Slow network connection to Green Bank

Message boards : News : Slow network connection to Green Bank
Message board moderation

To post messages, you must log in.

Previous · 1 · 2 · 3 · 4 · 5 . . . 14 · Next

AuthorMessage
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 1915992 - Posted: 31 Jan 2018, 4:27:15 UTC - in response to Message 1915966.  

Matt was able to get us 1.6 TB of data (enough for a day and a half) by first transferring the data to Google storage, then to Berkeley. The problem appears to be somewhere in the internet2 route between Berkeley and Green Bank. We're pushing on campus networking to consider it a real problem (they keep asking if we've checked our cables, even though the problem exists on several machines).

More news tomorrow, we hope.

Problem might be upstream from campus and they don't fix that so they give the standard fix your cables.

I do remember one time where I was having issues reaching LHC@Home from my home machine but my work machine a dozen miles away but on a different ISP had no issues. A trace route showed a failed router at some network interchange point on the East Coast before the packets should have crossed the Atlantic, the other ISP used a totally different route. Didn't even have a clue who it belonged to so no way to pitch a bitch.
ID: 1915992 · Report as offensive
Profile Keith Myers Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 29 Apr 01
Posts: 13161
Credit: 1,160,866,277
RAC: 1,873
United States
Message 1915995 - Posted: 31 Jan 2018, 4:37:50 UTC - in response to Message 1915992.  

Well, the experienced campus network people should know their network topology and if there is an issue upstream should be able determine where the slowdown is happening. Then they have to raise a trouble ticket with that network tier level about the issue. Who knows how fast that link tech support will respond?

So anyone want to comment on my idea of just FedEx'ing some drives to Berkeley?
Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)
ID: 1915995 · Report as offensive
Grant (SSSF)
Volunteer tester

Send message
Joined: 19 Aug 99
Posts: 13720
Credit: 208,696,464
RAC: 304
Australia
Message 1915998 - Posted: 31 Jan 2018, 4:41:58 UTC - in response to Message 1915995.  

So anyone want to comment on my idea of just FedEx'ing some drives to Berkeley?

How to fund it?
Grant
Darwin NT
ID: 1915998 · Report as offensive
Profile Zalster Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 27 May 99
Posts: 5517
Credit: 528,817,460
RAC: 242
United States
Message 1915999 - Posted: 31 Jan 2018, 4:43:20 UTC - in response to Message 1915995.  
Last modified: 31 Jan 2018, 4:43:42 UTC

That assumes the tapes are removable (like Arecibo) , or if they have extras to take their place. They might just read and dump down the pipeline then rewrite. Not to mention the cost of FedEx as opposed to cableline.

edit..

Grant beat me to it..
ID: 1915999 · Report as offensive
Profile Keith Myers Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 29 Apr 01
Posts: 13161
Credit: 1,160,866,277
RAC: 1,873
United States
Message 1916003 - Posted: 31 Jan 2018, 4:50:47 UTC - in response to Message 1915998.  

I would think the department has $50 in petty cash for the delivery costs. The only thing that I can see that might be a hindrance, as Zalster comments, is that they don't have any spare drives to move the data onto for shipment.
Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)
ID: 1916003 · Report as offensive
Grant (SSSF)
Volunteer tester

Send message
Joined: 19 Aug 99
Posts: 13720
Credit: 208,696,464
RAC: 304
Australia
Message 1916004 - Posted: 31 Jan 2018, 4:55:34 UTC - in response to Message 1916003.  

I would think the department has $50 in petty cash for the delivery costs.

Don't know about the US, but here $50 would get you 2 boxes, 2 drives in each box, air freight. Maybe.
Grant
Darwin NT
ID: 1916004 · Report as offensive
Profile Keith Myers Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 29 Apr 01
Posts: 13161
Credit: 1,160,866,277
RAC: 1,873
United States
Message 1916007 - Posted: 31 Jan 2018, 5:23:13 UTC - in response to Message 1916004.  

OK, haven't shipped anything FedEx for a few years. Rates have gone up. 1 lb Fedex medium box (assuming for 1 drive) from Green Bank to Berkeley, Priority Overnight = $89.04. 2nd day AM = $42.88

So not too far off my estimate, at least for 2nd day delivery.
Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)
ID: 1916007 · Report as offensive
Eric Korpela Project Donor
Volunteer moderator
Project administrator
Project developer
Project scientist
Avatar

Send message
Joined: 3 Apr 99
Posts: 1382
Credit: 54,506,847
RAC: 60
United States
Message 1916012 - Posted: 31 Jan 2018, 6:01:38 UTC - in response to Message 1915972.  

I don't know in what format the Green Bank data is stored. If it is on physical drives, couldn't you just overnight them via FedEx? Assuming you have the correct hardware on your end to mount them and get the data into the splitters, wouldn't that keep us going for a few more days until campus networking gets off their Level One script reader response?


Unfortunately Breakthrough needs the drives for daily observations, so we can't pull and ship the way we did with Arecibo. (That and they are in a fairly large raid configuration, so we'd probably need to ship all of them).
@SETIEric@qoto.org (Mastodon)

ID: 1916012 · Report as offensive
Grant (SSSF)
Volunteer tester

Send message
Joined: 19 Aug 99
Posts: 13720
Credit: 208,696,464
RAC: 304
Australia
Message 1916013 - Posted: 31 Jan 2018, 6:02:14 UTC - in response to Message 1916007.  

