Message boards : Projects : Anything and Everything to do with (WCG) World Community Grid
Message board moderation
Previous · 1 · 2 · 3 · 4 · 5 · 6 . . . 14 · Next
Author | Message |
---|---|
Send message Joined: 31 Dec 18 Posts: 293 |
...it will take about 74 hours. That’s just the move - allow time for the resultant databases to be checked for integrity etc. I’d be surprised if the move was not 24/7 but give them at least a day or so to verify before they restart. |
Send message Joined: 30 Mar 20 Posts: 392 |
New Update, as an updated first post in an old thread. All of a sudden those 74 hours counting from Friday March 24, is now estimated to something much higher. Estimation now for the restart, is "later this week" https://www.worldcommunitygrid.org/forums/wcg/viewthread_thread,45101_offset,0#683202 I do not expect any real life from BOINC WCG until mid April, at the earliest. More likely is mid May. |
Send message Joined: 30 Mar 20 Posts: 392 |
The WCG team have said many times now, that WCG will be back soon. So, how long time can "soon" mean? |
Send message Joined: 8 Mar 23 Posts: 10 |
will April be the month we return to crunching? Stay tuned to find out. |
Send message Joined: 30 Mar 20 Posts: 392 |
Not the final countdown I fear, but more like the continuous countup. Last update host XML 2023-02-28 13:11:22 UTC (28 days 19:29:42 old) Last update user XML 2023-03-01 01:21:01 UTC (28 days 07:20:03 old) Last update team XML 2023-03-01 01:21:01 UTC (28 days 07:20:03 old) |
Send message Joined: 30 Mar 20 Posts: 392 |
New update from Igor Jurisica. Unfortunately more HDD problems that is stopping the restart of BOINC. https://www.worldcommunitygrid.org/forums/wcg/viewthread_thread,45101_offset,80#683642 "Thank you for your patience. Unfortunately - the recovery ran into additional HDD failures. With the help of the data center we had to rebuild the drive onto a hot spare. This is now complete and the transfer of /science is about to resume. We managed to get 20 out of 36TB moved to the new storage already. In the meantime, we reworked the back-end that was causing some devices not being recognized. thank you igor" |
Send message Joined: 10 May 07 Posts: 1419 |
This is starting to look like one huge April Fool's joke. The lastest prediction from my crystal ball ... earliest to resume all of WCG: Now mid-April 2033. * Then it will crash & burn in less than 24 hours under the stress from everyone trying to send back all the old data and get new work. |
Send message Joined: 10 May 07 Posts: 1419 |
The most recent news update was almost 24 hours ago from Dr J@Krembil: There are many more frustrating aspects I do not need to share with you all. |
Send message Joined: 28 Jun 10 Posts: 2636 |
The lastest prediction from my crystal ballCan't be doing with this new newfangled technology. I think a trip to Delphi is in order. |
Send message Joined: 10 May 07 Posts: 1419 |
The lastest prediction from my crystal ballCan't be doing with this new newfangled technology. I think a trip to Delphi is in order. Lol, Maybe I should also put up a dart board with the WCG logo in the middle and throw 10 darts at it while blindfolded. Then see how many hit or get closest to the bullseye adding a bit more statistics to my S.W.A.G.? It might just improve the accuracy. |
Send message Joined: 3 Mar 23 Posts: 10 |
Late Friday and it looks like the website is down again. I went to get a quick status update from the News forum and that is giving me "Error 500: java.lang.NullPointerException" while my stats page is back to the "System error" message. One step forward, two steps back? |
Send message Joined: 30 Mar 20 Posts: 392 |
New Update: https://www.worldcommunitygrid.org/forums/wcg/viewthread_thread,45101_offset,0#683202 March 31, 2023 update Data transfer update & maintenance check Earlier in the week, we ran into HDD failures while transferring the data from recovery storage to the new storage system. The issue has since been resolved and we have resumed transferring data, expecting to finish it by 5pm today. At 4pm, we will be conducting a brief maintenance on the website and forums to transfer the DB2 filesystems to our new storage system, which will result in restricted access to the website for up to 30 minutes. If all goes well, this could be the final step towards the full storage system upgrade. We evaluated the possibility of starting download of processed WUs, while not sending new WUs out. It was determined that the risk of complications that might result from doing this with incomplete information available to our scheduler and BOINC or any other unforeseen issues is too high. We have extended the deadlines for workunits that were processed and await upload to WCG. While we wait for the data transfer to finish, we are working on resolving other long standing issues such as device recognition. |
Send message Joined: 30 Mar 20 Posts: 392 |
And another FB/Twitter update 25 minutes ago: "DB2 maintenance has finished, and both the website and forums are accessible again. Finally, we now run from the new SSD storage system. The science DB transfer has been delayed in the process, now estimated to finish by 6pm. Thank you for your support." |
Send message Joined: 26 Mar 11 Posts: 188 |
And another FB/Twitter update 25 minutes ago: For what it is worth the Page loading times are quick at the moment |
Send message Joined: 30 Mar 20 Posts: 392 |
My finished tasks are all uploaded now. Reporting of finished tasks will happen when it happens. That part of BOINC isn't turned on yet. |
Send message Joined: 7 Apr 13 Posts: 64 |
We evaluated the possibility of starting download of processed WUs, while not sending new WUs out. It was determined that the risk of complications that might result from doing this with incomplete information available to our scheduler and BOINC or any other unforeseen issues is too high. I was thinking it might help the overall situation if we set WCG to "No New Tasks." That way once the finished WUs start reporting, that traffic would not compete with new tasks downloading. Perhaps wait a few days after reporting to start downloading new WUs. But after reading the above quote, it appears either that won't help much or WCG doesn't have information on that situation. However the pipeline is likely to be completely overloaded for while - seems like reducing total traffic for a brief period (by delaying DL of new tasks) would help the entire system recover more quickly? Waddya think? |
Send message Joined: 25 May 09 Posts: 1295 |
It would be a marginal help, as they wouldn't be getting "send work" requests, to which they respond with "no work available" or "download server is down" messages. |
Send message Joined: 30 Mar 20 Posts: 392 |
Well, they did not follow their own evaluation, since they after all did download, processed WU's, without sending new WU's, or allow reporting of the downloaded processed WU's. So, I don't know what happened there. if they changed their mind, or if "the left hand doesn't know what the right hand is doing". Edit: The "feeder" is still not running. |
Send message Joined: 10 May 07 Posts: 1419 |
Something good is happening at WCG! My 3 tasks that have been trying to upload since early March 1st UTC time have uploaded & reported within the last hour I see. |
Send message Joined: 29 Aug 05 Posts: 74 |
|
Copyright © 2024 University of California.
Permission is granted to copy, distribute and/or modify this document
under the terms of the GNU Free Documentation License,
Version 1.2 or any later version published by the Free Software Foundation.