PING: BarryAZ
16 Jun 2007 0:17:54 UTC
Topic 192850
(moderation:
Since we have the same class date I took the liberty of pinging you here. I would have PM'ed you at SAH but they seem to have managed to break the web server over there at the moment. ;-)
Thanks again to all you folks who forced PAH to cut off posters to the fora with no RAC. Good move on your part regardless of how you felt personally about the incident, right or wrong.
Regarding the Checked, but No Consensus thread over on PAH. Please let David know that there is nothing wrong with the hosts or backend per se. It's a BOINC quirk that when the WU is in this state the host summaries show pending but don't appear on the Pendings summary. I read there is a fix checked in to address this, but I don't have a clue when it will become available.
Also let me assure you if everything passes validation eveybody gets credit for the WU. I've had it happen here at EAH and at SAH and I have always gotten credit for the result when the extra result comes in OK (all other things being equal).
Alinator
PING: BarryAZ
)
Yep, I thought I was connecting to LHC's web server instead of SETI's... ;-)
RE: Yep, I thought I was
)
Sigh....
I don't know why they wait until Friday afternoon to rollout backend changes.
That's just asking to be on the short end of the fuse of doom. ;-)
Alinator
RE: I don't know why they
)
I don't think it is so much BOINC code in general. The message is pretty specific, about a syntax error in countries.inc...
Parse error: syntax error, unexpected T_STRING, expecting ')' in /disks/setifiler1/home/boincadm/projects/sah/html/inc/countries.inc on line 152
I'm a bit surprised that nobody has corrected it, considering how long it's been hosed...
In my former job, I was disliked by management for being willing to take some "risks" (they were pretty much scared of doing anything without 5 miles of paperwork, especially after SOX came along), however one of the big no-nos was any kind of major change, unless absolutely essential, on weekends.
Agreed, I wonder if they were
)
Agreed, I wonder if they were trying to fix the issue of the webserver putting up seemingly random languages instead of your selected one on the site nav bar when viewing certain pages.
In any event whatever happened is proving to be somewhat stubborn to fix. ;-)
OTOH, the problems which seem like they should be trivial to fix are usually the ones which get you in the most trouble when you actually try to! :-)
Alinator
RE: Agreed, I wonder if
)
Hmmm.... I was wondering if anyone else had that happen. Strange stuff... Since I sometimes might have a...ummm...desire to impress...ummm...someone, with a Cyrillic character or two, I thought it might have been something I did, but when Hebrew started coming up, I didn't think it was something on my side... :)
RE: I don't know why they
)
This is an easy one to answer. It's cyclical.
1) Come to work on Monday, and everything is FUBAR.
2) Spend M-Th fixing problems
3) Finally have time to get to the work intended for Monday.
4) Implement changes.
5) Go home for the weekend.
6) GOTO #1
Reno, NV Team: SETI.USA
OK -- got your note -- thanks
)
OK -- got your note -- thanks for the explanation regarding Predictor -- like a number of others, I received a 'bulk credit' boost earlier this week -- it still looks like there is a bit of manual processing to clear periodic work unit credit issues there.
Sorry about the delay in responding -- I was mostly offline on vacation until earlier this week.
By the way, since it seems
)
By the way, since it seems you might have some added information sources with the various BOINC projects, do you know more about the 'unplanned maintenance' which has taken Climate offline for the past three days. Aside from the notice they posted on Wednesday, there has been no detail of what is going on there. According to that notice, they 'hoped' to be online Thursday or Friday -- looks now like it is an 'airline' day or two day outage -- that is, it is a day or two days from the time you look. (Like dealing with flight delays -- the flight will board in 20 minutes, you come back in 20 minutes and you are told nothing has changed, the flight will board in 20 minutes).
LOL... No, I haven't heard
)
LOL...
No, I haven't heard anything more about what going on with the CPDN database problem.
Sometimes I get lucky when researching a problem for someone here or over at SAH and run across a post in the BOINC mailists, etc. which gives some hint. So far, nada on that front. ;-)
BTW, it turned out there apparently isn't a RAC squelch function, I must have just drawn a 60 dayer for wading in on the other side of that thread. :-)
Call it a 'friendly fire' incident. :-D
OK -- Climate typically has
)
OK -- Climate typically has been reasonable about explaining outages -- especially when the extend past the first (typically optimistic) recovery date. For their current outage, 1 to 2 days is now working toward day 4 with no further status flash. I've done the typical response and suspended the two BBC Climate and all the Climate processing cycles I have going -- pushing the work to my other five active projects.
Regarding Predictor -- there seems to be a couple of things there -- first, they are not a large project, so there is a lot of support multi-tasked activity - which can sometimes undermine the quality of task support due to distraction. Second, the low staffing there means that when folks get rowdy in any way, there is nothing in the way of early control/mediation/response so that by the time the admins take notice things can be out of hand. There response in those situations tends to be hammer/bug/squish.
Looking at the stats there, my sense is that Predictor was a much more active project in the past (before I added it to my collection). A lot of folks have seemingly gone on to other projects -- that's the only way I could have a top 20 RAC there. It is only my number 5 (of 7) projects -- though it is getting a larger share with the Climate crash this week.