Rates have gone up. 1 lb Fedex medium box (assuming for 1 drive) from Green Bank to Berkeley, Priority Overnight = $89.04. 2nd day AM = $42.88

*sound of jaw hitting ground*
Grant
Darwin NT
ID: 1916013 · Report as offensive
Profile Zalster Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 27 May 99
Posts: 5517
Credit: 528,817,460
RAC: 242
United States
Message 1916014 - Posted: 31 Jan 2018, 6:15:34 UTC - in response to Message 1916012.  

Unfortunately Breakthrough needs the drives for daily observations, so we can't pull and ship the way we did with Arecibo. (That and they are in a fairly large raid configuration, so we'd probably need to ship all of them).


Thanks for that. I thought I had read that somewhere but didn't want to misspeak. We'll muddle through the best we can.
ID: 1916014 · Report as offensive
Profile Keith Myers Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 29 Apr 01
Posts: 13161
Credit: 1,160,866,277
RAC: 1,873
United States
Message 1916019 - Posted: 31 Jan 2018, 6:50:47 UTC - in response to Message 1916012.  

Thanks for chiming in on my idea Eric. I always like to know the technical particulars of any system I use.
Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)
ID: 1916019 · Report as offensive
Profile Keith Myers Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 29 Apr 01
Posts: 13161
Credit: 1,160,866,277
RAC: 1,873
United States
Message 1916020 - Posted: 31 Jan 2018, 6:53:43 UTC

The other day we were told that the outage was longer than usual because they had to run down the hill to the CoLo to change drives and reboot kit. Wasn't the whole idea of moving the servers from the closet to the Data Centre because of better cooling, UPS facilities, faster network links, and on-site staff to hot-swap drives and re-boot kit?


I wondered about that post too. Why wasn't the staff at the colo able to just swap drives and reboot. Aren't we paying for that 24 hour staff resource? Why didn't that happen?
Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)
ID: 1916020 · Report as offensive
rob smith Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer moderator
Volunteer tester

Send message
Joined: 7 Mar 03
Posts: 22158
Credit: 416,307,556
RAC: 380
United Kingdom
Message 1916023 - Posted: 31 Jan 2018, 7:36:45 UTC

That would only work if the project had sufficient spare drives of the "right type" down at the coloc, otherwise someone has to "run down the hill" to re-stock the spares, remove the deads and do some general face to face stuff.
Bob Smith
Member of Seti PIPPS (Pluto is a Planet Protest Society)
Somewhere in the (un)known Universe?
ID: 1916023 · Report as offensive
Profile Stargate (SA)
Volunteer tester
Avatar

Send message
Joined: 4 Mar 10
Posts: 1854
Credit: 2,258,721
RAC: 0
Australia
Message 1916037 - Posted: 31 Jan 2018, 8:49:28 UTC

ID: 1916037 · Report as offensive
kevin o'rourke

Send message
Joined: 28 Apr 08
Posts: 1
Credit: 185,049
RAC: 0
United Kingdom
Message 1916038 - Posted: 31 Jan 2018, 9:05:05 UTC

hope it comes back on soon. I've two computers sitting idle doing nothing at the moment.
ID: 1916038 · Report as offensive
Profile Keith Myers Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 29 Apr 01
Posts: 13161
Credit: 1,160,866,277
RAC: 1,873
United States
Message 1916039 - Posted: 31 Jan 2018, 9:06:35 UTC

Yes, I remember that too. There were supposed to be ready spares on site at the colo. When you contract for server space and maintenance at server farms, you normally contract for the labor to maintain those server resources. That means ready spares, staff to replace drives and even have them ship the defective hardware off to the manufacturer for warranty replacement under either your shipping account or the contracted organizations shipping account.

That has been my experience in past jobs with robotic data library companies. So what services did we actually contract for at the colo?
Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)
ID: 1916039 · 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 1916041 - Posted: 31 Jan 2018, 9:11:13 UTC - in response to Message 1916039.  

There may be problems with third parties maintaining our rather eclectic collection of bespoke servers. IIRC, it matters which order they're rebooted so that all the interconnect mount points can be established. Our staff may prefer to do it themselves, knowing that we're watching over their shoulders for the slightest glitch.
ID: 1916041 · 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 1916047 - Posted: 31 Jan 2018, 10:35:48 UTC - in response to Message 1916046.  

It depends whether all our failing drives are in accessible hot-swap bays.
ID: 1916047 · Report as offensive
Profile Mr. Kevvy Crowdfunding Project Donor*Special Project $250 donor
Volunteer moderator
Volunteer tester
Avatar

Send message
Joined: 15 May 99
Posts: 3776
Credit: 1,114,826,392
RAC: 3,319
Canada
Message 1916055 - Posted: 31 Jan 2018, 11:30:04 UTC - in response to Message 1915995.  

So anyone want to comment on my idea of just FedEx'ing some drives to Berkeley?


Done and it only took -5 years...... I guess those drives got repurposed.

Also if there is a mysterious slowdown on a GBT <-> Berkeley data link, I'm wondering if it's being caused by the dreaded post-net-neutrality that was being worried about, depending on who's the link provider. Worth checking if everything else fails I guess.
ID: 1916055 · Report as offensive
Boiler Paul

Send message
Joined: 4 May 00
Posts: 232
Credit: 4,965,771
RAC: 64
United States
Message 1916062 - Posted: 31 Jan 2018, 13:25:46 UTC

Just got new work. Awesome
ID: 1916062 · Report as offensive
Previous · 1 · 2 · 3 · 4 · 5 . . . 14 · Next

Message boards : News : Slow network connection to Green Bank


 
©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.