old NEWS!!!
August 4, 2004
Check out the new Planetary Society newsletter: Multi-Beam Receiver Promises New Vistas for SETI Research.
(also to be found here as a direct link)
August 2, 2004
This morning the entire Space Sciences Lab lost its network connectivity to the rest of the world for a few hours. This got fixed down on campus, but the SETI@home servers are still currently catching up, so you may have some difficulty connecting.
July 21, 2004
Another month has passed and I've been busy except with the PEG-SETI pages. This is way past apologizing :( It is plain laziness. I'm sorry.
Well, BOINC has gone public now and is on Version 3.20. (I did send an email about it to the PEG though)
Even though I've been testing BOINC and following its development closely, I haven't gotten around to create a statistics page like we have in the old SETI@home that we know. BOINC statistics are supposed to be available in XML, but right now they are exported files, chopped up in slices and bulky for my small system. They are supposed to come up with a CGI or PHP script which lets me query a team and the users. So eventually the statistics (graphs) page will be updated with SETI@home2 (BOINC).
From the Berkeley website:
June 22, 2004
A new version of SETI@home based on BOINC, is now available. We'll be transitioning to this new version (details are here). For now, you can use the new version or stick with SETI@home Classic.
June 22, 2004
First of all, my apologies. I haven't been updating the website lately.
Maybe some of you noticed that the results had some trouble to be returned as well as new work units were hard to get. The following newsflashes are from the SETI@Home website:June 20, 2004
We expect the faulty router to be repaired tomorrow. Normal data service will follow.June 18, 2004
The campus network folks did some great troubleshooting and narrowed down the problem to a faulty link on the path to our ISP. Repairs have been called in.June 17, 2004
We are working on fixing the network problems that are resulting in dropped connections to the data server.In the meantime the problem has been resolved and traffic flows again normally.
May 18, 2004
SETI@home turned five years old yesterday. Congratulations and thanks to all concerned. Our search continues!
May 12, 2004
OUTAGE NOTICE: On Monday, May 17th we are going to have a two-hour web site outage starting at 17:00 UTC. Client downloads, statistical pages, etc. will be off-line during this time.
April 21, 2004
ZoneAlarm Pro users may be experiencing weird behavior when looking up user stats on our web site. Click here for more info.
March 25, 2004
A new Mac GUI client, version 3.08a, is now available. This version fixes screensaver problems under Mac OS 10.3.x.
March 24, 2004
New Planetary Society newsletter: In Search of Dyson Spheres
March 10, 2004
OUTAGE NOTICE. There will be maintenance performed on the data server's link to the Internet tomorrow, 3/11/2004, from 11:00 to 15:00 UTC. The outage is not likely to cover the entire 4 hour period.
February 13, 2004
OUTAGE NOTICE. There will be a data server outage at some point between the hours of 08:00 and 12:00 UTC this coming Sunday, 2/15/2004. During this period our ISP, Cogent Communications, will be reconfiguring their core routers. It is unlikely that the outage will last the entire 4 hour period.
January 24, 2004
(PEG - SETI team)
After a two week outage on my system due to a not so pretty sight:
Windows protection error. You need to restart your computer.
System haltedand not so good support from HP, the company I got this box from,
I finally managed to get it back myself and feel confident that it is stable.I did lose a few days of statistics which I now tried to "cheat in" manually.
Robi out
December 4, 2003
Here is a brief summary and current status of our reobservations.
December 3, 2003
The data server went down last night for about 12 hours. It is up again now, but slowly catching up on backlogged connections.
November 20, 2003
BOINC beta test has opened the gates for more beta testers.
To join, visit http://setiboinc.ssl.berkeley.edu/ap/ and create an account.
Make sure to have read the Rules and policies [read this first] link on that page before joining.
The PEG-SETI team for BOINC is located at http://setiboinc.ssl.berkeley.edu/ap/team_display.php?teamid=13
November 15, 2003 (updated)
The SETI@home website http://setiathome.ssl.berkeley.edu/ seems to be down (it is unreachable).
According to the cricket graphs, from about 18:40 to 21:50 (3 hours 10 minutes) on November 15th, there was an outage in the communication to the SSL network.
Aha! A check in the alt.sci.seti/sci.astro.seti newsgroup shows the following post: news:Pine.GSO.4.58.0311152159170.25601@rancid.berkeley.edu or from Google
Quote:
«I just got it back--a media converter power supply went»
bad. I had to improvise to get it working again (fortunately these
"proprietary" media converters are actually standards-based and I was able
to swap in another one).
November 15, 2003 (updated)
A virus/worm appears to be currently installing SETI@home on computers:
November 14, 2003
It is possible that one or more viruses/worms are circulating around the internet installing SETI@home on infected computers. For more information, click here.
