No work again?

Message boards : Bug reports : No work again?

To post messages, you must log in.

AuthorMessage
Profile bigjohn1

Send message
Joined: 9 Jul 10
Posts: 3
Credit: 32,688
RAC: 0
Message 1706 - Posted: 11 Aug 2010, 0:08:20 UTC

My messages indicate no new work. Is the project down again? I searched the forum, and there is no indications that the project is down.
What's up? Could it be on my end?
ID: 1706 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
thinking_goose

Send message
Joined: 12 Nov 07
Posts: 116
Credit: 1,105,645
RAC: 0
Message 1707 - Posted: 11 Aug 2010, 0:55:26 UTC - in response to Message 1706.  

I'm getting the same message- it happens every so often. have you tried increasing the cache? I have about a days supply of work in the cache, so when the project goes down my computer still has work to do.
ID: 1707 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile Graeme of Boinc UK

Send message
Joined: 11 Oct 07
Posts: 29
Credit: 12,503,305
RAC: 0
Message 1708 - Posted: 11 Aug 2010, 16:23:16 UTC - in response to Message 1706.  

Back end servers are not running. See here...........

http://www.enigmaathome.net/server_status
ID: 1708 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile TJM
Project administrator
Project developer
Project scientist
Avatar

Send message
Joined: 25 Aug 07
Posts: 843
Credit: 69,304,327
RAC: 381,380
Message 1709 - Posted: 11 Aug 2010, 23:07:28 UTC

The server has serious hardware problems.
Today morning I thought that one of the SATA drives has died, because the kernel went into panic mode with tons of SATA errors on sdb1. I copied the data and removed the drive, the server was left running but disabled, just in case. I had to leave it like that because I had various other tasks to do.

When I came back home I thought that I'll setup the server to run without the faulty drive, but suddenly it got worse: it shows errors on random SATA drives and also crashes from time to time. I won't be able to do much right now. Tomorrow I'll try to clear everything off dust and then I'll start looking what's wrong.


M4 Project homepage
M4 Project wiki
ID: 1709 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile TJM
Project administrator
Project developer
Project scientist
Avatar

Send message
Joined: 25 Aug 07
Posts: 843
Credit: 69,304,327
RAC: 381,380
Message 1710 - Posted: 13 Aug 2010, 21:47:22 UTC

The old motherboard died, right now the server runs on a backup machine which has been sitting in the corner for occassions like this one. It should be fast enough to host the project. I hope that it won't fail anytime soon, because the backup of backup machine won't be able to run anything.
During the next few days everything shoult slowly go back to normal.

M4 Project homepage
M4 Project wiki
ID: 1710 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile Graeme of Boinc UK

Send message
Joined: 11 Oct 07
Posts: 29
Credit: 12,503,305
RAC: 0
Message 1711 - Posted: 14 Aug 2010, 7:32:20 UTC

Thank you for the update.

Regards,
Graeme.
ID: 1711 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile Ageless
Volunteer moderator
Volunteer tester
Avatar

Send message
Joined: 11 Sep 07
Posts: 104
Credit: 155,932
RAC: 0
Message 1712 - Posted: 14 Aug 2010, 12:20:30 UTC - in response to Message 1710.  
Last modified: 14 Aug 2010, 12:20:45 UTC

Time for (a) (another) donations round then? :)
Jord.

BOINC FAQ Service.
ID: 1712 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile Graeme of Boinc UK

Send message
Joined: 11 Oct 07
Posts: 29
Credit: 12,503,305
RAC: 0
Message 1713 - Posted: 14 Aug 2010, 17:50:48 UTC

Is this the motherboard from Debian 5 ?
Just wondering if that is still under warranty.
ID: 1713 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
fitz

Send message
Joined: 15 Apr 09
Posts: 31
Credit: 147,954
RAC: 0
Message 1714 - Posted: 14 Aug 2010, 22:29:55 UTC

whats the hardware specs of the server affected, what sort of mobo would it need? i'd be happy to help if i can?!
ID: 1714 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile TJM
Project administrator
Project developer
Project scientist
Avatar

Send message
Joined: 25 Aug 07
Posts: 843
Credit: 69,304,327
RAC: 381,380
Message 1715 - Posted: 15 Aug 2010, 23:37:33 UTC

The old server was a Phenom 9850 running on a cheap Gigabyte motherboard.
The 'new' one is an Athlon 3500+ on a decent s939 ASUS motherboard (I forgot the model name) , which has only two disadvantages: only 4 SATA ports (5th, the nvidia's RAID doesnt work for some reason) and DDR memory slots - this allows only 2GB (4x512) of RAM, because 1GB DDRs are way too expensive.



M4 Project homepage
M4 Project wiki
ID: 1715 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile Ageless
Volunteer moderator
Volunteer tester
Avatar

Send message
Joined: 11 Sep 07
Posts: 104
Credit: 155,932
RAC: 0
Message 1716 - Posted: 16 Aug 2010, 0:14:51 UTC - in response to Message 1715.  

