Don't forget about the Server Relocation on 1st of April !


log in

Advanced search

Message boards : Number crunching : Don't forget about the Server Relocation on 1st of April !

Previous · 1 · 2 · 3 · 4 · 5 · 6 · Next
Author Message
Richard HaselgroveProject donor
Volunteer tester
Send message
Joined: 4 Jul 99
Posts: 8438
Credit: 47,978,186
RAC: 60,628
United Kingdom
Message 1352966 - Posted: 3 Apr 2013, 20:57:35 UTC - in response to Message 1352960.

According to the name it's a gigabitethernet link:

Graphs for gigabitethernet6_17

Claggy

The old Cricket was a gigabit interface too - gigabitethernet2_3, to be precise.

But that only tells us about the capability of that single interface on that particular router - doesn't tell us anything about the other hardware, the cabling, and the software policies in place elsewhere on the route.

Profile Karsten Vinding
Volunteer tester
Send message
Joined: 18 May 99
Posts: 140
Credit: 16,479,740
RAC: 2,405
Denmark
Message 1352967 - Posted: 3 Apr 2013, 20:57:47 UTC - in response to Message 1352965.

The link just passed 200Mbit:

Cur: 203.20 Mbits/sec
____________

Profile HAL9000
Volunteer tester
Avatar
Send message
Joined: 11 Sep 99
Posts: 3996
Credit: 110,078,454
RAC: 137,220
United States
Message 1352988 - Posted: 3 Apr 2013, 21:41:44 UTC - in response to Message 1352967.

The link just passed 200Mbit:

Cur: 203.20 Mbits/sec

It did peak at at a bit over 210Mb around 14:00 on the graph.

It was mentioned that with the Hurricane connection only running 100Mb they were paying a reduced rate. So even if the CoLo allows full Gb use we might still be limited in the bandwidth we use. At the moment they could be playing with a limit on it to see where it holds steady & then cap it off around that. Probably by tomorrow we will see a nice steady graph. Which I expect to be in the form of a wave.
____________
SETI@home classic workunits: 93,865 CPU time: 863,447 hours

Join the BP6/VP6 User Group today!

ClaggyProject donor
Volunteer tester
Send message
Joined: 5 Jul 99
Posts: 4058
Credit: 32,805,314
RAC: 5,214
United Kingdom
Message 1352994 - Posted: 3 Apr 2013, 21:45:36 UTC - in response to Message 1352966.

According to the name it's a gigabitethernet link:

Graphs for gigabitethernet6_17

Claggy

The old Cricket was a gigabit interface too - gigabitethernet2_3, to be precise.

But that only tells us about the capability of that single interface on that particular router - doesn't tell us anything about the other hardware, the cabling, and the software policies in place elsewhere on the route.

Once we get a Shortie Storm, with AP thrown in, we'll see how high the limit is, i reckon we can maybe generate around 500Mbits/sec then, we'll see.

Claggy

Profile HAL9000
Volunteer tester
Avatar
Send message
Joined: 11 Sep 99
Posts: 3996
Credit: 110,078,454
RAC: 137,220
United States
Message 1352999 - Posted: 3 Apr 2013, 21:47:56 UTC - in response to Message 1352994.

According to the name it's a gigabitethernet link:

Graphs for gigabitethernet6_17

Claggy

The old Cricket was a gigabit interface too - gigabitethernet2_3, to be precise.

But that only tells us about the capability of that single interface on that particular router - doesn't tell us anything about the other hardware, the cabling, and the software policies in place elsewhere on the route.

Once we get a Shortie Storm, with AP thrown in, we'll see how high the limit is, i reckon we can maybe generate around 500Mbits/sec then, we'll see.

Claggy

At the moment it looks like the feeder limits are keeping us around 200Mb. It could be another factor, but I am speculation it is the feeder being emptied at the moment.
____________
SETI@home classic workunits: 93,865 CPU time: 863,447 hours

Join the BP6/VP6 User Group today!

N9JFE David S
Volunteer tester
Avatar
Send message
Joined: 4 Oct 99
Posts: 11031
Credit: 13,834,729
RAC: 15,063
United States
Message 1353069 - Posted: 4 Apr 2013, 1:39:30 UTC - in response to Message 1352999.

At the moment it looks like the feeder limits are keeping us around 200Mb. It could be another factor, but I am speculation it is the feeder being emptied at the moment.

