Panic Mode On (81) Server Problems?

Message boards : Number crunching : Panic Mode On (81) Server Problems?
Message board moderation

To post messages, you must log in.

Previous · 1 . . . 13 · 14 · 15 · 16 · 17 · 18 · 19 . . . 21 · Next

AuthorMessage
Cosmic_Ocean
Avatar

Send message
Joined: 23 Dec 00
Posts: 3027
Credit: 13,516,867
RAC: 13
United States
Message 1336922 - Posted: 11 Feb 2013, 3:12:24 UTC - in response to Message 1336523.  

Just out of curiosity.. in the HTTP Proxy tab.. "don't use proxy for:" field.. how do I tell it not to use it for uploads and scheduler contacts? Is that going to be something like "sched_op" or something along those lines? I get HTTP errors most of the time on scheduler contacts when using a proxy, and only one that I've used so far was able to upload.

Proxy only really benefits for download, but upload and scheduler contact seem to work just fine without one.. so how do I limit proxy usage to just downloads? Or can I even?


And yes, I know the crickets have died and something is seriously broken. I got one failed scheduler attempt earlier that said "HTTP service unavailable" followed by "error when receiving data from the peer."


setiboincdata.ssl.berkeley.edu will stop it from using proxy for uploads. don't know about scheduler contact.

Looking through my messages log, I got a date stamp from a successful contact, and then cross-referenced to my router's connection logs, and came up with setiboinc.ssl.berkeley.edu for scheduler. Then boinc2.ssl.berkeley.edu is for downloads.

So this is working, but I don't know how to put multiple items in the box. I've tried a simple space, I've tried a comma, I've tried a semi-colon. Anything after the first FQDN is ignored, apparently.
Linux laptop:
record uptime: 1511d 20h 19m (ended due to the power brick giving-up)
ID: 1336922 · Report as offensive
Cosmic_Ocean
Avatar

Send message
Joined: 23 Dec 00
Posts: 3027
Credit: 13,516,867
RAC: 13
United States
Message 1336971 - Posted: 11 Feb 2013, 6:20:21 UTC

This proxy deal is working out quite well for AP downloads. Here's a snippet of my messages tab, with nothing trimmed or removed for this batch of downloads:

2013-02-10 21:00:36 SETI@home Sending scheduler request: To fetch work.
2013-02-10 21:00:36 SETI@home Requesting new tasks
2013-02-10 21:00:36 SETI@home [sched_op_debug] CPU work request: 195139.09 seconds; 0.00 CPUs
2013-02-10 21:00:41 SETI@home Scheduler request completed: got 4 new tasks
2013-02-10 21:00:41 SETI@home [sched_op_debug] Server version 701
2013-02-10 21:00:41 SETI@home Project requested delay of 303 seconds
2013-02-10 21:00:41 SETI@home [sched_op_debug] estimated total CPU job duration: 178631 seconds
2013-02-10 21:00:41 SETI@home [sched_op_debug] Deferring communication for 5 min 3 sec
2013-02-10 21:00:41 SETI@home [sched_op_debug] Reason: requested by project
2013-02-10 21:00:43 SETI@home Started download of ap_02ja13ad_B0_P0_00282_20130210_04474.wu
2013-02-10 21:00:43 SETI@home Started download of ap_31dc12ab_B2_P0_00162_20130210_11991.wu
2013-02-10 21:00:43 SETI@home Started download of ap_31dc12ab_B1_P1_00384_20130210_00911.wu
2013-02-10 21:00:43 SETI@home Started download of ap_30dc12ab_B4_P0_00086_20130209_11736.wu
2013-02-10 21:05:46 SETI@home [sched_op_debug] Starting scheduler request
2013-02-10 21:05:46 SETI@home Sending scheduler request: To fetch work.
2013-02-10 21:05:46 SETI@home Requesting new tasks
2013-02-10 21:05:46 SETI@home [sched_op_debug] CPU work request: 15879.33 seconds; 0.00 CPUs
2013-02-10 21:05:49 SETI@home Scheduler request completed: got 0 new tasks
2013-02-10 21:05:49 SETI@home [sched_op_debug] Server version 701
2013-02-10 21:05:49 SETI@home Message from server: Project has no tasks available
2013-02-10 21:05:49 SETI@home Project requested delay of 303 seconds
2013-02-10 21:05:49 SETI@home [sched_op_debug] Deferring communication for 5 min 3 sec
2013-02-10 21:05:49 SETI@home [sched_op_debug] Reason: requested by project
2013-02-10 21:06:48 SETI@home Finished download of ap_02ja13ad_B0_P0_00282_20130210_04474.wu
2013-02-10 21:07:21 SETI@home Finished download of ap_31dc12ab_B1_P1_00384_20130210_00911.wu
2013-02-10 21:07:26 SETI@home Finished download of ap_30dc12ab_B4_P0_00086_20130209_11736.wu
2013-02-10 21:08:00 SETI@home Finished download of ap_31dc12ab_B2_P0_00162_20130210_11991.wu