October 30, 2003
If you are unable to connect to our server because of "100" errors, please click here.
(or if your SETI client just doesn't want to return a finished result)
(note from Robi: I had this problem, but instead of deleting the "result.sah" and the "user_info.sah" file I searched for a valid "user_info.sah" file on my computer and copied the values of
- id=[id-number]
- key=[key-number]
into the "result.sah" file that didn't want to be returned using the wordpad - yeah, windows :(
as
- user_id=[id-number] (this one was 0)
- user_key=[key-number] (this one was either -1 or -0)
I also copied the correct "user_info.sah" over the bad "user_info.sah" file and then returning the results completed successfully )
October 23, 2003
Our data server is down because of a disk problem on the user database host. We are working on a fix.
October 7, 2003
(BOINC Beta-test mailing list)
We may have fixed the problem that resulted in lots of results being issued after their input files had already been deleted (causing download failures).
Other project news:
- we're very close (and I mean it!) to releasing the Windows client for SETI@home/BOINC.[1]
- Some details of the planned transition to BOINC are described at http://setiathome.berkeley.edu/boinc_transition_plan.html Please let me know if any comments/questions.
- We have someone working on a BOINC GUI client for Mac OS X. It will look almost exactly like the Win GUI.
-- David
[1] with "releasing" he probably means released for Beta testing
October 2, 2003
We are in the process of fixing/upgrading our webservers. Some web functions may be broken. We apologize for any inconvenience.
(note from Robi: I tried to update all "broken" links
[cgi ⇒ fcgi]. If something isn't working, please let me know at the address at the bottom of the page)
September 26, 2003
New article just released at the Planetary Society: New and Improved SETI@home will form Backbone of Distributed Computing Network.
September 10, 2003
Recent web server glitches (slow response times, broken message boards, etc.) have been fixed.
September 8, 2003
(PEG - SETI team)
My Computer/Server, a Windows® system, had been stalling and freezing for some time and 3 days ago I wasn't able to start it up anymore, but with persistence I managed to get it back up. Some people might have tried to access my local pages and some noticed that the Geocities and/or Tripod pages had not been updated. I am hoping to get this system to become more stable (Linux maybe). Anyway, the system is back up, the pages are back online and the PEG - SETI team page is being updated again.
Robi
September 5, 2003
Outage Notice. This Sunday, 9/7/2003, network maintenance will be performed at Space Sciences Lab. This web site may be unavailable for short periods throughout the day. Our data server will probably not be affected.
August 29, 2003
(PEG - SETI team)
For several days the last 4 tables in the graphs page had an error which I didn't notice until today.
When José María left, there was no member in Spain left, which caused an unfortunate div/0 in my Progress 4GL calculations. I have localized and fixed the cause, and the page is generated again correctly.
August 28, 2003
New SETI@home Science Newsletter released today: Telescope Pointing Corrections.
August 27, 2003
We made an error in the generation of reobservation workunits. We have corrected the error and are redistributing these workunits. Here are some details.
August 22, 2003
One of the four CPUs on the main data server malfunctioned this morning, causing several crashes. We have pulled the bad CPU out of the system. A replacement from Sun should be arriving later today. We will need to have another outage to install the new CPU.
August 21, 2003
We made an error in the generation of reobservaton workunits. We have corrected the error and are redistributing these workunits. Here are some details.
August 19, 2003
As of Monday, August 19, 2003, SETI@home has received over one billion results (one milliard) from SETI@home participants world-wide. See our current total statistics page for daily accumulated results.
July 21, 2003
(SETI@Home / The Planetary Society)
Read about how the analysis of the reobservations data is coming along in The Planetary Society's latest SETI@home update! (Analyzing the Reobservations)
Jun 19, 2003
Website Outage Notice. This Sunday, 6/22/2003, there will be a 2 hour website outage starting at 16:00 UTC. Router hardware changes will be taking place.
Jun 16, 2003
SETI@home has just finished sending out workunits for reobserved candidates.
May 23, 2003
(SETI@Home / The Planetary Society)
(May 21, 2003)
SETI@home Prepares to Send Out "Reobservations" Work Units!
May 11, 2003
See lots of photos from our reobservation trip to Arecibo.
April 22, 2003
(BOINC)
BOINC beta testing has been resumed
Quoting a message from the BOINC beta email list:
"The legal situation has somewhat progressed. We can now resume the beta test and start testing and bug fixing again, however we have to keep the SourceForge site down (for now). The beta server is still working, so feel free to start running BOINC again and posting bugs to this mailing list."
April 21, 2003
We have a new signal candidates page, providing links to news, articles, and other candidate-related information.
