Panic Mode On (82) Server Problems?

Message boards : Number crunching : Panic Mode On (82) Server Problems?
Message board moderation

To post messages, you must log in.

Previous · 1 . . . 18 · 19 · 20 · 21 · 22 · 23 · 24 · Next

AuthorMessage
Profile [B^S] madmac
Volunteer tester
Avatar

Send message
Joined: 9 Feb 04
Posts: 1175
Credit: 4,754,897
RAC: 0
United Kingdom
Message 1356039 - Posted: 12 Apr 2013, 10:24:44 UTC

Is that why when I tried to upload I got cannot resolve hostname yet when I uploaded just now I got no problems?
ID: 1356039 · Report as offensive
Richard Haselgrove Project Donor
Volunteer tester

Send message
Joined: 4 Jul 99
Posts: 14649
Credit: 200,643,578
RAC: 874
United Kingdom
Message 1356041 - Posted: 12 Apr 2013, 10:36:16 UTC - in response to Message 1356039.  

Is that why when I tried to upload I got cannot resolve hostname yet when I uploaded just now I got no problems?

Yup.
ID: 1356041 · Report as offensive
Gone

Send message
Joined: 31 May 99
Posts: 150
Credit: 125,779,206
RAC: 0
United Kingdom
Message 1356043 - Posted: 12 Apr 2013, 10:48:56 UTC

The Berkeley IST status page http://ucbsystems.org/
is showing no problems.
Perhaps they haven't updated yet or it was an upstream problem with their ISP.
ID: 1356043 · Report as offensive
Profile Cliff Harding
Volunteer tester
Avatar

Send message
Joined: 18 Aug 99
Posts: 1432
Credit: 110,967,840
RAC: 67
United States
Message 1356044 - Posted: 12 Apr 2013, 11:17:37 UTC - in response to Message 1356021.  

Yeah, everything on berkeley.edu went down.

Yes the whole university dropped off the internet. I checked several is-it-up sites and they all reported it as down.



Scared the S(&%E$%T out of me. I had just uninstalled a MS update from the last update cycle per MS email and had recycled the machine and noticed that BOINC could not connect to the servers. Couldn't connect/browse to anything @ Berkeley, including the IST. Was able to reach the Lunatics site where it was noted that the whole domain was down. IST has not been updated as to case as of yet.


I don't buy computers, I build them!!
ID: 1356044 · Report as offensive
Profile Jord
Volunteer tester
Avatar

Send message
Joined: 9 Jun 99
Posts: 15184
Credit: 4,362,181
RAC: 3
Netherlands
Message 1356052 - Posted: 12 Apr 2013, 12:13:26 UTC - in response to Message 1356043.  
Last modified: 12 Apr 2013, 12:16:05 UTC

The Berkeley IST status page http://ucbsystems.org/ is showing no problems.

They weren't available either earlier when all of Berkeley was down. I checked that site, it was gone as well. Couldn't ping either.

I even checked if it hadn't been an earthquake, but nothing there.
ID: 1356052 · Report as offensive
kittyman Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 9 Jul 00
Posts: 51468
Credit: 1,018,363,574
RAC: 1,004
United States
Message 1356063 - Posted: 12 Apr 2013, 12:57:34 UTC - in response to Message 1356052.  

The Berkeley IST status page http://ucbsystems.org/ is showing no problems.

They weren't available either earlier when all of Berkeley was down. I checked that site, it was gone as well. Couldn't ping either.

I even checked if it hadn't been an earthquake, but nothing there.

No quakes here...
And all kitty caches have since recovered.
"Freedom is just Chaos, with better lighting." Alan Dean Foster

ID: 1356063 · Report as offensive
Profile Gary Charpentier Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 25 Dec 00
Posts: 30608
Credit: 53,134,872
RAC: 32
United States
Message 1356086 - Posted: 12 Apr 2013, 13:42:43 UTC - in response to Message 1356052.  

The Berkeley IST status page http://ucbsystems.org/ is showing no problems.

They weren't available either earlier when all of Berkeley was down. I checked that site, it was gone as well. Couldn't ping either.

I even checked if it hadn't been an earthquake, but nothing there.