Longest one took 8 minutes, which is an average of 17KB/sec. The point is no stalls, no time-outs.
Linux laptop:
record uptime: 1511d 20h 19m (ended due to the power brick giving-up)
ID: 1336971 · Report as offensive
Profile TRuEQ & TuVaLu
Volunteer tester
Avatar

Send message
Joined: 4 Oct 99
Posts: 505
Credit: 69,523,653
RAC: 10
Sweden
Message 1337111 - Posted: 11 Feb 2013, 17:34:43 UTC - in response to Message 1336971.  

This proxy deal is working out quite well for AP downloads. Here's a snippet of my messages tab, with nothing trimmed or removed for this batch of downloads:

2013-02-10 21:00:36 SETI@home Sending scheduler request: To fetch work.
2013-02-10 21:00:36 SETI@home Requesting new tasks
2013-02-10 21:00:36 SETI@home [sched_op_debug] CPU work request: 195139.09 seconds; 0.00 CPUs
2013-02-10 21:00:41 SETI@home Scheduler request completed: got 4 new tasks
2013-02-10 21:00:41 SETI@home [sched_op_debug] Server version 701
2013-02-10 21:00:41 SETI@home Project requested delay of 303 seconds
2013-02-10 21:00:41 SETI@home [sched_op_debug] estimated total CPU job duration: 178631 seconds
2013-02-10 21:00:41 SETI@home [sched_op_debug] Deferring communication for 5 min 3 sec
2013-02-10 21:00:41 SETI@home [sched_op_debug] Reason: requested by project
2013-02-10 21:00:43 SETI@home Started download of ap_02ja13ad_B0_P0_00282_20130210_04474.wu
2013-02-10 21:00:43 SETI@home Started download of ap_31dc12ab_B2_P0_00162_20130210_11991.wu
2013-02-10 21:00:43 SETI@home Started download of ap_31dc12ab_B1_P1_00384_20130210_00911.wu
2013-02-10 21:00:43 SETI@home Started download of ap_30dc12ab_B4_P0_00086_20130209_11736.wu
2013-02-10 21:05:46 SETI@home [sched_op_debug] Starting scheduler request
2013-02-10 21:05:46 SETI@home Sending scheduler request: To fetch work.
2013-02-10 21:05:46 SETI@home Requesting new tasks
2013-02-10 21:05:46 SETI@home [sched_op_debug] CPU work request: 15879.33 seconds; 0.00 CPUs
2013-02-10 21:05:49 SETI@home Scheduler request completed: got 0 new tasks
2013-02-10 21:05:49 SETI@home [sched_op_debug] Server version 701
2013-02-10 21:05:49 SETI@home Message from server: Project has no tasks available
2013-02-10 21:05:49 SETI@home Project requested delay of 303 seconds
2013-02-10 21:05:49 SETI@home [sched_op_debug] Deferring communication for 5 min 3 sec
2013-02-10 21:05:49 SETI@home [sched_op_debug] Reason: requested by project
2013-02-10 21:06:48 SETI@home Finished download of ap_02ja13ad_B0_P0_00282_20130210_04474.wu
2013-02-10 21:07:21 SETI@home Finished download of ap_31dc12ab_B1_P1_00384_20130210_00911.wu
2013-02-10 21:07:26 SETI@home Finished download of ap_30dc12ab_B4_P0_00086_20130209_11736.wu
2013-02-10 21:08:00 SETI@home Finished download of ap_31dc12ab_B2_P0_00162_20130210_11991.wu


Longest one took 8 minutes, which is an average of 17KB/sec. The point is no stalls, no time-outs.