April 16, 2003
Browse our new reobserved candidates, organized by the SETI@home participants who helped identify them.
April 16, 2003
SETI Cam from the SETI Institute is back online at Arecibo, featuring Project Phoenix observations during April 14 - 30 (2 ½ weeks).
April 9, 2003
Due to a legal issue, we are suspending the BOINC beta test, and we have requested that BOINC source code no longer be available at SourceForge.net. We hope to resolve this issue soon.
April 7, 2003
The start of today's 2 hour data server outage has been moved to 1800 UT.
April 7, 2003
Our user database crashed this past weekend, causing trouble with user stats updates. All stats will be properly regenerated sometime today.
April 7, 2003
I have found the CLI version 3.08 – which fixes a buffer overflow error – for the following systems:
- Windows - setiathome-3.08.i386-winnt-cmdline.exe
- Solaris - setiathome-3.08.sparc-sun-solaris2.6.tar
- MAC - setiathome_mac_3_08.hqx
- Linux - setiathome-3.08.i686-pc-linux-gnu.tar
They can be found at ftp://alien.ssl.berkeley.edu/pub/
I wasn't able to download the windows client directly from http://setiathome.berkeley.edu/unix.html but going to the FTP directory helped.
April 6, 2003
(March 31, 2003)
We are preparing a BOINC-based version of SETI@home. See a preview of the graphics.
April 4, 2003
There is a software update with a precautionary security fix. To obtain it, go the the download page.
The new version of the GUI client is 3.08, which fixes a buffer overflow error that could be a potential security threat.
"Version 3.08 is a precautionary security release. There was a potential buffer overrun in the networking code of the client that is fixed with version 3.08. Note that to exploit this vulnerability, a potential attacker would have to trick the client into contacting a fake server rather than the actual SETI@home server. To our knowledge, no SETI@home client has ever been attacked in this manner." [Link]
April 4, 2003
OUTAGE NOTICES. There will be intermittent service drops for one hour starting tomorrow at 0800 UT while maintenance is performed on our Internet link.
On Monday 4/7 there will be a 2 hour data server outage starting at 1700 UT while Sun performs some maintenance.
March 29, 2003
"Within the next two years the SETI@home team hopes to phase out the aging receiver at the base of the line feed. In its place, SETI@home observations will be conducted using a new multi-beam array that will be located within the Gregorian dome.
Unlike the single needle-shaped feed now in use, the array will be composed of seven separate feeds, each connected to its own highly sensitive L-Band receiver. This means that whereas the line feed can only observe a single location in the sky at any given time, the multi-beam array will be able to point at seven locations simultaneously. It will, furthermore, observe them at a much higher degree of sensitivity than is possible with the current arrangement."
March 29, 2003
(SETI@Home Tech news March 27, 2003)
We had another unexpected data server outage last night due to residual damage from the user database crash a couple days ago. The damaged table in question was expendible - it just had to be dropped and recreated and then everything worked again.
(see tech news dated March 25, 2003 for the preceding "problem")
March 11, 2003 (updated March 26 with even more links)
New information about what is happening (and will happen) with the results we have analyzed:
http://www.berkeley.edu/news/media/releases/2003/03/10_search.shtml
http://www.space.com/scienceastronomy/setiathome_030310.html
http://www.msnbc.com/news/883316.asp
http://www.bayarea.com/mld/mercurynews/news/local/5360300.htm
http://www.wired.com/news/technology/0,1282,57992,00.html (2 pages)
http://www.oaklandtribune.com/Stories/0,1413,82~1726~1235808,00.html
http://www.theage.com.au/articles/2003/03/11/1047144976260.html
http://www.newscientist.com/news/news.jsp?id=ns99993486
Also some news at The Planetary Society:
SETI@home is preparing to reobserve some of its most promising candidates on March 18-20. See SETI@home Returns to the Scenes of its Most Promising Candidates at the Planetary Society for more details.
- What to do when ET Calls: The SETI Protocol
- March 10, 2003: Shifting Gears at Arecibo
- March 12, 2003: Reobserving, Recording, Reprocessing
- March 14, 2003: Selecting the Finalist Candidates
- March 17, 2003: Real Time Results
- March 18, 2003: First Observation Session Completed!
- March 19, 2003: Solar Intervention Delays Reobservations
- March 24, 2003: On Last Day at Arecibo SETI@home Turns to Distant Planetary System
- March 25, 2003: Reobservations End at Arecibo
http://planetary.org/audio/pr20030317.html
http://planetary.org/html/society/press/2003/stellar_countdown.html
March 20, 2003
OUTAGE NOTICE. There will be 14 hour outage beginning at 02:00 UT on 3/24/2003. This outage is for major electrical work at the Space Sciences Lab.