Whatever it was, I note looking at the crickets all of them have a hole in the data, so it was some on campus event. IST may never say. Could be an embarrassment or could be security reasons to keep quiet.

ID: 1356086 · Report as offensive
ExchangeMan
Volunteer tester

Send message
Joined: 9 Jan 00
Posts: 115
Credit: 157,719,104
RAC: 0
United States
Message 1356097 - Posted: 12 Apr 2013, 14:05:16 UTC

All 3 of my crunchers reported name resolution problems. I was a little relieved when I couldn't get to the seti@home website from my cell.

ID: 1356097 · Report as offensive
kittyman Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 9 Jul 00
Posts: 51468
Credit: 1,018,363,574
RAC: 1,004
United States
Message 1356107 - Posted: 12 Apr 2013, 14:28:22 UTC

We gotz shorties goin' on.....

The kitties just luv shorties. Especially since the improved bandwidth makes it much less of a problem to keep the caches filled.

My rigs just crunch the bejeezus out of them.

Meow!
"Freedom is just Chaos, with better lighting." Alan Dean Foster

ID: 1356107 · Report as offensive
ExchangeMan
Volunteer tester

Send message
Joined: 9 Jan 00
Posts: 115
Credit: 157,719,104
RAC: 0
United States
Message 1356184 - Posted: 12 Apr 2013, 16:28:14 UTC - in response to Message 1356107.  

We gotz shorties goin' on.....

The kitties just luv shorties. Especially since the improved bandwidth makes it much less of a problem to keep the caches filled.

My rigs just crunch the bejeezus out of them.

Meow!

Same here. My massive cruncher just blazes through them. I seem to average about 40 shorties every 5 minute update cycle - WOW!!! I don't mind them as long as there are no bottlenecks created within the Seti system somewhere and you can get all you can eat.

ID: 1356184 · Report as offensive
Profile zoom3+1=4
Volunteer tester
Avatar

Send message
Joined: 30 Nov 03
Posts: 65709
Credit: 55,293,173
RAC: 49
United States
Message 1356220 - Posted: 12 Apr 2013, 17:32:28 UTC - in response to Message 1356052.  

The Berkeley IST status page http://ucbsystems.org/ is showing no problems.

They weren't available either earlier when all of Berkeley was down. I checked that site, it was gone as well. Couldn't ping either.

I even checked if it hadn't been an earthquake, but nothing there.

I did a tracert and a ping, 13 hops was as much as I could see, sorry that I don't know what the last hop was, I noted cenic.net was before it.

Oh and I'm able to download gpu wu's and I can't get any cpu wu's, reporting and uploading seem to work ok though, I checked and I'm setup for both here, this didn't happen before this mini night outage.
The T1 Trust, PRR T1 Class 4-4-4-4 #5550, 1 of America's First HST's
ID: 1356220 · Report as offensive
Cherokee150

Send message
Joined: 11 Nov 99
Posts: 192
Credit: 58,513,758
RAC: 74
United States
Message 1356222 - Posted: 12 Apr 2013, 17:40:18 UTC

The new Colo is a godsend to those of you at the highest end of the top cruncher list. Now that you can get all you can eat, your TC and RAC are going to skyrocket.

I wonder who will end up on top now? It will be most interesting to watch.

Sorry, I can't participate. I'm nowhere near fast enough to enter into the race, heh heh. ;) I'll just sit in the stands and watch the fur fly. (I'm just using that as a figure of speech. Nothing is aimed at -any- of the beautiful furry creatures or those who love them, Mark!)
ID: 1356222 · Report as offensive
Profile Bernie Vine
Volunteer moderator
Volunteer tester
Avatar

Send message
Joined: 26 May 99
Posts: 9954
Credit: 103,452,613
RAC: 328
United Kingdom
Message 1356223 - Posted: 12 Apr 2013, 17:45:25 UTC - in response to Message 1356222.  
Last modified: 12 Apr 2013, 17:45:41 UTC

The new Colo is a godsend to those of you at the highest end of the top cruncher list. Now that you can get all you can eat, your TC and RAC are going to skyrocket.

I wonder who will end up on top now? It will be most interesting to watch.

Sorry, I can't participate. I'm nowhere near fast enough to enter into the race, heh heh. ;) I'll just sit in the stands and watch the fur fly. (I'm just using that as a figure of speech. Nothing is aimed at -any- of the beautiful furry creatures or those who love them, Mark!)

