Message boards : News : Outage resolved
Author | Message |
---|---|
Before anybody wonders: yes there was an another outage last night. Thanks for the patience. | |
ID: 53899 | Rating: 0 | rate:
![]() ![]() ![]() | |
NIce to have you back Toni... | |
ID: 53900 | Rating: 0 | rate:
![]() ![]() ![]() | |
It's not clear but I don't think so, because it happens suddenly. When it happens, the web server gets stuck without real load. | |
ID: 53902 | Rating: 0 | rate:
![]() ![]() ![]() | |
Being a retired facilities manager, I was all ready to blame the custodial night crew for plugging in a floor buffer on the same circuit as Grosso and tripping the breaker and then forgetting to report it. Used to happen to our researchers now and then. | |
ID: 53907 | Rating: 0 | rate:
![]() ![]() ![]() | |
Wasn't just the web server. All my uploads and downloads were stalled out and backed off on all hosts. Finally cleared them this morning just now and getting new work finally. | |
ID: 53908 | Rating: 0 | rate:
![]() ![]() ![]() | |
Uploads not going through again. | |
ID: 53913 | Rating: 0 | rate:
![]() ![]() ![]() | |
GPUGRID 1u5tA02_413_1-TONI_MDADpr4su-2-10-RND2363_0_0 0.000 3.15 K 00:08:03 - 05:18:50 0.00 Kbps Upload pending (Project backoff: 03:07:59) Pipsqueek | |
ID: 53914 | Rating: 0 | rate:
![]() ![]() ![]() | |
GPUGRID 1yg9A03_379_4-TONI_MDADpr4sy-2-pdb_file 0.000 1595.02 K 00:48:24 - 03:17:48 0.00 Kbps Download pending (Retry in: 00:02:24), retried: 24 Darksider | |
ID: 53915 | Rating: 0 | rate:
![]() ![]() ![]() | |
@Toni ===>I seem to be having issues with more tasks showing up on the server as in progress than actually appear in my BOINC manager. I have 4 GPUs and 11 tasks in progress since the outage. I will list the unaccounted-for tasks if they persist tomorrow. | |
ID: 53917 | Rating: 0 | rate:
![]() ![]() ![]() | |
@Toni ===>I seem to be having issues with more tasks showing up on the server as in progress than actually appear in my BOINC manager. I have 4 GPUs and 11 tasks in progress since the outage. I will list the unaccounted-for tasks if they persist tomorrow. I have a few "ghost" WUs myself right now, (3 actually), with only 4 units downloaded. See link: http://www.gpugrid.net/results.php?hostid=263612&offset=0&show_names=0&state=1&appid= And a grand total of 3 "ghost" computer accounts: http://www.gpugrid.net/hosts_user.php?userid=19626 | |
ID: 53918 | Rating: 0 | rate:
![]() ![]() ![]() | |
@Toni ===>I seem to be having issues with more tasks showing up on the server as in progress than actually appear in my BOINC manager. I have 4 GPUs and 11 tasks in progress since the outage. I will list the unaccounted-for tasks if they persist tomorrow. As of a little more than hours, from the time I wrote this post, I had 2 active computers and 5 "ghost" computers. So, I used the merge computers by name to get the rid of the "ghosts" computers and my RAC and total credit doubled on the computer ID: 534811, see linked below. This happened by accident, not by intent. https://www.gpugrid.net/show_host_detail.php?hostid=534811&ipprivate=1 Though it did not effect my user account RAC and total credit, which are correct. http://www.gpugrid.net/show_user.php?userid=19626 Can this be corrected? | |
ID: 54128 | Rating: 0 | rate:
![]() ![]() ![]() | |
@Toni ===>I seem to be having issues with more tasks showing up on the server as in progress than actually appear in my BOINC manager. I have 4 GPUs and 11 tasks in progress since the outage. I will list the unaccounted-for tasks if they persist tomorrow. No as far as I know, it can't. That number comes from the servers and you don't have the ability to modify. The server just glitched. This same thing happened to one of my teammates (juan BFP) when he merged a host and he suddenly picked up almost a billion credits. Didn't affect the host RAC or any other adverse effect than instantly catapult him to the top of the total credits leaderboard. He just says the "the servers liked him" https://stats.free-dc.org/stats.php?page=hosts&proj=sah&sort=today | |
ID: 54142 | Rating: 0 | rate:
![]() ![]() ![]() | |
Can't upload for at least last 6hours. | |
ID: 54165 | Rating: 0 | rate:
![]() ![]() ![]() | |
Yes. My uploads are (project delayed). | |
ID: 54166 | Rating: 0 | rate:
![]() ![]() ![]() | |
Same. Rosetta is also stalled, but it look like they're out of tasks at the moment. Any other BOINC covid-19 projects suggested? Let's go smash through their jobs too! | |
ID: 54174 | Rating: 0 | rate:
![]() ![]() ![]() | |
Also TN-GRID is studying the relationship between some specific genes and COVID-19. | |
ID: 54180 | Rating: 0 | rate:
![]() ![]() ![]() | |
Also TN-GRID is studying the relationship between some specific genes and COVID-19. Thanks for sharing. Rosetta, TN-Grid and F@H. https://gene.disi.unitn.it/test/forum_thread.php?id=265 | |
ID: 54181 | Rating: 0 | rate:
![]() ![]() ![]() | |
Ralph@Home, beta project of Rosetta@Home, is testing Android tasks. If you have an Android device, you can help this project too. | |
ID: 54189 | Rating: 0 | rate:
![]() ![]() ![]() | |
I'm maintaining a list of COVID-19 related projects here. Hat tip to https://twitter.com/TheBoinc for helping me keep this current. | |
ID: 54196 | Rating: 0 | rate:
![]() ![]() ![]() | |
I'm maintaining a list of COVID-19 related projects here. Hat tip to https://twitter.com/TheBoinc for helping me keep this current. Nice! The WCG link is broken. This one maybe? https://www.ibm.org/OpenPandemics | |