That is how it was for me before New Year. 15KB/sec.
Now it is kinda hard getting ap tasks. Transfers stalls most of the time in countown for server backoffs.
But again I am in Northen Europe, kinda far away from Berkely servers.
ID: 1337111 · 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 1337671 - Posted: 13 Feb 2013, 7:19:34 UTC

Not really panic mode....since today's outage the kitties have refilled the crunchers' caches, so things must be flowing OK.

But I am wondering if it is just coincidental that the Cricket graphs dropped dead during the outage, OR, was there some change made in the path that the servers are transferring data?

Curious, as it is so rare for the Cricket graphs to go down.
"Freedom is just Chaos, with better lighting." Alan Dean Foster

ID: 1337671 · Report as offensive
Profile KWSN THE Holy Hand Grenade!
Volunteer tester
Avatar

Send message
Joined: 20 Dec 05
Posts: 3187
Credit: 57,163,290
RAC: 0
United States
Message 1337677 - Posted: 13 Feb 2013, 7:34:24 UTC

here's an odd one: the daily cricket graph has had no data since (roughtly...) 1300 2/12/13 - but the weekly has shown full usage during the period... (except when the outage was occurring...)
.

Hello, from Albany, CA!...
ID: 1337677 · Report as offensive
Profile Wiggo
Avatar

Send message
Joined: 24 Jan 00
Posts: 34744
Credit: 261,360,520
RAC: 489
Australia
Message 1337681 - Posted: 13 Feb 2013, 7:38:03 UTC - in response to Message 1337677.  

here's an odd one: the daily cricket graph has had no data since (roughtly...) 1300 2/12/13 - but the weekly has shown full usage during the period... (except when the outage was occurring...)

You really need to take a closer look there as I see it much differently. ;)

Cheers.
ID: 1337681 · 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 1337685 - Posted: 13 Feb 2013, 7:41:44 UTC - in response to Message 1337677.  

here's an odd one: the daily cricket graph has had no data since (roughtly...) 1300 2/12/13 - but the weekly has shown full usage during the period... (except when the outage was occurring...)

I don't see that with this link...the weekly graph stops at the same time as the daily, and has not come back up since.
"Freedom is just Chaos, with better lighting." Alan Dean Foster

ID: 1337685 · Report as offensive
Cosmic_Ocean
Avatar

Send message
Joined: 23 Dec 00
Posts: 3027
Credit: 13,516,867
RAC: 13
United States
Message 1337704 - Posted: 13 Feb 2013, 8:31:45 UTC

The graphs are fed by SNMP multicast packets and are captured using SNMP traps. Either the device stopped sending the packets out, or the trap stopped listening. Hard telling.


I did notice something that I haven't been able to do in a long time though. I had reported some tasks minutes before maintenance went down (I pulled up the task pages and was adding the new tasks that I received overnight to my spreadsheet, and when I hit "next 20" to go to the ones that I reported, it had gone down during that 2-3 minutes), and when it all came back up afterward, those completed tasks were still there.