The limits of 100 tasks in progress per processor(not 100 per core) and 100 per machine for GPU (no matter no of GPU's) are still in place.
ID: 1356223 · Report as offensive
ExchangeMan
Volunteer tester

Send message
Joined: 9 Jan 00
Posts: 115
Credit: 157,719,104
RAC: 0
United States
Message 1356231 - Posted: 12 Apr 2013, 17:57:07 UTC - in response to Message 1356223.  

The new Colo is a godsend to those of you at the highest end of the top cruncher list. Now that you can get all you can eat, your TC and RAC are going to skyrocket.

I wonder who will end up on top now? It will be most interesting to watch.

Sorry, I can't participate. I'm nowhere near fast enough to enter into the race, heh heh. ;) I'll just sit in the stands and watch the fur fly. (I'm just using that as a figure of speech. Nothing is aimed at -any- of the beautiful furry creatures or those who love them, Mark!)

The limits of 100 tasks in progress per processor(not 100 per core) and 100 per machine for GPU (no matter no of GPU's) are still in place.

The 100 limit even for GPU will work as long as Seti is up and can give out work units on demand with no bottlenecks, but you won't have the ability to handle any type of outage or hiccup without going dry for some period of time. That's the danger for big crunchers, especially if something happens in the middle of a shorty storm.

ID: 1356231 · Report as offensive
kittyman Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 9 Jul 00
Posts: 51468
Credit: 1,018,363,574
RAC: 1,004
United States
Message 1356233 - Posted: 12 Apr 2013, 18:00:06 UTC - in response to Message 1356231.  

The new Colo is a godsend to those of you at the highest end of the top cruncher list. Now that you can get all you can eat, your TC and RAC are going to skyrocket.

I wonder who will end up on top now? It will be most interesting to watch.

Sorry, I can't participate. I'm nowhere near fast enough to enter into the race, heh heh. ;) I'll just sit in the stands and watch the fur fly. (I'm just using that as a figure of speech. Nothing is aimed at -any- of the beautiful furry creatures or those who love them, Mark!)

The limits of 100 tasks in progress per processor(not 100 per core) and 100 per machine for GPU (no matter no of GPU's) are still in place.

The 100 limit even for GPU will work as long as Seti is up and can give out work units on demand with no bottlenecks, but you won't have the ability to handle any type of outage or hiccup without going dry for some period of time. That's the danger for big crunchers, especially if something happens in the middle of a shorty storm.

Yeah, my better crunchers can't even make it through a Tuesday outage without running the GPUs dry most of the time.

"Freedom is just Chaos, with better lighting." Alan Dean Foster

ID: 1356233 · Report as offensive
Profile zoom3+1=4
Volunteer tester
Avatar

Send message
Joined: 30 Nov 03
Posts: 65709
Credit: 55,293,173
RAC: 49
United States
Message 1356234 - Posted: 12 Apr 2013, 18:00:20 UTC - in response to Message 1356223.  

The new Colo is a godsend to those of you at the highest end of the top cruncher list. Now that you can get all you can eat, your TC and RAC are going to skyrocket.

I wonder who will end up on top now? It will be most interesting to watch.

Sorry, I can't participate. I'm nowhere near fast enough to enter into the race, heh heh. ;) I'll just sit in the stands and watch the fur fly. (I'm just using that as a figure of speech. Nothing is aimed at -any- of the beautiful furry creatures or those who love them, Mark!)

The limits of 100 tasks in progress per processor(not 100 per core) and 100 per machine for GPU (no matter no of GPU's) are still in place.

Wonderful, I've got 6 cpu wu's left and the cpu can't get anymore, yet I download up to the limit for the gpu, something is busted I think.
The T1 Trust, PRR T1 Class 4-4-4-4 #5550, 1 of America's First HST's
ID: 1356234 · Report as offensive
Profile Gary Charpentier Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 25 Dec 00
Posts: 30608
Credit: 53,134,872
RAC: 32
United States
Message 1356237 - Posted: 12 Apr 2013, 18:06:32 UTC
Last modified: 12 Apr 2013, 18:06:42 UTC

For those that haven't seen it yet ...
http://ucbsystems.org/
Outage Type: UNSCHEDULED OUTAGE (RESOLVED)
Date Submitted: Friday, April 12, 2013
Outage Start/End Time: 0015 – 0210
Groups Impacted: Campus Wide
Equipment: Various Campus Sites

Description: There were multiple reports of access issues from off campus to many sites on the UC Berkeley campus network, from about 12:15am until about 2:10am Friday, April 12th. IST staff is investigating the cause of the interruption.