March 20, 2003
http://www.scifitoday.com/story/2003/3/19/7147/26354
However, the final two observation sessions had to be postponed due to the eruption of a solar flare. For the next few days, the Arecibo radio telescope will be used to track this rare event on the Sun. "It looks like we and everybody else using the telescope from now though the weekend is getting bumped," Werthimer said. "We can still observe a bit for tomorrow (March 20th), but only two hours. We are getting bumped because of a solar flare. They need to reschedule us for early next week, so we probably won't get the observing finished until Tuesday next week. This is rare--happens once every two years that they have to bump everyone so they can observe a flare. The schedule is still tentative."
March 13, 2003
See our skymap of best candidates to be reobserved on March 18-20.
March 11, 2003 (updated March 20 with more links)
New information about what is happening (and will happen) with the results we have analyzed:
http://www.berkeley.edu/news/media/releases/2003/03/10_search.shtml
http://www.space.com/scienceastronomy/setiathome_030310.html
http://www.msnbc.com/news/883316.asp
http://www.bayarea.com/mld/mercurynews/news/local/5360300.htm
http://www.wired.com/news/technology/0,1282,57992,00.html (2 pages)
http://www.oaklandtribune.com/Stories/0,1413,82~1726~1235808,00.html
http://www.theage.com.au/articles/2003/03/11/1047144976260.html
http://www.newscientist.com/news/news.jsp?id=ns99993486
Also some news at The Planetary Society:
SETI@home is preparing to reobserve some of its most promising candidates on March 18-20. See SETI@home Returns to the Scenes of its Most Promising Candidates at the Planetary Society for more details.
- http://planetary.org/stellarcountdown/reobservations_1.html
- http://planetary.org/stellarcountdown/reobservations_2.html
- http://planetary.org/stellarcountdown/reobservations_3.html
- http://planetary.org/stellarcountdown/reobservations_4.html
- http://planetary.org/stellarcountdown/reobservations_5.html
- http://planetary.org/stellarcountdown/reobservations_6.html
http://planetary.org/audio/pr20030317.html
http://planetary.org/html/society/press/2003/stellar_countdown.html
March 11, 2003
(PEG - SETI team)
I have been selected for Beta-Testing of the new BOINC (Berkeley Open Infrastructure for Network Computing) client and have set up some information about what will come:
January 9, 2003
OUTAGE NOTICE . Tomorrow there will be intermittent disruptions to the availability of our data server because of emergency maintenance on our Cogent link. The disruptions will occur between 11:30 and 12:30 UTC.
February 9, 2003
Yesterday there was an unscheduled outage that started at 17:30 UT. During normal construction work, an undocumented section of fiber was disrupted. This was the the fiber that carried network traffic from Space Sciences Lab (including SETI@home) down to campus. The campus network engineers worked tirelessly to fix the problem by rerouting traffic over intact strands of fiber and service was restored at 04:00 UT today.
February 5, 2003
OUTAGE NOTICE. We will be having a 2 hour data server outage tomorrow, begining at 18:00 UT . During this outage we will be replacing a CPU that failed this past weekend as well as an ethernet interface that is beginning to fail. We'll take this opportunity to perform some system software upgrades as well.
February 3, 2003 - Correction!!! (February 1, 2003)
News from BOINC and S@H
Interesting NG post from a
GermanTLC (Team Lamb Chop) member, Mike Kaehlerwho visited Berkeley and interviewed David Anderson (SETI@Home and BOINC project director) and Eric Heien (BOINC developer/student) about the BOINC (Berkeley Open Infrastructure for Network Computing) Project.Sorry Folks, BIG mistake on my side! When I saw the post in the newsgroup alt.sci.seti, I assumed it was from Mike. This isn't so. He had posted this message from the ars technica [TLC] Open discussion forum, written by Sungod - Team Ars Technica Lamb Chop.
The reading is still interesting, either on the NG or on the TLC Open Forum.
Links to
NGmessage:
- Original message
- NG post <news:f65ae017.0302010607.6304f3ce@posting.google.com>
- groups.google.com (link to thread)
Sorry about the mistake. Robi
February 2, 2003
(PEG - SETI team)
PEG - SETI team has passed the total CPU time of 100 years.
On February 2nd, 2003 20:00, the team's CPU time is 101.942 years.
Scanning back through the history data, on January the 26th we reached together the 100 Year CPU time. Actually thebithead joined us and catapulted us right into the 100's.Congratulations to everybody and thanks to thebithead for the extra boost ;^)
February 1, 2003
The bandwidth problem that we have been having for the past day or so was due to a faulty ethernet interface on our data server. We swapped to a new interface today and bandwidth availability returned to normal.
