1)
Message boards :
Number crunching :
Problems and Technical Issues with Rosetta@home
(Message 81445)
Posted 15 Apr 2017 by Keith E. Laidig Post: Hey folks - Happy Easter to all. Yesterday, in response contributors concerns about our aged SSL cipher handling on the R@H services, we made changes the webserver configurations. For reasons I don't yet understand, these changes resulted in upload timeouts and the slow collapse of project backend. Regrettably, I had 'gone offline' yesterday evening to celebrate the season with family and was unaware of the problems until this AM [ADT]. We've reverted to previous configurations and restarted the project. We apologize for the 'outage' and will keep monitoring the situation closely until convinced things are working. FYI - we are in the final stages of building out a new, shiny, high-powered BOINC system for R@H. More info coming.... -KEL [Update] It looks like there are still problems.... |
2)
Message boards :
Number crunching :
Problems and Technical Issues with Rosetta@home
(Message 81285)
Posted 10 Mar 2017 by Keith E. Laidig Post: Folks! Thanks for your patience and support during this outage. We appreciate the support of the BOINC community! You are an impressive crew! -KEL |
3)
Questions and Answers :
Web site :
SSL certificate rating (C)
(Message 79962)
Posted 28 Apr 2016 by Keith E. Laidig Post: Howdy. I'll look into this and see what I we can do. -KEL |
4)
Message boards :
Number crunching :
Problems and Technical Issues with Rosetta@home
(Message 77142)
Posted 30 Jul 2014 by Keith E. Laidig Post: This appears to be an upstream issue for us. Something has changed with the campus routing to 'throttle' traffic into our servers. We're working with the UW's Network Operations team to hunt down the cause.... -KEL |
5)
Message boards :
Number crunching :
Looking into Rosetta's source code?
(Message 72008)
Posted 8 Jan 2012 by Keith E. Laidig Post: Markus, Sorry, but no, you can't get the code that is being run in R@H at the moment. It is in active development and the scientists do not release all development branches of the code tree to the public prior to publication of their work. You could, in principle, get a copy of the base code through the UW - as you know. Thanks for your interest. |
6)
Message boards :
Number crunching :
What's been up with R@H?
(Message 56043)
Posted 26 Sep 2008 by Keith E. Laidig Post: I can confirm the project was down (by my own observation), and presume that the scheduler is very busy right now. My hosts have been requesting work, but sometimes getting responses back from the scheduler with no new work, just confirming completed results. It will take a couple of hours for things to get more normal. Work is now flowing. But perhaps somewhat sporatically. The back-end fileserver experienced a kernel panic while updating the filesystem journal. There is a tremendous amount of I/O on this old machine and sometimes it doesn't keep up well. We're implementing a new back-end filesystem for R@H - along the line of the SAN that was attempted last year - and plan to move R@H over in the next few weeks. The amount of data that exists and the rapidity with which your clients update it makes the transfer of data very challenging indeed... |
7)
Message boards :
Rosetta@home Science :
new supercomputer
(Message 54958)
Posted 6 Aug 2008 by Keith E. Laidig Post: Actually, we are in the planning stages of moving our public automatic Rosetta server, Robetta, to BlueWaters. Our present "Robetta system" allows scientists anywhere to access Rosetta's methods but is brutally over-subscribed. A large number of scientists waiting for the service to return results have given up on us. This has distressed us but we just don't have the resources to fill the need. We have been collaborating w/ NCSA to use the cycles in some of their installed clusters to try and knock down the queue length but it has only been partially successful. So, we're going to go big-time on this and see if we can't make BlueWaters/Robetta a nearly instantaneous Rosetta service for researchers around the globe. There's LOTS and LOTS of scaling issues to be addressed but we're looking at them and applying for $$$ to hire some folks to help w/ all of this. Anyway, I thought it was interesting enough to post a bit about BlueWaters.... we'll keep you abreast of our progress along these lines.... |
8)
Message boards :
Number crunching :
Problems with web site
(Message 54901)
Posted 4 Aug 2008 by Keith E. Laidig Post: Well that was interesting -- Rosetta was flat offline for about 6 hours -- back up now -- but no indication that anyone else noticed. The back-end fileserver had crashed - kernel panic. I've just returned from a quick run to campus to reboot that server... |
9)
Message boards :
Rosetta@home Science :
DISCUSSION of Rosetta@home Journal (3)
(Message 52781)
Posted 29 Apr 2008 by Keith E. Laidig Post: On the updated "Articles about Rosetta and Rosetta@home" page (http://boinc.bakerlab.org/rosetta/rah_articles.php), the link to "RCSB Protein Data Bank Molecule of the Month: Designer Proteins" (http://www.rcsb.org/pdb/molecules/pdb70_1.html) is dead. Here is a working URL: http://www.rcsb.org/pdb/static.do?p=education_discussion/molecule_of_the_month/pdb70_1.html. Thanks, incorporated. -KEL |
10)
Message boards :
Rosetta@home Science :
amyloid fibril structure prediction
(Message 50971)
Posted 25 Jan 2008 by Keith E. Laidig Post: I am excited to be a part of your research I have a personal interest in your successes. My Father passed away in 1987 from complications due to amyloidosis and I proudly support your efforts. If donating my PCs' computational capacity is a way I can help, then you have what I can give. Time to invest in a new dedicated quad-core box or two. Welcome to the project. We appreciate your contribution! -KEL |
11)
Message boards :
Number crunching :
need help with host-id/credit issue
(Message 50560)
Posted 11 Jan 2008 by Keith E. Laidig Post: What was the old hosts' id? -KEL Hi! |
12)
Message boards :
Rosetta@home Science :
CAN YOU HELP ROSETTA??