In the past.. year+, I know tasks are supposed to stay visible for ~24h, but during the maintenance, even if it has only been 20 minutes, if it is validated and assimilated, it gets purged during the clean-up maintenance. This time, that did not happen. I was glad, too, because there were several of them and I hate having blanks in my spreadsheet.
Linux laptop:
record uptime: 1511d 20h 19m (ended due to the power brick giving-up)
ID: 1337704 · Report as offensive
Profile James Sotherden
Avatar

Send message
Joined: 16 May 99
Posts: 10436
Credit: 110,373,059
RAC: 54
United States
Message 1337739 - Posted: 13 Feb 2013, 14:02:16 UTC

APs are still downloading slow as heck. My fastest speed so far is 8 KBps. Well at least they are downloading.
[/quote]

Old James
ID: 1337739 · Report as offensive
Cruncher-American Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor

Send message
Joined: 25 Mar 02
Posts: 1513
Credit: 370,893,186
RAC: 340
United States
Message 1337757 - Posted: 13 Feb 2013, 14:56:06 UTC

It seems like Cricket has been dead since about 6PM (Eastern) yesterday. Who should/ how should we contact to notify them (assuming they don't know)?
ID: 1337757 · Report as offensive
Josef W. Segur
Volunteer developer
Volunteer tester

Send message
Joined: 30 Oct 99
Posts: 4504
Credit: 1,414,761
RAC: 0
United States
Message 1337759 - Posted: 13 Feb 2013, 15:19:46 UTC

Given that all the Cricket graphs for all the UCB routers went dead at the same time, I'm sure they know.
                                                                   Joe
ID: 1337759 · Report as offensive
juan BFP Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 16 Mar 07
Posts: 9786
Credit: 572,710,851
RAC: 3,799
Panama
Message 1337762 - Posted: 13 Feb 2013, 15:28:04 UTC - in response to Message 1337739.  

APs are still downloading slow as heck. My fastest speed so far is 8 KBps. Well at least they are downloading.

Nothing was changed on the bandwidth we all use the same all ready overloaded 100MBps link... Snif!

ID: 1337762 · 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 1337769 - Posted: 13 Feb 2013, 15:47:05 UTC - in response to Message 1337759.  
Last modified: 13 Feb 2013, 15:48:14 UTC

Given that all the Cricket graphs for all the UCB routers went dead at the same time, I'm sure they know.
                                                                   Joe

Thank you, Joe. I was unaware of that. I had not looked at any links for any of the other Cricket monitors.

And, we should not try to contact anybody about them regardless.

Most folks, if they are not aware, should know a couple of other things as well.

The Cricket graphs are primarily for the use of the campus IT staff. Not the Seti project. They are courteous enough that our use of them either is accepted from outside of the campus, or the few of us that use them causes no problems for them.

And, the use of live links with the img tag to display the actual graphs was discouraged some time ago, to prevent the graph server from being pinged every time a Seti forum thread containing them is viewed. Most have respected that since then. That's why I only post a link that can be viewed when clicked upon, not show an updated graph on each and every post view.

I do hope they come back up, as they are a great tool for us Seti obsessed types off campus.
"Freedom is just Chaos, with better lighting." Alan Dean Foster

ID: 1337769 · Report as offensive
Profile James Sotherden
Avatar

Send message
Joined: 16 May 99
Posts: 10436
Credit: 110,373,059
RAC: 54
United States
Message 1337772 - Posted: 13 Feb 2013, 15:52:41 UTC - in response to Message 1337769.  

Given that all the Cricket graphs for all the UCB routers went dead at the same time, I'm sure they know.
                                                                   Joe

Thank you, Joe. I was unaware of that. I had not looked at any links for any of the other Cricket monitors.

And, we should not try to contact anybody about them regardless.

Most folks, if they are not aware, should know a couple of other things as well.

The Cricket graphs are primarily for the use of the campus IT staff. Not the Seti project. They are courteous enough that our use of them either is accepted from outside of the campus, or the few of us that use them causes no problems for them.

And, the use of live links with the img tag to display the actual graphs was discouraged some time ago, to prevent the graph server from being pinged every time a Seti forum thread containing them is viewed. Most have respected that since then. That's why I only post a link that can be viewed when clicked upon, not show an updated graph on each and every post view.

I do hope they come back up, as they are a great tool for us Seti obsessed types off campus.

