Rosetta remains blacked out this morning (6AM PDT)

Indeed - still down (18h20 GMT/UTC).

Based on previous experience with Rosetta, I expect we will not hear anything (e.g.at this forum) for a few days (up to a week) and then the site will come back with no explanation.

Hi!

I think there was a explanation on the rosetta website on Sunday. It said that they were moving hardware to a new building on the campus I think. They expected an outage of 4-5 days.

Regards,
Rayburner

MW is down for the install of the new server

I thought that might be the case -- is anyone back at the project running a pool on when it will be back up (not asking for a schedule -- schedule information providing software is part of a subsequent upgrade <G>).

MW is down for the install of the new server

Ah -- thanks for that. Sounds like Monday next then.

Hi!

I think there was a explanation on the rosetta website on Sunday. It said that they were moving hardware to a new building on the campus I think. They expected an outage of 4-5 days.

Regards,
Rayburner

I thought that might be the case -- is anyone back at the project running a pool on when it will be back up (not asking for a schedule -- schedule information providing software is part of a subsequent upgrade <G>).

MW is down for the install of the new server

New server is up but there are some IP Address Propogation issues that may take a couple more days to resolve.

Thanks for that -- by the way, if you don't have completed work not yet reported (ie nothing to lose in terms of completed work), you can get new work to process even now -- detatch and then re-attach -- will pick up the changes. Users might still have issues if they are working with proxy servers though.

New server is up but there are some IP Address Propogation issues that may take a couple more days to resolve.

Actually, I think I remember the explanation being that they were migrating Rosetta@home to more powerful servers in a campus datacenter. I would hope so because the last outage shows that its current servers take an extreme amount of time to recover from an outage and catch up on all the backlogs that each outage creates.

If I am wrong, please correct me. I do not want to raise false hopes.

I thought that might be the case -- is anyone back at the project running a pool on when it will be back up (not asking for a schedule -- schedule information providing software is part of a subsequent upgrade <G>).

MW is down for the install of the new server

New server is up but there are some IP Address Propogation issues that may take a couple more days to resolve.

We are up!

We are up!

Good, now be nice and send me some N-Body tasks. I need to test something. ;-)

Rosetta@home: Status? Where can I get an updated staus on the project? I have a test invironment running and use rosetta@home to stress test virtuel servers :-) and need more workload and have a lot of result to transfer.

Rosetta@home: Status? Where can I get an updated staus on the project? I have a test invironment running and use rosetta@home to stress test virtuel servers :-) and need more workload and have a lot of result to transfer.

I see an abbreviated homepage, with a message explaining the outage is now accessible but that is all. Don't expect anything until Monday.

Unfortunately, the User of the day's name on Rosetta@home's home pate as of this writing could be considered NSFW.

The NSFW name has been replaced with something safe at Rosetta@home's home page probably due to the UTC date incrementing.

Don't expect anything until Monday.

I'm pleased to have been proven wrong. Rosetta is back up and running.

Indeed, and aside from one IP glitch on one of the servers, resolved with a hosts file workaround, Rosetta appears in good shape.

Don't expect anything until Monday.

I'm pleased to have been proven wrong. Rosetta is back up and running.

MW and their home page went offline an hour or two ago (4PM PST)

MW remains offline -- they do have a 'down for maintenance' homepage up (which suggests the problem was MW project based, not RPI)

It's up:

The database was inadvertently put on a small partition on the new server, and it filled up. Matt A. fixed it, but somehow got the computer into a state where it needs a hard reboot, and we don't have anyone on duty who can do that at night.

MAJOR SCHEDULED BOINC SERVER OUTAGE AT WORLD COMMUNITY GRID:

Extended BOINC Server Outage Planned from 21:30 UTC Jan 13 through 8:00 UTC Jan 14th

We are going to be upgrading our database for the BOINC systems from MySQL 5.0 to MySQL 5.5. There have been substantial improvements in the MySQL database performance between these releases, and as we continue to grow, we need this additional performance.

In order to accomplish this change, we will need to stop all access to the BOINC database during this change. This will allow us to take a clean backup of the database and then upgrade the system. Due to the size of the database, we expect it to take some time to do this upgrade and thus the length of the outage.

During the outage the website and forums will work (except for the result status page which queries the BOINC database). Additionally, clients will be able to update result files and download input files. However, clients will not be able to communicate with the server to report completed results nor obtain new work to process.

Additionally, this change window occurs during the end of day statistics process. We will defer the end of day statistics process until after the database upgrade is complete.

We apologize for the disruption, but it is necessary to support the continued growth in your contribution.

thanks.
knreed [Jan 10, 2012 11:00:57 PM]