(Message 50260)
Posted 2 Jan 2008 by Keith E. Laidig Post: Specifically, all data-distribution and communication from our servers are 'pulled' by the client, at the client's request, from the server. We don't make un-solicited communications with client computers, excepting rare project update emails to those who've elected to receive them. For BOINC to work on your computer, firewall configs need only allow outbound connection to ports 80 (http) and 443 (https). [BOINC running locally on your computer needs to be able to connect back to other ports on your computer as well.] Sure. The primary weakness in computer security is the individual touching the keyboard. So, in a controlled environment such as a hospital I'd say: Yes, if a given computer is user primarily to retrieve data: information kiosks, nursing/physician stations, office desktops, cafeteria registers, etc., the vast majority of the installed computer systems. No, if the computer is a source for customer/patient information or provides dynamic information (monitors, etc). {Not because of BOINC's possible security issues, but because simplicity is the key to timely and efficient troubleshooting of critical infrastructure - problems could be masked or made more complex by BOINC's behavior.} [It won't happen here in the US, at least in hospitals, because of the need for medical practitioners and researchers to comply with the HIPPA act (the Health Insurance Portability and Accountability Act which mandates that healthcare providers and health plans protect the privacy of patient records). We, the Baker Group, have to comply with the HIPPA act as well.] But, if a company/organization wished to harness the idle cycles of their installed computer base, they could implement a BOINC project within their own network, or donate their idle time to a 'trusted' BOINC project. For example, the University of Washington has a large base of installed desktop machines under central control and administration. They run BOINC on their installed base (if the individual users allows) and their contribution to our work is considerable - see the Nebula project. |
13)
Message boards :
Cafe Rosetta :
Happy new year
(Message 50232)
Posted 1 Jan 2008 by Keith E. Laidig Post: A hearty Happy New Year from the Baker group IT here in Seattle to all of you who've made Rosetta@home such a success! Here's hoping that 2008 bring health and happiness to you and yours. Cheers! -KEL |
14)
Questions and Answers :
Web site :
Result plots not working?
(Message 49752)
Posted 17 Dec 2007 by Keith E. Laidig Post: When trying to visit the Results page I always get a time-out. The result page opened from within the client works, but you can't view the plotted results and so far found minimum RMSD there. Whoops! Perhaps the server is down. I've asked DK to check it out. Thanks for the information... -KEL |
15)
Message boards :
Number crunching :
Project encountered internal error: shared memory
(Message 49165)
Posted 28 Nov 2007 by Keith E. Laidig Post: Looks like the problem is back. Get this error trying to return any results to this project. I've alerted DK to the problem. Thanks, -KEL. |
16)
Message boards :
Number crunching :
Whats up with Ralph website?
(Message 49142)
Posted 28 Nov 2007 by Keith E. Laidig Post: Anyone know why Ralph has been down all day? Hmmm, the system is down. I'm not on campus today - I'll see if I can get someone there to take a quick look... -KEL |
17)
Message boards :
Number crunching :
Who Can I Contact?
(Message 48604)
Posted 13 Nov 2007 by Keith E. Laidig Post: Who can I contact regarding a team that is listing itself as a government agency that in no way belongs in that category? send the info to issuesATbakerlab.org w/ AT=@. And I'll take a look... -KEL |
18)
Message boards :
Rosetta@home Science :
CAN YOU HELP ROSETTA??
(Message 48561)
Posted 11 Nov 2007 by Keith E. Laidig Post: I myself is a contributor in the rosetta@home science project and I would like to see that Boinc gets installed on workplaces here in Sweden. Although, since some things concerns personal privacy and therefore security must be very high in some places, I guess that it isn't possible to install Boinc everywhere, unfortunately. The general security issues involved in distributed computing such as BOINC (and R@H) are discussed here along with the approaches taken to address these issues. Specifically, all data-distribution and communication from our servers are 'pulled' by the client, at the client's request, from the server. We don't make un-solicited communications with client computers, excepting rare project update emails to those who've elected to receive them. For BOINC to work on your computer, firewall configs need only allow outbound connection to ports 80 (http) and 443 (https). [BOINC running locally on your computer needs to be able to connect back to other ports on your computer as well.] -KEL |
19)
Message boards :
Number crunching :
Pending Credits ?
(Message 46444)
Posted 17 Sep 2007 by Keith E. Laidig Post: Thanks for the head's up on this. DK will set things right when he gets to lab... -KEL |
20)
Message boards :
Number crunching :
Welcome Back!
(Message 45828)
Posted 9 Sep 2007 by Keith E. Laidig Post: thanks for the detailed messages. The filesystem failure caused some files to be lost - but even more troublesome is that a number of file ownership and permission problems arose. What a pain. We're slowly finding the permission/ownership conflicts that stop the system from running and I appreciate your detailed help here! -KEL |
©2024 University of Washington
https://www.bakerlab.org