Thanks for that reminder Mark about live links to the cricket graphs. That was a big no-no back then for posting them, And it still is today.
[/quote]

Old James
ID: 1337772 · Report as offensive
Richard Haselgrove Project Donor
Volunteer tester

Send message
Joined: 4 Jul 99
Posts: 14644
Credit: 200,643,578
RAC: 874
United Kingdom
Message 1337782 - Posted: 13 Feb 2013, 16:39:56 UTC - in response to Message 1337772.  

I don't think we've ever been officially discouraged from posting live cricket links - certainly Eric himself has posted them in the past (see message 918297). I have a vague memory that we had a discussion on the subject last time somebody got a bee in their bonnet and started hiding posts that contained them, but that would be harder to search for.

Obviously, discretion is requested - don't over-use the resource, and there are times when it's actually more useful for troubleshooting purposes to re-host a static copy and link that instead. I'd just ask people to use their intelligence, and think before posting, rather than try to impose a blanket ban.
ID: 1337782 · 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 1337785 - Posted: 13 Feb 2013, 16:46:31 UTC - in response to Message 1337782.  

I don't think we've ever been officially discouraged from posting live cricket links - certainly Eric himself has posted them in the past (see message 918297). I have a vague memory that we had a discussion on the subject last time somebody got a bee in their bonnet and started hiding posts that contained them, but that would be harder to search for.

Obviously, discretion is requested - don't over-use the resource, and there are times when it's actually more useful for troubleshooting purposes to re-host a static copy and link that instead. I'd just ask people to use their intelligence, and think before posting, rather than try to impose a blanket ban.

I really don't recall anymore how the conversation came up or who brought it up.
Point being, we would not like to overburden a resource that we are being allowed to access by unofficial courtesy.

And it really is not necessary, the simple link allows anybody to see it when desired, without loading the graph from the server anytime somebody simply looks at the thread. Posting a static copy of course, would not do that either.
"Freedom is just Chaos, with better lighting." Alan Dean Foster

ID: 1337785 · Report as offensive
Profile James Sotherden
Avatar

Send message
Joined: 16 May 99
Posts: 10436
Credit: 110,373,059
RAC: 54
United States
Message 1337802 - Posted: 13 Feb 2013, 17:21:23 UTC

I would agree to use discretion. Nobody is on the look out to start hiding posts for live links to the cricket graphs.

I had just starting using the boards when it was told to us stop using live links.

Eric as the project scientist, and all the other titles he holds can post what ever live links he desires. You wont see me hiding a post of his:)
[/quote]

Old James
ID: 1337802 · 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 1337806 - Posted: 13 Feb 2013, 17:27:28 UTC - in response to Message 1337802.  

I would agree to use discretion. Nobody is on the look out to start hiding posts for live links to the cricket graphs.

I had just starting using the boards when it was told to us stop using live links.

Eric as the project scientist, and all the other titles he holds can post what ever live links he desires. You wont see me hiding a post of his:)

LOL...indeed.
And you will notice that the post in question is almost 4 years old.
And, as Richard indicated, a static link would still show the point Eric was making at that time. The live display shown now is totally irrelevant to what he was posting about then.
"Freedom is just Chaos, with better lighting." Alan Dean Foster

ID: 1337806 · Report as offensive
Richard Haselgrove Project Donor
Volunteer tester

Send message
Joined: 4 Jul 99
Posts: 14644
Credit: 200,643,578
RAC: 874
United Kingdom
Message 1337940 - Posted: 13 Feb 2013, 23:54:57 UTC

The crickets are back.
ID: 1337940 · Report as offensive
.clair.

Send message
Joined: 4 Nov 04
Posts: 1300
Credit: 55,390,408
RAC: 69
United Kingdom
Message 1337945 - Posted: 14 Feb 2013, 0:13:55 UTC

Looks like someone gave them a 24 hour pass
they deserve a day off every now and then
they have to put up with a lot of bits from all sorts of people.
ID: 1337945 · Report as offensive
Previous · 1 . . . 13 · 14 · 15 · 16 · 17 · 18 · 19 . . . 21 · Next

Message boards : Number crunching : Panic Mode On (81) 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.