January 31, 2003
We've been wrestling with some data server problems the past 24 hours. We're still working on it.
January 9, 2003
(PEG - SETI team)
Ok, it seems that the Dutch Power-play is diminishing, which has catapulted Robin since it's start back in mid-September from the 31st place to 4th place, having returned an amazing 41/4K results in those 115 days.
Congratulations Robin. Actually, congratulations to everyone!
Besides that, I have found 2 other teams with progress people, which — compared to our team — are pretty small:
PROGRESS-Hackers (mainly a German company using Progress)
9 Members
9'292 Results
Progress (SETI enthusiasts at Progress Software Corporation)
29 Members
11'674 Results
Progress Email Group (placed here for comparison)
46 Members
61'668 Results
Update Jan 12, 2003. Hey, if someone from those teams happens to read this, why don't all y'all join us?
To be in the top 200 of all teams, we would need about 230'000 results (that's about 145'000 more than the the three teams above combined), but to be in the top 200 of the teams of our class "Clubs", where the top 200 mark is at around 132'000, combined we'd need just 50'000 more.
So if anybody knows another large SETI group related to Progress, or some other people who want to join the team for the fun of it, let them know.
December 18, 2002
OUTAGE NOTICE. On Thursday, 12/19/2002, we will be having a one hour data server outage for user database maintenance. The outage will begin at 18:00 UT.
Note from Robi: There has been some "cheating" going on, with users returning 1500+ results per day, and Berkeley has
a) removed cheated results in user statistics
b) removed whole cheating accounts.
I suppose the user DB maintenance has to do with that and the crashes on December 13th
December 15, 2002
(PEG - SETI team)
The team graphs take a little longer to load (they are 4 to 5 times larger in KB than they were before). I noticed that the graphs were blurry and unreadable, so I had to remove some JPG compression :( Sorry about that. At least the lines are better visible and the text readable.
On a side note, the US is returning close to 100 results per day after the Dutch call for help ;o)
Robin, sorry it backfired. Your computer power is still amazing.
- Our friends at Berkeley had 2 system crashes on Friday the 13. Must have been because of the date. I had trouble on the 14th though. They had posted the crashes on the SETI@home website, but seems that they removed it again.
December 1, 2002
(PEG - SETI team)
Ah, the USA has claimed it's top rank in Results Per Day back. Robin is still returning about 60 to 65 results per day, but that "power play" has been overpowered by the US teamwork. Still, nice going Robin.
BTW, in our PEG - SETI team, we have the following participating countries:
North America:
Canada
Europe: United Kingdom USA
Netherlands Mexico Spain Oceania: Australia Czech Republic New Zealand Germany
November 15, 2002
Outage Alert. This Sunday, 11/17/02, from 17:00 to 19:00 UT, there will be a series of short web site outages for lab wide network maintenance.
November 15, 2002
(PEG - SETI team)
In the last 2 weeks we had a major "breakthrough" — literally speaking. Congrats to that achievement goes to Robin Doesburg who, together with a friend of his, has managed to return a whopping 71 results in the last 24 hours, averaging 41 results per day for this week. Robin managed to catapult the Nederlands to First Place on the Country Statistics in the Results per Day table with an average of 55 results for NL, knocking the USA to Second Place with 53 results per day.
Steve West-Fisher has managed to pass Danny Williams, our long time Numero Uno who had kept the Top Rank since the PEG - SETI team was created. Congratulations to both, Danny and Steve, our new Number One — Engage!
Milestones:
Name Results Date Lynn McKelvie
100
Oct. 30
Robin Doesburg 250 Nov. 3 Robin Doesburg 500 Nov. 11 Robin Doesburg 750 Nov. 15 Jim Litzie 750 Oct. 28 Nigel Allen 1000 Nov. 15 Personal Milestone Certificates can be viewed/printed here
November 9, 2002
A paper about SETI@home is featured in the Nov. 2002 issue of Communications of the ACM.
September 10, 2002
We've added more milestone levels, all the way up to 100,000.
(see also the 'Certificates' form on this page)
September 2, 2002
(PEG - SETI team)
Well, so much for an unintended temporary (4 days) interruption of the PEG - SETI team homepage "housekeeping") :(
What happened:
Pat and I bought a new house and had planned to move somewhere around September 10th - 15th, but we got the house two weeks earlier to move in, and since we were so eager to move I had completely forgot to ask someone in the team to keep the statistics while wewaswere moving (like last November when we went to FL for a week). There might me some minor glitches on the statistics page during the next month. Sorry about that and thanks for bearing with me ;o)Robi
August 25, 2002
Read about our exciting new future projects.