If you didn't see his comment in the News thread, Matt said the upload server seems to be a bottleneck and he has some other throttles in place for now.

____________
David
Sitting on my butt while others boldly go,
Waiting for a message from a small furry creature from Alpha Centauri.


Profile rebestProject donor
Volunteer tester
Avatar
Send message
Joined: 16 Apr 00
Posts: 1296
Credit: 32,174,082
RAC: 8,976
United States
Message 1353081 - Posted: 4 Apr 2013, 2:27:11 UTC

Wow. My entire 3-day backlog of uploads on 5952933 cleared in just 6 minutes. Fantastic!

No downloads available yet. I still have about 10 hours of Astropulse work left so if all goes well tomorrow... >>BIG SMILE<<
____________

Join the PACK!

Profile RottenMutt
Avatar
Send message
Joined: 15 Mar 01
Posts: 992
Credit: 207,654,737
RAC: 3
United States
Message 1353093 - Posted: 4 Apr 2013, 3:24:50 UTC - in response to Message 1350154.
Last modified: 4 Apr 2013, 3:42:55 UTC

i just got some new work! MB and AP!!!

mostly i'm getting no new work... nothing new. yet the status page says it is there, before I blamed it on the their network. well not now.
____________

Horacio
Send message
Joined: 14 Jan 00
Posts: 536
Credit: 71,088,748
RAC: 82,250
Argentina
Message 1353098 - Posted: 4 Apr 2013, 4:05:33 UTC

For some reasson, things are weird here...

All my hosts have been getting CPU work in all the RPCs that only asked for CPU, but when they ask for GPU or CPU+GPU they dont get anything and the "NVIDIA GPU work fetch deferral interval" rises, so BOINC stops asking for GPU work and then it gets CPU work again...
I guess/hope this is related to the several throttles mentioned by Matt and/or the catch up after the long outage...
____________

Profile RottenMutt
Avatar
Send message
Joined: 15 Mar 01
Posts: 992
Credit: 207,654,737
RAC: 3
United States
Message 1353104 - Posted: 4 Apr 2013, 4:20:09 UTC

i've been noticing the same behavior. also we seem to hover around 295,242 work units ready to send and when it spikes to 305,xxx then some work will squirt out.
____________

Lionel
Send message
Joined: 25 Mar 00
Posts: 544
Credit: 221,919,577
RAC: 207,151
Australia
Message 1353105 - Posted: 4 Apr 2013, 4:21:00 UTC - in response to Message 1353098.

For some reasson, things are weird here...

All my hosts have been getting CPU work in all the RPCs that only asked for CPU, but when they ask for GPU or CPU+GPU they dont get anything and the "NVIDIA GPU work fetch deferral interval" rises, so BOINC stops asking for GPU work and then it gets CPU work again...
I guess/hope this is related to the several throttles mentioned by Matt and/or the catch up after the long outage...


same here, it's been like that for a while

____________

Profile Wiggo
Avatar
Send message
Joined: 24 Jan 00
Posts: 6697
Credit: 92,254,418
RAC: 73,269
Australia
Message 1353108 - Posted: 4 Apr 2013, 4:30:46 UTC - in response to Message 1353105.
Last modified: 4 Apr 2013, 4:35:02 UTC

For some reasson, things are weird here...

All my hosts have been getting CPU work in all the RPCs that only asked for CPU, but when they ask for GPU or CPU+GPU they dont get anything and the "NVIDIA GPU work fetch deferral interval" rises, so BOINC stops asking for GPU work and then it gets CPU work again...
I guess/hope this is related to the several throttles mentioned by Matt and/or the catch up after the long outage...


same here, it's been like that for a while

Maybe they're trying to get the CPU's filled 1st now, or the work just isn't suitable for what I want on my GPU's.

Cheers.

Profile RottenMutt
Avatar
Send message
Joined: 15 Mar 01
Posts: 992
Credit: 207,654,737
RAC: 3
United States
Message 1353109 - Posted: 4 Apr 2013, 4:33:37 UTC

well then you would think i would get cpu work then :(

____________

Profile BilBg
Volunteer tester
Avatar
Send message
Joined: 27 May 07
Posts: 2617
Credit: 5,956,852
RAC: 3,608
Bulgaria
Message 1353112 - Posted: 4 Apr 2013, 4:39:48 UTC - in response to Message 1353105.


