Thread 'News on Project Outages'

Message boards : Projects : News on Project Outages
Message board moderation

To post messages, you must log in.

Previous · 1 . . . 6 · 7 · 8 · 9

AuthorMessage
Jimmy G (BA)

Send message
Joined: 26 Sep 11
Posts: 41
Message 41289 - Posted: 23 Nov 2011, 19:11:36 UTC

As mentioned elsewhere on these boards, 96-hour Milkyway Server Status can be found here...

Milkyway@Home Server Graphs:
http://www.bluenorthernsoftware.com/scarecrow/mwstats/graphs.php?t=96
ID: 41289 · Report as offensive
BarryAZ

Send message
Joined: 4 Sep 09
Posts: 381
United States
Message 41290 - Posted: 23 Nov 2011, 19:17:01 UTC - in response to Message 41289.  

That's interesting -- though at this point it simply confirms what is known. MW is offline, and has been offline for 30 hours or more. Given the holiday weekend, and that it is past 2PM back east, I suspect this outage might extend for another 120 hours.

Which is to say, if you are doing MW GPU's, you'd best have (have had already) an alternate GPU project running.

As mentioned elsewhere on these boards, 96-hour Milkyway Server Status can be found here...

Milkyway@Home Server Graphs:
http://www.bluenorthernsoftware.com/scarecrow/mwstats/graphs.php?t=96

ID: 41290 · Report as offensive
ProfileBlurf

Send message
Joined: 18 Jul 11
Posts: 217
United States
Message 41292 - Posted: 23 Nov 2011, 21:55:43 UTC

I am still trying to get an update. I've been locked up with work stuff all day and am just getting to it now.
ID: 41292 · Report as offensive
ProfileBlurf

Send message
Joined: 18 Jul 11
Posts: 217
United States
Message 41293 - Posted: 23 Nov 2011, 22:09:58 UTC

MW Update:

Our support staff has been looking into it since Tuesday. We are currently able to log onto the machine, but it is not displaying the website.
ID: 41293 · Report as offensive
BarryAZ

Send message
Joined: 4 Sep 09
Posts: 381
United States
Message 41294 - Posted: 23 Nov 2011, 22:54:02 UTC - in response to Message 41293.  

Thanks for the update -- it sounds like something pretty major then which will require an onsite look/see.


MW Update:

Our support staff has been looking into it since Tuesday. We are currently able to log onto the machine, but it is not displaying the website.

ID: 41294 · Report as offensive
KAMasud

Send message
Joined: 13 Feb 07
Posts: 21
Pakistan
Message 41295 - Posted: 24 Nov 2011, 2:29:55 UTC - in response to Message 41294.  

An on site look see sounds like some very hard exertion, very bad for health. As to finding out the status, whats the use. The completed WU's are uploading into some blackhole about which and where, i dont know a thing.
ID: 41295 · Report as offensive
BarryAZ

Send message
Joined: 4 Sep 09
Posts: 381
United States
Message 41297 - Posted: 24 Nov 2011, 3:30:13 UTC - in response to Message 41295.  

The problem at this point being the long holiday weekend. Since this problem started early on Tuesday, one would have thought that two working days would be enough to figure out and solve the problem. Unfortunately, that turns out not to be the case for the MW folks and thus, perhaps sometime on Monday things will be dealt with.... or not.


An on site look see sounds like some very hard exertion, very bad for health. As to finding out the status, whats the use. The completed WU's are uploading into some blackhole about which and where, i dont know a thing.

ID: 41297 · Report as offensive
ProfileCarlos R. Moreira

Send message
Joined: 23 Nov 11
Posts: 2
Brazil
Message 41304 - Posted: 25 Nov 2011, 10:23:12 UTC
Last modified: 25 Nov 2011, 10:31:05 UTC

Separation Status and New Runs
The separation runs that analyse the northern galactic cap (Sloan Digital Sky Survey, or SDSS Stripes 9-27) have nearly finished - soon we'll be able to wrap all of that data together, and with data from Nathan Cole's PhD thesis, we'll have a nice scientific paper out in the next few months. We'll keep you notified.

That doesn't mean that the separation code will be retired - far from it. We're starting new separation runs (the ones with "mix" in the name) that are testing the robustness of the separation code. We'll be running simulated data sets along side real data to test several aspects of the stream-fitting process. The main questions are: What would Milkyway@home do if something different from our model exists in the data? And how different would that something have to be in order to modify our results?