Note from Robi: SETI-2 is coming soon. This means that when that time arrives, we will have to get a new analyzing client (called BOINC - yes, some of you are laughing hysterically now ;o)
I don't know how "comparable" the new client will be towards the old one, since some of you use the client mainly to test the speed and performance of new systems (burn-in test) but it has several new features, which until now we had as add-ons only, like caching work units.Robi
August 19, 2002
(PEG - SETI team)
I have added a Country Standings table to the graphs page (there was a "little" error on the calculations, but I think I sorted that out :)
August 15, 2002
Outage Alert: There will be another quick one hour outage on Friday, August 16, starting at 17:00 UTC. The data server will be unreachable during this time.
August 8, 2002
Outage alert: We are scheduling another one-hour outage on Friday, August 9th, starting at 17:00 UTC. This will be for further user database maintenance, so the data server and web CGIs will be off during this time.
July 26, 2002
(PEG - SETI team)
Finally! 'THE MAP' (How is the Team scattered) has been updated (and enhanced ;o)
Please let me know if something is wrong
July 16, 2002
We discovered that since rebooting the data server yesterday, the netmask on the Cogent facing interface was wrong. After correcting this, we began to see reports that the very recent connectivity problem may have gone away. We are a still concerned that some folks reported that the problem started a day or so before the reboot. We'll keep out eye on the message board.
July 16, 2002
We discovered that since rebooting the data server yesterday, the netmask on the Cogent facing interface was wrong. After correcting this, we began to see reports that the very recent connectivity problem may have gone away. We are a still concerned that some folks reported that the problem started a day or so before the reboot. We'll keep out eye on the message board.
July 16, 2002
We completely moved our main web server over to a new machine, over the course of the past week. You may have noticed a few brief glitches when it was difficult to connect. During the process, we also began enforcing that people use the correct user lookup CGI request, which unfortunately broke some third-party add-ons, but those seem to be getting fixed by their respective authors. Outside of all that, the move should have been relatively transparent.
The upshot of this is the web server is now on a more powerful, dedicated machine. We will probably add a second machine as well. Expect to see lots of new web features in the near future, now that we have increased our web serving capacity.
July 15, 2002
Changing an obscure configuration option from the default on our firewall has, from all reports, fixed the partial download problem. Thanks go to Ken at UCB's Communication and Network Services for helping us track this down and to all of the SETI@home participants that posted messages about the problem. And special thanks go to David H. Lipman for his tireless field testing over the past several weeks!
July 13, 2002
(PEG - SETI team)
I am updating the Map of our team members, although it is time consuming, and I don't want to just place "x" flags in the center of the US, AU, NZ and so on. I want to place the flags more or less accurately on the cities/regions where our members reside. I do have several emails from members who have told me the city they live in, so if you think I don't know where you're at and want me to place a flag over your town/region, please let me know.
I am also looking for an alternative map, a generic map or something like that instead of the one I'm using now.
July 12, 2002
OUTAGE NOTICE.
On 7/16/2002 (Tuesday, July 16.) there will be some maintenance work performed along our network path to the PAIX, where Cogent picks us up. This work will be done between 10:00 and 14:00 UT. Within this window there will likely be a 30 to 90 minute data server connectivity outage.
July 11, 2002
We're reconfiguring our webservers over the next week; we apologize for any occasional delays you may experience.
Note from Robi: Since the web servers at Berkeley are being worked on, there have been several reconfigurations, specially the CGI scripts have been moved to
fcgi-bin/fcgi, but the original CGI provides a link to the new target. I do hope that they will have the CGI scripts back to their original directoryUpdate: I have got the information that they are moving certain CGI scripts to fast-CGI:
(source hereunfortunately the message doesn't seem to exist on the messageboard anymore)
The user stats cgis have been moved to fastCGI.
The old URLs (using cgi instead of fastcgi) now load a page with the correct link.
The problem, as many of you know, is that we've had webserver load issues intermittently in the past. Our website generates over 1,500,000 hits per day (which we greatly appreciate and wish to maintain, don't get us wrong! :-) ), and there are a number of database access functions that comprise the majority of the load.
Server load is THE reason why we haven't yet been able to re-activate our "view your last 10 results" feature (though we're working toward our goal of reactivation).
FastCGI helps decrease the load tremendously, but unfortunately many users (apparently all users using SETISpy, which grieves me greatly, since your message states that SETISpy is no longer being maintained) still use the old non-fast CGI links. (Also, by the way, we've noticed that a number of users and team founders try to retrieve stats from our site MULTIPLE TIMES PER SECOND, which seems unnecessary & hurts our load. We'd love it if people could hit at a reasonable rate of once per 15 minutes or so.....)