Not same here.
I get sometimes tasks (obviously when feeder is not empty at the exact time of my request) and they download VERY fast:

904 SETI@home 04.04.2013 06:31:49 Requesting new tasks for GPU
905 SETI@home 04.04.2013 06:31:52 Scheduler request completed: got 0 new tasks
906 SETI@home 04.04.2013 06:31:52 Message from server: Project has no tasks available
907 SETI@home 04.04.2013 06:36:57 Sending scheduler request: To fetch work.
908 SETI@home 04.04.2013 06:36:57 Requesting new tasks for GPU
909 SETI@home 04.04.2013 06:37:00 Scheduler request completed: got 2 new tasks
910 SETI@home 04.04.2013 06:37:02 Started download of ap_01dc12ag_B4_P0_00042_20130314_27195.wu
911 SETI@home 04.04.2013 06:37:02 Started download of ap_24jl12ai_B0_P1_00246_20130403_19995.wu
912 SETI@home 04.04.2013 06:37:17 Finished download of ap_01dc12ag_B4_P0_00042_20130314_27195.wu
913 SETI@home 04.04.2013 06:37:18 Finished download of ap_24jl12ai_B0_P1_00246_20130403_19995.wu





956 SETI@home 04.04.2013 07:18:10 Requesting new tasks for CPU
957 SETI@home 04.04.2013 07:18:13 Scheduler request completed: got 0 new tasks
958 SETI@home 04.04.2013 07:18:13 Message from server: Project has no tasks available
959 SETI@home 04.04.2013 07:23:18 Sending scheduler request: To fetch work.
960 SETI@home 04.04.2013 07:23:18 Requesting new tasks for GPU
961 SETI@home 04.04.2013 07:23:22 Scheduler request completed: got 4 new tasks
962 SETI@home 04.04.2013 07:23:24 Started download of ap_01dc12ac_B5_P0_00178_20130403_03240.wu
963 SETI@home 04.04.2013 07:23:24 Started download of ap_24jl12ai_B0_P1_00356_20130403_19995.wu
964 SETI@home 04.04.2013 07:23:24 Started download of ap_29no12af_B0_P1_00110_20130307_03446.wu
965 SETI@home 04.04.2013 07:23:24 Started download of ap_29no12af_B0_P1_00168_20130307_03446.wu
966 SETI@home 04.04.2013 07:23:43 Finished download of ap_29no12af_B0_P1_00110_20130307_03446.wu
967 SETI@home 04.04.2013 07:23:47 Finished download of ap_24jl12ai_B0_P1_00356_20130403_19995.wu
968 SETI@home 04.04.2013 07:23:49 Finished download of ap_29no12af_B0_P1_00168_20130307_03446.wu
969 SETI@home 04.04.2013 07:23:50 Finished download of ap_01dc12ac_B5_P0_00178_20130403_03240.wu

(do not mind the green - it is not SETI trafic)




____________



- ALF - "Find out what you don't do well ..... then don't do it!" :)

Profile Wiggo
Avatar
Send message
Joined: 24 Jan 00
Posts: 6697
Credit: 92,254,418
RAC: 73,269
Australia
Message 1353113 - Posted: 4 Apr 2013, 4:40:56 UTC - in response to Message 1353109.

1st come 1st serve then, but may be a lot of PC's that havn't been able to connect to the servers for ages are now connecting.

Anyone out there willing enough to undo their TCP settings that they modified not long ago to find out how the "average person" will go with the new connection?

I'm sorry but I can't do it due to meteorology software that I run here (it'll just get changed back again).

Cheers.

TBar
Volunteer tester
Send message
Joined: 22 May 99
Posts: 1198
Credit: 44,248,982
RAC: 114,815
United States
Message 1353119 - Posted: 4 Apr 2013, 5:11:49 UTC
Last modified: 4 Apr 2013, 5:14:49 UTC

Things were looking great here, then went belly up. One machine grabbed a full cache of CUDA tasks, a few replacements...now nothing. The other machine got a few ATI APs, a couple CPU APs, but No CUDA MBs. Then it dried up. When it's full, it should be around 180 tasks, mostly APs. They are going fast...

Where's the Beef?