ID: 54216 | Rating: 0 | rate:
![]() ![]() ![]() | |
I'm maintaining a list of COVID-19 related projects here. from what it looks, with none of these projects one can choose COVID-research exclusively :-( | |
ID: 54221 | Rating: 0 | rate:
![]() ![]() ![]() | |
I'm maintaining a list of COVID-19 related projects here. Hat tip to https://twitter.com/TheBoinc for helping me keep this current. Thanks for pointing this out, it's fixed now. That was actually supposed to link to a WCG forum post and I must have just not pasted it in, but this is better. I'm doing my best to keep this list current, let me know if you're aware of other projects that should be listed, or errors in the list. None of these projects allow you to exclusively select COVID-19 work as far as I know, but FAH and Rosetta seem to be giving that work top priority - if you can get it. ____________ Team USA forum | Team USA page Join us and #crunchforcures. We are now also folding:join team ID 236370! | |
ID: 54222 | Rating: 0 | rate:
![]() ![]() ![]() | |
FAH and Rosetta seem to be giving that work top priority - if you can get it. lately, I had little success with Rosetta. Most of the time 900.000+ tasks in process, but no new ones in the download queue. | |
ID: 54223 | Rating: 0 | rate:
![]() ![]() ![]() | |
FAH and Rosetta seem to be giving that work top priority - if you can get it. I think it's because R@h sends an absurd number of WUs to each computer, far more than it can complete by the deadline. Many BOINCers are probably setting a long queue time and hording them. I abort a lot of them so they can go to others. It's a shame the queue time is a global setting and not project specific. | |
ID: 54235 | Rating: 0 | rate:
![]() ![]() ![]() | |
I think it's because R@h sends an absurd number of WUs to each computer, far more than it can complete by the deadline. Many BOINCers are probably setting a long queue time and hording them. I abort a lot of them so they can go to others. Im am wondering why R@h can't apply a certain limit to the number of downloaded tasks, like GPUGRID is doing (2 tasks per GPU). This probably would solve the problem immediately. | |
ID: 54236 | Rating: 0 | rate:
![]() ![]() ![]() | |
That was certainly the case in my first scheduler connection after joining. Got 260 tasks in my first download before I could set NNT. | |
ID: 54238 | Rating: 0 | rate:
![]() ![]() ![]() | |
Starting a new project for a PC will often result in too many tasks as the ETA don't represent real run times. After a bit that will correct itself. It's recommended to set a low queue when starting new projects instead of a multi day queue to make it through Tuesday maint. | |
ID: 54240 | Rating: 0 | rate:
![]() ![]() ![]() | |
Starting a new project for a PC will often result in too many tasks as the ETA don't represent real run times. After a bit that will correct itself. It's recommended to set a low queue when starting new projects instead of a multi day queue to make it through Tuesday maint. I didn't have a choice. It picked up my Einstein cache settings since cache settings is global and propagates across all projects. | |
ID: 54245 | Rating: 0 | rate:
![]() ![]() ![]() | |
I am running a custom client now in which I have direct control over the number of tasks I want for every project. Do tell. I didn't think that was even possible from the client side. Is this client available? | |
ID: 54248 | Rating: 0 | rate:
![]() ![]() ![]() | |
I am running a custom client now in which I have direct control over the number of tasks I want for every project. No. It is restricted to GPUUG team members. | |
ID: 54252 | Rating: 0 | rate:
![]() ![]() ![]() | |
It's open source. SETI users made a custom version to get more tasks since SETI shut down every Tues. | |
ID: 54311 | Rating: 0 | rate:
![]() ![]() ![]() | |
It's open source. SETI users made a custom version to get more tasks since SETI shut down every Tues. That is a different version. The one Keith refers to is restricted only to GPUUG members. ____________ ![]() ![]() | |
ID: 54313 | Rating: 0 | rate:
![]() ![]() ![]() | |
It's open source. SETI users made a custom version to get more tasks since SETI shut down every Tues. I'm saying anyone can go make the edits themselves. | |
ID: 54358 | Rating: 0 | rate:
![]() ![]() ![]() | |
It's open source. SETI users made a custom version to get more tasks since SETI shut down every Tues. Yes, the BOINC client code is open source and licensed under the GPL. So up to you what you do with it. That is exactly what we have done in the GPUUG. Added new features and fixed a lot of BOINC issues to get it to run well with Seti. A lot of new features work well with other projects. We simply restricted it to GPUUG because of the possible damage to the Seti database if used indiscriminately and if the modified client got out into the wild and used by the masses of Seti volunteers. Our concern was validated when the project unexpectedly increased the amount of work sent out to all users without warning and the database and project crumpled under the increased load from swelled caches. | |
ID: 54364 | Rating: 0 | rate:
![]() ![]() ![]() | |
So how about remove the seti idiosyncrasies to the benefit of all BOINCers??? | |
ID: 54376 | Rating: 0 | rate:
![]() ![]() ![]() | |
Message boards : News : Outage resolved