ID: 1716 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
noderaser
Avatar

Send message
Joined: 24 Dec 08
Posts: 88
Credit: 629,026
RAC: 0
Message 1717 - Posted: 16 Aug 2010, 2:17:47 UTC

Likely too expensive in terms of a volunteer project.

ID: 1717 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile TJM
Project administrator
Project developer
Project scientist
Avatar

Send message
Joined: 25 Aug 07
Posts: 843
Credit: 69,304,327
RAC: 381,380
Message 1718 - Posted: 16 Aug 2010, 7:52:35 UTC

The definition: I believe that I would be able to find mobo with faster processor and 4GB DDR2 cheaper than 4x1GB DDRs.
M4 Project homepage
M4 Project wiki
ID: 1718 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
fitz

Send message
Joined: 15 Apr 09
Posts: 31
Credit: 147,954
RAC: 0
Message 1719 - Posted: 16 Aug 2010, 19:46:39 UTC

Ok so if understand it correctly...the best solution would be an AM2+ board, which would allow you to re-use the DDR2 and Processor from the failed board?...is the raid merely there for redundancy or is there a large I/O demand on the server? Would on board RAID suffice or would a controller be Ideal.?!

Presumably the easiest solution (from my point of view of wanting to help out), would be to provide a donation, rather than source a board + post?! (from UK)

Thanks for all the effort you put in!
ID: 1719 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
quel

Send message
Joined: 19 May 09
Posts: 34
Credit: 32,879,565
RAC: 1,838
Message 1721 - Posted: 18 Aug 2010, 16:49:31 UTC - in response to Message 1719.  

fitz,

I agree I'd like to donate money to the project and have asked for information on how to do so in the past. Not looking for any tax-deductible charity or anything of the sort just a gift to the project as a show of thanks and to help out in any way I can.
ID: 1721 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile TJM
Project administrator
Project developer
Project scientist
Avatar

Send message
Joined: 25 Aug 07
Posts: 843
Credit: 69,304,327
RAC: 381,380
Message 1726 - Posted: 1 Sep 2010, 20:08:35 UTC - in response to Message 1719.  
Last modified: 1 Sep 2010, 20:09:41 UTC

Ok so if understand it correctly...the best solution would be an AM2+ board, which would allow you to re-use the DDR2 and Processor from the failed board?...


I think that the processor might also be fried, however I don't have any other board around here to test it. During the last hours of it's life, the server had symptoms typical to overheating CPU, except that it wasn't overheating. It's also possible that the board is fine and just the CPU is dead.


is the raid merely there for redundancy or is there a large I/O demand on the server? Would on board RAID suffice or would a controller be Ideal.?!


I didn't use RAID here and I don't plan to, mainly because JBOD "array" of disks with database tables, logs and work folders spread across physical disks seems to be faster and regularly backed up shouldn't be much less safe.

Right now the server itself is not a problem, the backup machine should do all the work just fine. The only thing that is really needed is UPS, the old one failed few months ago and since then sudden power failures already caused lots of problems.
M4 Project homepage
M4 Project wiki
ID: 1726 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
quel

Send message
Joined: 19 May 09
Posts: 34
Credit: 32,879,565
RAC: 1,838
Message 1735 - Posted: 4 Sep 2010, 3:34:42 UTC - in response to Message 1726.  

Ugh, I can feel for you on the UPS front. My 2nd APC Smart-UPS 1500 to die. Each lasted about 5 years total. "System controller board failure." Sad thing is this one I replaced the batteries earlier this year. I think I'm done with APC though all the brownouts and under-voltage conditions in the TX summer are harsh on a UPS. None of the many fold more expensive Lieberts at the office died though but I can't afford those or the special wiring :P

The excellent part of when the APCs die is that they can't even pass current so its death kills the power to the load. I experienced MySQL corruption that in over a decade of administration have never seen to that degree when this happened. (I was cheap on the raid card...it was already $500 so I didn't add $150 for a battery backup unit...which was fine until my UPS died.)
ID: 1735 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile TJM
Project administrator
Project developer
Project scientist
Avatar

Send message
Joined: 25 Aug 07
Posts: 843
Credit: 69,304,327
RAC: 381,380
Message 1736 - Posted: 4 Sep 2010, 10:05:42 UTC

I used a small 300VA UPS (Ever) modified for long running. I made two 80mm holes on the both sides and attached standard 80x80 fans with temperature controller attached to one of the heatsinks inside. Then I replaced it's original battery (I think it was 12V 6Ah) with much larger battery - 104Ah which of course didn't fit inside, so it was just standing next to UPS.
With proper cooling, large battery and a little scripting on the server to reduce average load when running on battery, it could run for 6-8 hours.
There was only one problem - internal charging circuit couldn't charge battery that large and soon after first tests it broke, so I had to use external charger.
M4 Project homepage
M4 Project wiki
ID: 1736 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote

Message boards : Bug reports : No work again?




Copyright © 2017 TJM