4/4/2013 12:51:17 AM | SETI@home | Sending scheduler request: To fetch work.
4/4/2013 12:51:17 AM | SETI@home | Requesting new tasks for NVIDIA and ATI
4/4/2013 12:51:19 AM | SETI@home | Scheduler request completed: got 0 new tasks
4/4/2013 12:51:19 AM | SETI@home | Project has no tasks available
4/4/2013 12:59:24 AM | SETI@home | Sending scheduler request: To fetch work.
4/4/2013 12:59:24 AM | SETI@home | Requesting new tasks for ATI
4/4/2013 12:59:25 AM | SETI@home | Scheduler request completed: got 0 new tasks
4/4/2013 12:59:25 AM | SETI@home | Project has no tasks available
4/4/2013 1:06:30 AM | SETI@home | Sending scheduler request: To fetch work.
4/4/2013 1:06:30 AM | SETI@home | Requesting new tasks for NVIDIA
4/4/2013 1:06:32 AM | SETI@home | Scheduler request completed: got 0 new tasks
4/4/2013 1:06:32 AM | SETI@home | Project has no tasks available
...

Profile Grand Admiral Thrawn
Avatar
Send message
Joined: 19 Feb 01
Posts: 53
Credit: 22,263,601
RAC: 0
Austria
Message 1353171 - Posted: 4 Apr 2013, 7:29:51 UTC

Hm, most of my clients can at least communicate with the servers. However, my Linux box running BOINC 7.1.0 gets stuck at "Fetching scheduler list":

Thu 04 Apr 2013 09:25:54 AM CEST | SETI@home | update requested by user
Thu 04 Apr 2013 09:25:57 AM CEST | SETI@home | Fetching scheduler list
Thu 04 Apr 2013 09:26:14 AM CEST | Project communication failed: attempting access to reference site
Thu 04 Apr 2013 09:26:15 AM CEST | Internet access OK - project servers may be temporarily down.

____________
3dfx Voodoo5 6000 AGP HiNT Rev.A 3700 prototype, dead HiNT bridge

Grant (SSSF)
Send message
Joined: 19 Aug 99
Posts: 5774
Credit: 57,517,217
RAC: 48,601
Australia
Message 1353174 - Posted: 4 Apr 2013, 7:45:21 UTC - in response to Message 1350901.

I really DO miss Scarecrow's graphs.

Same here, but the alternative graphs do help a lot.

Seti graphs
____________
Grant
Darwin NT.

Profile ivan
Volunteer tester
Avatar
Send message
Joined: 5 Mar 01
Posts: 597
Credit: 134,078,295
RAC: 116,209
United Kingdom
Message 1353180 - Posted: 4 Apr 2013, 8:06:43 UTC - in response to Message 1353171.

Hm, most of my clients can at least communicate with the servers. However, my Linux box running BOINC 7.1.0 gets stuck at "Fetching scheduler list":

Thu 04 Apr 2013 09:25:54 AM CEST | SETI@home | update requested by user
Thu 04 Apr 2013 09:25:57 AM CEST | SETI@home | Fetching scheduler list
Thu 04 Apr 2013 09:26:14 AM CEST | Project communication failed: attempting access to reference site
Thu 04 Apr 2013 09:26:15 AM CEST | Internet access OK - project servers may be temporarily down.

Try stopping boinc, flushing your DNS cache (perhaps /etc/init.d/nscd restart) and restarting boinc.
____________

Lionel
Send message
Joined: 25 Mar 00
Posts: 544
Credit: 221,919,577
RAC: 207,151
Australia
Message 1353197 - Posted: 4 Apr 2013, 9:45:47 UTC - in response to Message 1353180.

Hm, most of my clients can at least communicate with the servers. However, my Linux box running BOINC 7.1.0 gets stuck at "Fetching scheduler list":

Thu 04 Apr 2013 09:25:54 AM CEST | SETI@home | update requested by user
Thu 04 Apr 2013 09:25:57 AM CEST | SETI@home | Fetching scheduler list
Thu 04 Apr 2013 09:26:14 AM CEST | Project communication failed: attempting access to reference site
Thu 04 Apr 2013 09:26:15 AM CEST | Internet access OK - project servers may be temporarily down.

Try stopping boinc, flushing your DNS cache (perhaps /etc/init.d/nscd restart) and restarting boinc.


rebooted the machines a while ago ... saw no change in behaviour ... still behaving as above ...
____________

Previous · 1 · 2 · 3 · 4 · 5 · 6 · Next

Message boards : Number crunching : Don't forget about the Server Relocation on 1st of April !

Copyright © 2014 University of California