themselves.

There is this in their Tech News forum posted 18th of Nov

BRP4 & FGRP1 download (server) problems
We are experiencing problems with the download server for BRP4 and FGRP1. We're working on it. During that time we may send out only work for S6Bucket or no work at all.

BM

FYI: As of a.m. Nov 19 my CPDN Transfers queue is populated with 23 items totaling over 500MB of .zip files, and my CPDN Tasks queue is populated with 3 completed HADAM3P's, all waiting for upload. The pile continues to grow since the Nov 10th "offline". This is to be expected?

FYI: As of a.m. Nov 21 all CPDN Transfers queue uploaded successfully, all CPDN Tasks queue uploaded successfully. Kudos CPDN team!

:)
JG

The Milkyway site is unavailable.

---

The Milkyway site is unavailable.

Sorry to get back to y'all so late. I've been in and out of meetings since Noon Eastern. This was escalated by myself @ 2:30.

Here's hoping that things are fixed on Wednesday, elsewise this outage might stretch into Monday. Hope folks have other GPU processing projects in place.

Sorry to get back to y'all so late. I've been in and out of meetings since Noon Eastern. This was escalated by myself @ 2:30.

:). Yes thankyou. Education curve.

Pirates@Home is in drydock.

22 November 2011 20:18 UTC
    Pirates@Home is undergoing a server software upgrade.
    Check back later to see if we are still hosed.
                          - Captain Wormholio 

And so we begin day two of the Milkyway outage -- since no home page access, there is no information available regarding the outage. The lack of home page access suggests the possibility of an RPI problem rather than something within the perview of project folks.

SETI, which had not yet recovered from their weekly maintenance outage is now fully offline. There are times that it seems their 4 hour maintenance cycle on Tuesdays is more of an 'unmaintenance' cycle as recovering from the maintenance on occasion takes many times longer than the actual maintenance.