In any case, I'll mention to the powers-that-be that SETISpy is foreveryoked to non-fastCGI. Unfortunate news!!
Thanks for alerting us,
SAH Administration
July 11, 2002
The results of the MSS change were inconclusive but tended towards the negative with some reports of a general slowdown. So we reversed this change. We are trying other changes with Netscreen's advice, so please let us know via the Help Desk message board if the problem goes away.
July 10, 2002
Some progress is being made in the "partial download" problem. See technical news.
June 15, 2002
We are still working on the partial workunit download problem. We will put updates in technical news as we see progress.
June 12, 2002
OUTAGE NOTICE. Tomorrow, 6/13/2002, there will be a 15 minute network outage, starting at 08:00 GMT, to upgrade equipment on our Cogent link. The data server will be unavailable during this time.
June 4, 2002
There are reports of clients hanging during workunit download. We are looking into this. If you are experiencing this problem, giving us feedback via this message board thread (messageboard was messed up, and thread doesn't seem to be available anymore) would be very helpful. Thanks!
June 3, 2002
Over in technical news is a description of a major reorganization of the online science database.
June 1, 2002
(PEG - SETI team)
As promised: Pictures! Link is at "Other links"
May 31, 2002
Network Bandwidth Problems Solved
Yesterday at about 23:30 GMT, we switched our data server's route to the internet from the general UC Berkeley link to Cogent Communications with the hope of solving our long standing bandwidth problem. See the Planetary Society's article for a great description of the problem and the Cogent solution.
It is now twenty-two hours after the switch and things look very stable. We are sending on the order of 25 Mb/s outbound through the new link and should have plenty of room to grow. We are no longer dropping connections and the server is operating very efficiently. This ought to translate into successful and speedy data downloads for all of our users.
Many, many thanks go to the networking folks who put this together. The entire route from Space Sciences Lab to Cogent's Palo Alto point of presence was designed and implemented by UC Berkeley's Communication and Network Services (CNS). In particular, we want to thank Siegrid Rickenbach who was the CNS technical lead for this project and got the path between campus and Cogent up and running. Jay Bryon of CNS configured and installed the equipment that was used to trunk both the Cogent and regular campus traffic over the fiber-optic link between campus and SSL. He had to come in very early to schedule outages so that a minimum number of users would be affected. Our own Space Sciences Lab network manager, Greg Paschall, did all of the local work needed for our server to get to the new link. Kudos also go to CNS gurus Michael Sinatra and Ken Lindahl for their routing expertise and critical moment wizardry. Because we had to move off the the SSL firewall, we needed a firewall of our own. Netscreen stepped up with a firewall donation and great support besides.
Thanks also to Ken Reneris who wrote the SetiQueue software package, and to the countless volunteers who used SetiQueue to set up workunit caches, allowing others to get workunits when our server was unreachable. And, of course, thanks to all our patient users who stuck with the project despite the frequent lack of connectivity!
All big changes have unforeseen problems and yesterday was no exception. An early morning test revealed a potential show stopper (or at least a show postponer). A test client from offsite, hardwired to come in over the new route, was not able to establish a connection. After thinking a bit at the white board, Michael at CNS figured it out. Incoming connection initiation was happening over the Cogent link. But the outgoing response was happening over our server's default route which was still through campus. This route passed through the SSL firewall which noticed a response to a request that it never saw (the SSL firewall does not see the Cogent link) and denied passage. Well, no problem we thought. We had to change our server's default route at the point of cutover anyway, which would send the response back out over the Cogent link. But not so fast. Our clients utilize the Domain Name Service (DNS) to find our server. DNS still had the old IP address (and by extension the old route). Thus, incoming packets would come in over the campus link and outgoing packets would go out over the Cogent link. The firewall would still see these "half open connections" and deny them. As long as there was asymmetry between the inbound and outbound traffic, we would have a problem. Now, we needed to tell DNS about our new IP address anyway and once this change propagated throughout the Internet we would be OK. But this propagation would take some time, during which nobody would be able to connect. CNS came to the rescue by telling the campus routers to redirect all incoming SETI@home packets to the Cogent link. This change was made at the same moment we changed our default route to send all outgoing packets over the Cogent Link. Symmetry was achieved! The link came up and performed very well. Watching the traffic graphs was an exciting and dramatic moment. We had a speaker phone connection with CNS during the cutover and hoorays could be heard on both ends of the line.
You can see the switchover on both sides: watch the Space Lab bandwidth plummet or the Cogent bandwidth jump.
May 30, 2002
Newsflash: We just linked our data server up to the new 100 Megabit/sec Cogent link (as of about 23:30pm GMT). We will be monitoring its progress throughout the evening. We'll let you know how it goes!