Also, the release of SDSS Data Release 8 earlier this year gives us access to several stripes of data in the southern galactic cap. We are currently processing this data (It's not as continuous as the northern data, so we have to cut out areas with spotty data), and it will eventually run on Milkway@home.

We are also looking to improve the code on Milkyway@home. My brand new paper studies the distribution function of stars in the Milky Way halo, and provides a new convolution kernel that should make Milkyway@home more effective. There are a few other results from that paper, and inside work, that we would like to implement in the Milkyway@home searches. When we update these functions, we will run stripes over again and look for differences.

Long story short: We're almost done with one part of Milkyway@home's mission, but there's still plenty to do. Happy Crunching!

Cheers - Matthew N



I tought they where just wrapping the data & upgrading the code, and hopefully in couple days we would have more work to do again ^^.


Best Regards,
Carlos R. Moreira
ID: 41304 · Report as offensive
KAMasud

Send message
Joined: 13 Feb 07
Posts: 21
Pakistan
Message 41309 - Posted: 26 Nov 2011, 3:28:06 UTC


They wrap the data and upgrade the code quite frequently. So far i have noted a cycle of ten to fifteen days. Ask BarryAZ, he is the resident expert, i only joined the project a month back with three breakdowns so far.
Cheers and regards.
ID: 41309 · Report as offensive
BarryAZ

Send message
Joined: 4 Sep 09
Posts: 381
United States
Message 41310 - Posted: 26 Nov 2011, 3:59:27 UTC - in response to Message 41309.  

Actually, Blurf has closer contact with the project than I do. Though I suspect I have participated longer than he has (I joined it over three years ago).

That being said, I suspect this particular outage is of a different nature -- either connectivity to the net, or something at their server.

I recall that they mentioned plans for hardware upgrades -- but I think that needs funds first.



They wrap the data and upgrade the code quite frequently. So far i have noted a cycle of ten to fifteen days. Ask BarryAZ, he is the resident expert, i only joined the project a month back with three breakdowns so far.
Cheers and regards.

ID: 41310 · Report as offensive
HassanShebli

Send message
Joined: 5 Oct 10
Posts: 3
Message 41315 - Posted: 26 Nov 2011, 14:48:49 UTC - in response to Message 41310.  

It is not matter of fund as they said frequently. Many users there were ready to donate the needed hardware. MW complaint was lack of supervision staff as far as my knowledge . And I don't really know why the don't have a one or more backup server?!!!

It so frustrating That they didn't add another server. They have a minimum of a monthly breakdown since a joined them a year ago. At least they should keep the website on a separate server to let us know status of the project!

I really wish that other real scientific project have AMD support so we can switch to. I am really considering selling my AMD 6970 card and getting Nvidia 580.

Good luck
ID: 41315 · Report as offensive
BarryAZ

Send message
Joined: 4 Sep 09
Posts: 381
United States
Message 41317 - Posted: 26 Nov 2011, 18:30:19 UTC - in response to Message 41315.  

Well, staffing is funding -- and a level of funding that is not only not available from the user community, but also one that might bounce up against constraints from their host university (RPI).

As to having a second server -- I agree on that point. At the very least, MW probably would be well served to have their home page hosted on different hardware so that when it hits the fan, as it has this week, there is a place where updates could be posted. That being said, in the US this is a major holiday weekend and as I suggested on Wednesday, if MW was not back online Wednesday, I didn't expect to see anything at all until this coming Monday.

Like many, I was dismayed that they were not able to get things back to any degree at all during Tuesday and Wednesday.

For me, I caught the offline status early on with my server farm and suspended MW to redirect processing to MooWrapper and Collatz.

As to ATI GPU science projects -- in the coming month, POEM should have something available.


It is not matter of fund as they said frequently. Many users there were ready to donate the needed hardware. MW complaint was lack of supervision staff as far as my knowledge . And I don't really know why the don't have a one or more backup server?!!!

It so frustrating That they didn't add another server. They have a minimum of a monthly breakdown since a joined them a year ago. At least they should keep the website on a separate server to let us know status of the project!

I really wish that other real scientific project have AMD support so we can switch to. I am really considering selling my AMD 6970 card and getting Nvidia 580.

Good luck

ID: 41317 · Report as offensive
KAMasud

Send message
Joined: 13 Feb 07
Posts: 21
Pakistan
Message 41327 - Posted: 27 Nov 2011, 9:34:18 UTC


They have their captive/addicted loyalists and are not pushed. Take it or leave it is their attitude.
ID: 41327 · Report as offensive
HassanShebli

Send message
Joined: 5 Oct 10
Posts: 3
Message 41335 - Posted: 27 Nov 2011, 15:44:02 UTC - in response to Message 41317.  