Network connectivity appears normal again as of 2:10am. IST staff will update as we are able to provide more details of the network interruption.

Please direct any questions you may have regarding this announcement to: Siegrid Rickenbach

CMR: 2065

ID: 1356237 · Report as offensive
Cherokee150

Send message
Joined: 11 Nov 99
Posts: 192
Credit: 58,513,758
RAC: 74
United States
Message 1356238 - Posted: 12 Apr 2013, 18:10:16 UTC - in response to Message 1356223.  
Last modified: 12 Apr 2013, 18:23:49 UTC

Yes, Bernie, you are correct.

However, the fastest machines' GPUs can burn through those 100 limits in just minutes. This is far faster than they have been able to download new units for a -very- long time. The download bandwidth problems, especially combined with the 100 unit limits, often affected even my lowly RAC speeds.

Now that the bandwidth bottleneck has been removed, I expect to see the big guys process faster than ever!

[EDIT] Yes, outages cause headaches for all of us, and the big guys get hurt first. However, I have read complaint after complaint about the tremendous problems keeping the fastest rigs fed under the old bandwidth constraints. Many times my own fastest rig, with only one GTX 460, would process all the units I had faster than the next batch would download, even with constant manual intervention using all the tricks to nurse them through.

Outages are a completely different problem. Nonetheless, I just have to believe that the new bandwidths from the Colo will benefit all of us, especially the big guys, and I believe it will make all our lives much less stressful. :)
ID: 1356238 · Report as offensive
kittyman Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 9 Jul 00
Posts: 51468
Credit: 1,018,363,574
RAC: 1,004
United States
Message 1356244 - Posted: 12 Apr 2013, 18:31:00 UTC - in response to Message 1356234.  


Wonderful, I've got 6 cpu wu's left and the cpu can't get anymore, yet I download up to the limit for the gpu, something is busted I think.

I think it's on your end, Vic.
Something not right in Boinc.
None of my rigs is having any trouble maintaining their cache limits on both CPU and GPU.
"Freedom is just Chaos, with better lighting." Alan Dean Foster

ID: 1356244 · Report as offensive
Profile zoom3+1=4
Volunteer tester
Avatar

Send message
Joined: 30 Nov 03
Posts: 65709
Credit: 55,293,173
RAC: 49
United States
Message 1356252 - Posted: 12 Apr 2013, 18:47:12 UTC - in response to Message 1356244.  
Last modified: 12 Apr 2013, 18:52:09 UTC


Wonderful, I've got 6 cpu wu's left and the cpu can't get anymore, yet I download up to the limit for the gpu, something is busted I think.

I think it's on your end, Vic.
Something not right in Boinc.
None of my rigs is having any trouble maintaining their cache limits on both CPU and GPU.

Well I cleared the settings on this end, Boinc still asks for cpu work, but Seti won't send any, yet I get up to the limit for cuda.


211 SETI@home 4/12/2013 11:51:08 AM Reporting 1 completed tasks, requesting new tasks for CPU and GPU
212 SETI@home 4/12/2013 11:51:09 AM Scheduler request completed: got 0 new tasks
213 SETI@home 4/12/2013 11:51:09 AM Message from server: Not sending work - last request too recent: 57 sec


All for the gpu, I'm set for both @ Seti of course, My app info file has 603 file settings as well as 610, so outside of no cpu downloads I'm stumped.
The T1 Trust, PRR T1 Class 4-4-4-4 #5550, 1 of America's First HST's
ID: 1356252 · Report as offensive
Previous · 1 . . . 18 · 19 · 20 · 21 · 22 · 23 · 24 · Next

Message boards : Number crunching : Panic Mode On (82) Server Problems?


 
©2024 University of California
 
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.