News on Project Outages

Message boards : Projects : News on Project Outages
Message board moderation

To post messages, you must log in.

Previous · 1 . . . 8 · 9 · 10 · 11

AuthorMessage
Profile Blurf

Send message
Joined: 18 Jul 11
Posts: 217
United States
Message 46587 - Posted: 3 Dec 2012, 20:57:40 UTC

This was not just the lab folks--it was a whole section of campuys. Doesn't matter if the lab has power when the existing area doesn't.
ID: 46587 · Report as offensive
Profile Gary Charpentier
Avatar

Send message
Joined: 23 Feb 08
Posts: 2291
United States
Message 46590 - Posted: 4 Dec 2012, 1:20:44 UTC - in response to Message 46587.  
Last modified: 4 Dec 2012, 1:20:57 UTC

This was not just the lab folks--it was a whole section of campuys. Doesn't matter if the lab has power when the existing area doesn't.

The entire campus was back up days ago. The issue is the DB was in a inconsistent state when power went down and had to be rebuilt. Remember the three day outages ...
ID: 46590 · Report as offensive
BarryAZ

Send message
Joined: 4 Sep 09
Posts: 381
United States
Message 46599 - Posted: 4 Dec 2012, 22:56:34 UTC - in response to Message 46590.  

They lost power for an hour -- last week. I really think they should 'go pessimistic' on estimates. I suspect the 12/4 estimate was optimistic (if things go right), instead of realistic (if things go as they have been for the past month), or pessimistic (if things go worse than they have for the past month).

That way, they might have indicated that while they *hope* to be up and running by 12/4, it is at least quite possible that it won't be until 12/11 (or something like that).

This would also have the benefit of pushing some users away -- which this project, for all the noise it generates about 'too little resource for too many users' would seem to be something the project folks would welcome.

Not a big deal for me, SETI long ago went into my low-middle set of BOINC projects, well below Einstein and MilkyWay for astrophysics projects, and for that matter below my cadre of ATI GPU projects (MilkyWay, MooWrap, Collatz, WorldGrid, POEM). These days even Rosetta gets a larger CPU share from my collection. I had been pushing SETI cycles up quite a bit until this past month when things started breaking down. At this point, a return will be quite tentative -- perhaps simply completing existing work and then taking another look next month.

Still curious that a 1 hour outage not only outlasted their UPS support (not a big surprise there) AND uncovered apparently broken power outage shutdown agents. One would have thought those would be in place - especially as power outages at Berkeley are not all that infrequent.



This was not just the lab folks--it was a whole section of campuys. Doesn't matter if the lab has power when the existing area doesn't.

The entire campus was back up days ago. The issue is the DB was in a inconsistent state when power went down and had to be rebuilt. Remember the three day outages ...

ID: 46599 · Report as offensive
BarryAZ

Send message
Joined: 4 Sep 09
Posts: 381
United States
Message 46601 - Posted: 5 Dec 2012, 2:11:46 UTC - in response to Message 46599.  

I see that the SETI folks this afternoon have determined that restoration of the databases is taking longer than expected (hoped for) and the new date they expect (hope for) is tomorrow, December 5. Sort of reminds me of airport time. Flight bording is delayed -- told that boarding will happen in 20 minutes. Come back to the podium in 20 minutes, and are told that boarding will happen in 20 minutes. Come back 20 minutes later, same story. Ask about the ongoing delays, reply -- we told that boarding will be in 20 minutes-- nothing has changed.

The restart process after a major database mess up is often quite time consuming. I still say that the better tactical approach would have been to lay out a pessimistic assessment (say December 11) and then come back on line on December 6 or 7. Instead, by suggesting December 4, then December 5, then perhaps December 6, the impression is that of optimism over realism and this project needs more of the latter and less of the former.
ID: 46601 · Report as offensive
Profile Gary Charpentier
Avatar

Send message
Joined: 23 Feb 08
Posts: 2291
United States
Message 46603 - Posted: 5 Dec 2012, 4:48:50 UTC - in response to Message 46601.  
Last modified: 5 Dec 2012, 4:49:05 UTC

the impression is that of optimism over realism

Why are you posting this here? Forum withdrawal? You could e-mail the project admin with your concerns. I don't believe he reads this forum.
ID: 46603 · Report as offensive
[CSF] Thomas H.V. Dupont

Send message
Joined: 30 May 12
Posts: 356
France
Message 46604 - Posted: 5 Dec 2012, 6:33:17 UTC - in response to Message 46601.  

Flight bording is delayed

Once you're on the plane, it's only an old story...
You should not be so picky about dates.
If they announce a return date so close, it may be that they also want a return to normal quickly.
Everyone wants the same thing.
Gary is right.
ID: 46604 · Report as offensive
Profile Blurf

Send message
Joined: 18 Jul 11
Posts: 217
United States
Message 46613 - Posted: 5 Dec 2012, 20:57:53 UTC

Seti forums are up. We seem to have progress :)
ID: 46613 · Report as offensive
Profile Blurf

Send message
Joined: 18 Jul 11
Posts: 217
United States
Message 46628 - Posted: 6 Dec 2012, 21:49:30 UTC
Last modified: 6 Dec 2012, 21:49:43 UTC

Validator is down on Milkyway. Issue has been escalated
ID: 46628 · Report as offensive
Previous · 1 . . . 8 · 9 · 10 · 11

Message boards : Projects : News on Project Outages

Copyright © 2021 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.