With all my respect that's ridicules . If there is a well, there is a way. I am not convinced what so ever that there are some restrictions from RPI regarding this issue!

Users asked MW people frequently to present a clear paper of their requirements ( hardware, money, staff...etc ) so users can contribute as possible, but no response, just a shady speeches !!

Good to hear from you the news about POEM's new application for AMD. Finally we'll set ourselves free from MW. Hope more to come.


ID: 41335 · Report as offensive
ProfileBlurf

Send message
Joined: 18 Jul 11
Posts: 217
United States
Message 41337 - Posted: 27 Nov 2011, 17:08:24 UTC - in response to Message 41327.  
Last modified: 27 Nov 2011, 18:04:43 UTC

They have their captive/addicted loyalists and are not pushed. Take it or leave it is their attitude.


Kamasud and HassanShebli-your assumptions are incorrect. At Milkyway, we all work very hard. We have a requisition in to hire another person for server maintenance needs

Please remember this is a holiday weekend and Milkyway is located at a college campus. The campus shuts down for the holiday.

It is an issue that the labstaff (above our MW staff) are involved with. We are working on it furiously and I plan on getting an update as soon as possible. We apologize for any inconvenience.
ID: 41337 · Report as offensive
HassanShebli

Send message
Joined: 5 Oct 10
Posts: 3
Message 41350 - Posted: 27 Nov 2011, 23:23:09 UTC - in response to Message 41337.  
Last modified: 27 Nov 2011, 23:25:26 UTC

Dear Blurf:

Our frustration is caused by the continuous breakdowns of MW not the current situation.

I used to run 3 servers by myself for a relatively big company ( for accounting, data, and mail ) despite being pretty amature in server-related domain. And I used to go for a vacation for a complete month or so with no problem whatsoever !! that is because of something called ( BACKUP and REMOTE ACCESS) !!

Here are some advices I whish you consider:

*- Separate the website server form that data server so we can keep up and occasionally help.

*- Increase the WU quantity ( one or two weeks stack or even more ), and increase the size of the single WU so it take 1 or 2 hours to avoid the continuous connection from crunching machines every single second. My 6970 can finish a WU in 1.4 second approximately, and I donate 10 hours a day from my machine. That means 428 WUs per day for my PC, thus 428 connection request to the data server daily !!! So you do the math. It is consuming time ( connection problems and delay ) and money ( bandwidth, electricity, manpower ..etc ).

I know the second point were discussed many times previously, but no one could convince me and many other users why this is not yet done.

whish you the best
ID: 41350 · Report as offensive
ProfileJord
Volunteer tester
Help desk expert
Avatar

Send message
Joined: 29 Aug 05
Posts: 15569
Netherlands
Message 41351 - Posted: 27 Nov 2011, 23:31:15 UTC
Last modified: 27 Nov 2011, 23:32:03 UTC

Please people, may I request that you set up a thread outside this one for the discussion of what's best for Milkyway and go have your discussions there? This thread is here to allow people to report that projects are out and have come back, not to give advice on how projects should be run.
ID: 41351 · Report as offensive
KAMasud

Send message
Joined: 13 Feb 07
Posts: 21
Pakistan
Message 41357 - Posted: 28 Nov 2011, 4:42:02 UTC


My vote is with Hassan, we have to keep our connection open just for WU's with malware, spyware and virus attacks.
Ageless, this problem is not just with Milkyway but with other worthwhile projects also, CPDN for one. It seems that the projects cannot or do not want to keep up with new Tech. They are just sucking their old machines with resultent breakdowns. As to going some where else, where? That project is a blank. Dont worry, when they come back online then we will take it up with them.
Regards.
ID: 41357 · Report as offensive
KAMasud

Send message
Joined: 13 Feb 07
Posts: 21
Pakistan
Message 41358 - Posted: 28 Nov 2011, 5:44:45 UTC

Homeless orphans describes us best. Ageless is correct, lets make our own Milkyway thread on Boinc. Where we can take it apart in peace.
ID: 41358 · Report as offensive
Previous · 1 . . . 6 · 7 · 8 · 9

Message boards : Projects : News on Project Outages

Copyright © 2024 University of California.
Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.2 or any later version published by the Free Software Foundation.