May 21, 2002
A new SETI@home article at the Planetary Society: A Question of Bandwidth.
May 20, 2002
OUTAGE UPDATE: There will be several network outages on Wednesday, May 22nd. The first is at 12:30 GMT, and may last upward to two hours. The second is at 15:30 GMT, and may last up to an hour. There will be a final short outage later in the afternoon. We are upgrading our network connection and reorganizing the server closet during these outages, and both the data server and the web site will be unavailable at these times. We'll take pictures of the whole procedure.
(I'm curious on the systems they have for SETI@Home. As soon as the pictures are made available at Berkeley, I'll link them under Other links)
May 19, 2002
(PEG - SETI team)
I added the homepages which show up on the team-stats page from SETI@Home to the two tables on the graphs page. It is not perfect, since the scrolling JavaScript doesn't change the background of the link and I must admit, I have no idea how to do it (IOW input would be gratefully accepted). I also know that the JavaScript is slow on the rows, but works on the 3 most used browsers (IE, Netscape and Opera). Both features are a test which, if I do not receive any negative input, will probably stay.
Ok, I haven't tested the links on those browsers yet - so it might be flaky - but will do ASAP (and fix it).If you have any comments or questions, let me know.
Also if you didn't want your homepage linked to your name, drop me a message and I'll fix that right away (I supposed it was ok since it's on the team-stats - sorry).
May 17, 2002
Today, May 17, 2002, is SETI@home's official three-year anniversary. All the credit for SETI@home's success and endurance belongs to our millions of dedicated users. We thank you and look forward to sharing our future with you.
May 16, 2002
OUTAGE UPDATE: There will be another network outage on Wednesday, May 22nd, at 12:30 GMT, and may last upward to two hours. Both the data server and the web site will be unavailable during this outage. These outages pertain to upgrading our network bandwidth.
May 16, 2002
Quote from NewsGroups alt.sci.setI and sci.astro.seti
[Only relevant part quoted - for original message, follow link above]
From: Michael Sinatra
Newsgroups: sci.astro.seti,alt.sci.seti
... I also stated that there might be a need for a second outage--and there will be.
So here are the details:
1. There will be an outage NEXT WEDNESDAY, 22 May 2002. The outage window
is 5:30 - 7:30 PDT (1230-1430 UTC). I doubt that the outage will last
anywhere near 2 hours, but we need to have a long window because of some
particularly cranky equipment we need to use (due to fiber distances to
the SSL network).
2. No changes to the bandwidth situation should be expected until at least
2-3 *days* AFTER the outage on Wednesday.
May 11, 2002
Quote from NewsGroup alt.sci.seti
[Only relevant part quoted - for original message, follow link above]
What we did today, combined with next Thursday's outage, will help the BW
problem--I hope. It will not stop server outages. It will not remove
single points of failure. But there's reason to hope that it will greatly
alleviate the bandwidth problem.
michael
May 10, 2002
(redacted May 11, 2002) Quote from NewsGroups sci.astro.seti and alt.sci.seti
[Only relevant part quoted - for original message, follow link above]
This has already been posted in alt.sci.seti, but I thought I would make
it 'official.'
There will be an interruption to network connectivity between the UC
Berkeley campus and the SSL[1]/LHS[2] network that carries seti@home. This
interruption will begin at 0600 PDT (1300 UTC) on 10 May 2002. The outage
will last anywhere from 15 minutes to 1 hour, depending on the flakiness
of a particular fiber transceiver.
This is the first of (more than 2 actually) outages, and this particular
interruption is to re-home the SSL/LHS nets (with seti) onto a new router.
Subsequent outages will separate the seti traffic from the SSL/LHS nets.
Michael Sinatra
Network Services
UC BerkeleyNote[n]:
May 9, 2002
OUTAGE ALERT. There will be 2 network outages in the next week for hardware upgrades. Each one will start at 13:00 GMT and should not last more than 1 hour. The first one is tomorrow, May 10, and the second one will be Thursday, May 16. Both the data server and the web site will be unavailable during these outages.
Editors note: fill up your cache, we programmers and IT guys know how long these hardware upgrades last... <g> especially if "outages in the next week" and "The first one is tomorrow"...
May 7, 2002
The 500 millionth result has been returned on May 2, sometime around 10:19 UTC, and the submitting user has been identified. The information is found here.
Unfortunately it was no one from our team :-(....
Well, maybe one of us will be the 1 billionth result submitter...
(That is of course if SETIIIhasn't started by then)
Labelled with ICRA |
PEG and the PEG logo belong to PEG Holdings, Inc. Used with permission. |
by The JavaScript Source |