Database Crash

Author	Message
ChrisD Send message Joined: 28 Sep 04 Posts: 8 Credit: 61,409 RAC: 0	Message 12552 - Posted: 28 Jan 2006, 17:19:37 UTC On Jan 26 LHC had a database crash, and it should now be fixed and the base is up again... Well, not quite :( 21 of my results are lost. They have been returned according to my log, but tomorrow the time limit is up, and the results are not listed. I haven´t checked, but does my computer save a copy of the results or are they deleted when sent? I am, of course, thinking about the posibility of making the computer resend the missing answers. Info anybody? Thanks and have a nice week-end ChrisD ID: 12552 · Reply Quote

Michael Karlinsky Send message Joined: 18 Sep 04 Posts: 163 Credit: 1,682,370 RAC: 0	Message 12553 - Posted: 28 Jan 2006, 18:25:26 UTC - in response to Message 12552. Last modified: 28 Jan 2006, 18:26:26 UTC On Jan 26 LHC had a database crash, and it should now be fixed and the base is up again... Well, not quite :( 21 of my results are lost. That's what a crash is all about :( They have been returned according to my log, but tomorrow the time limit is up, and the results are not listed. I haven´t checked, but does my computer save a copy of the results or are they deleted when sent? The results were uploaded and reported. So they are gone from your client. I am, of course, thinking about the posibility of making the computer resend the missing answers. If you are worried that any science might be lost: BOINC will resend the work when your result expires. If you are worried about your credits: You will probably not get any for them. Michael Team Linux Users Everywhere ID: 12553 · Reply Quote

River~~ Send message Joined: 13 Jul 05 Posts: 456 Credit: 75,142 RAC: 0	Message 12554 - Posted: 28 Jan 2006, 18:55:49 UTC - in response to Message 12553. If you are worried that any science might be lost: BOINC will resend the work when your result expires. If you are worried about your credits: You will probably not get any for them. If you are worried about your (or your team's) ranking in the stats, it affected everybody who sent in work during a ~12hr period, so the effect will be shared out across most participants, and rankings will not be affected as much as if it was just you that suffered. Almost all of us are affected, and it was nobody's fault. The notice saying it was fixed meant that results were once more being recorded properly, not that the lost results were recovered. I am sure Chrulle has had a look to see if any results can be recovered - but unless we see another announcement I think we should all assume that anything that is lost will stay lost. The best thing to say is that it doesn't happen often... ID: 12554 · Reply Quote

tengil Send message Joined: 27 Jul 05 Posts: 4 Credit: 440,478 RAC: 0	Message 12556 - Posted: 28 Jan 2006, 19:22:20 UTC I thought that CERN wasn't able to lose electricity. How did that happened? ID: 12556 · Reply Quote

Paul D. Buck Send message Joined: 2 Sep 04 Posts: 545 Credit: 148,912 RAC: 0	Message 12557 - Posted: 28 Jan 2006, 20:00:07 UTC - in response to Message 12556. I thought that CERN wasn't able to lose electricity. How did that happened? Forgot to pay the bill I suppose ... ID: 12557 · Reply Quote

Ananas Send message Joined: 17 Jul 05 Posts: 102 Credit: 542,016 RAC: 0	Message 12558 - Posted: 29 Jan 2006, 2:26:28 UTC SSL must have bundled their breaker with BOINC ID: 12558 · Reply Quote

River~~ Send message Joined: 13 Jul 05 Posts: 456 Credit: 75,142 RAC: 0	Message 12561 - Posted: 29 Jan 2006, 10:09:25 UTC - in response to Message 12556. I thought that CERN wasn't able to lose electricity. How did that happened? It is true that CERN has separate connections from both the French and Swiss national power grids - it would take both those grid circuits to go down for the big machines to lose power. Also, if it is the same as it was when I was there in the '70s, the domestic powor around the CERN site does not come of the grid power that supplies the big machines, but from the normal municipal supply like any other business in the canton. That way the power surges as they turn the big machines on and off do not make the lights flicker over the whole site... Even if that were not so, parts of the CERN site can go down just like a few streets can go down when the rest of the city is still powered. Or someone can trip the breakers for a single building... or part of a building. R~~ ID: 12561 · Reply Quote

Chrulle Send message Joined: 27 Jul 04 Posts: 182 Credit: 1,880 RAC: 0	Message 12578 - Posted: 30 Jan 2006, 10:44:47 UTC It was indeed only the IT building that lost its power. The problem was with the new local substation that is needed to supply the computer centre with power. Here is the explanation from the technical guys: The computer centre power failure on Tuesday night has been traced to a manufacturing fault with a circuit breaker in one of the main 2000A switchboards in the new substation. This problem was entirely unexpected---no equivalent failure has been seen at CERN in over 20 years. TS has asked the switchboard suppliers for an explanation and will be undertaking a systematic review of the remaining 94 circuit breakers in the near future. This exercise can be done without any service interruption but will require short periods when the physics computing farms run without UPS coverage. With many apologies for the inconvenience; we really hope that we are through our teething problems with the new substation. Chrulle Research Assistant & Ex-LHC@home developer Niels Bohr Institute ID: 12578 · Reply Quote

LHC@home