Apologies to wingmen: Hard Drive Failure all work lost

Message boards : Number crunching : Apologies to wingmen: Hard Drive Failure all work lost
Message board moderation

To post messages, you must log in.

AuthorMessage
Profile Len
Avatar

Send message
Joined: 15 Mar 10
Posts: 52
Credit: 11,725,173
RAC: 86
United Kingdom
Message 1229960 - Posted: 10 May 2012, 11:23:26 UTC

My apologies to all those waiting on my results.
On the plus side it will now be working marginally faster once I have it fully up to speed again.

This Cruncher is the one affected, so all these tasks will time out. Sorry, but the damned drive just totaled on me but I decided to go with Windoze7 64bit as the replacement OS, so it should run that bit faster than the old 32bit XP OS.
I think I am. Therefore I am. I think.
ID: 1229960 · Report as offensive
Profile red-ray
Avatar

Send message
Joined: 24 Jun 99
Posts: 308
Credit: 9,029,848
RAC: 0
United Kingdom
Message 1229963 - Posted: 10 May 2012, 11:35:12 UTC
Last modified: 10 May 2012, 11:35:36 UTC

As far as I know it's possible to get the new OS installation to have the same ID as the old one. If you did this then all the tasks would be resent to the new system. I don't know how to do it, but if you stop the new system getting new tasks by the time it's processed what you have I am sure someone will tell you how to do this.
ID: 1229963 · Report as offensive
Profile Len
Avatar

Send message
Joined: 15 Mar 10
Posts: 52
Credit: 11,725,173
RAC: 86
United Kingdom
Message 1229991 - Posted: 10 May 2012, 13:17:15 UTC

Thanks. I shall do that. Shame to waste the processing time, but Hey Ho. I might even garner a positive reply before it runs dry. It got a whole herd of work already.


I think I am. Therefore I am. I think.
ID: 1229991 · Report as offensive
Profile Fred E.
Volunteer tester

Send message
Joined: 22 Jul 99
Posts: 768
Credit: 24,140,697
RAC: 0
United States
Message 1229995 - Posted: 10 May 2012, 13:48:03 UTC

You can merge the two computer records (6203645 and 6647938). There's an option on your computer page to merge them by name. We don't see the computer names, so you may have to name it (or rename it) in Win 7 first. Don't know if that will get you resends, but it will fix cumulative credits, etc.
Another Fred
Support SETI@home when you search the Web with GoodSearch or shop online with GoodShop.
ID: 1229995 · Report as offensive
KneeDeep

Send message
Joined: 27 Sep 99
Posts: 131
Credit: 4,887,778
RAC: 0
United States
Message 1230014 - Posted: 10 May 2012, 14:38:42 UTC - in response to Message 1229995.  

You can merge the two computer records (6203645 and 6647938). There's an option on your computer page to merge them by name. We don't see the computer names, so you may have to name it (or rename it) in Win 7 first. Don't know if that will get you resends, but it will fix cumulative credits, etc.


Did so recently with one of my machines and all the WU's that were in progress when the old system failed were resent. Need to name the new system the same as the old system and do the merge.
ID: 1230014 · Report as offensive
Profile Len
Avatar

Send message
Joined: 15 Mar 10
Posts: 52
Credit: 11,725,173
RAC: 86
United Kingdom
Message 1230094 - Posted: 10 May 2012, 18:02:54 UTC

Well, I got them merged.
(I'd already tried, but Windoze had 'helpfully' capitalised the first letter in the new name, so it didn't work when I tried it before posting.)

Now to see if I get re-sends.
I think I am. Therefore I am. I think.
ID: 1230094 · Report as offensive
Claggy
Volunteer tester

Send message
Joined: 5 Jul 99
Posts: 4654
Credit: 47,537,079
RAC: 4
United Kingdom
Message 1230147 - Posted: 10 May 2012, 19:49:46 UTC - in response to Message 1229963.  

As far as I know it's possible to get the new OS installation to have the same ID as the old one. If you did this then all the tasks would be resent to the new system. I don't know how to do it, but if you stop the new system getting new tasks by the time it's processed what you have I am sure someone will tell you how to do this.


From the BoincFAQ: How to revert to an older HostID number?

You've nurtured your low hostID number for a couple of years and all of a sudden BOINC decides it's time you got a newer one. Is it possible to revert to the older number? Well yes, here's how.

1. Make sure you have no tasks in your queue (Finish & Upload, abort, report them all)
2. Go to the web site of the project, your list of computers and find your old Host ID number.
3. Stop and Close BOINC.
4. Edit client_state.xml, find the <hostid> tags and replace the number here with the old ID.
5. Find the <rpc_seqno> tag and increment the number here with one or two.
6. Save client_state.xml
7. Restart BOINC.

Now whatever you do, do not merge the hostIDs on the web site, or else you'll be stuck with the higher ID number. Just wait for all the tasks in the new ID's number to time out, then delete the ID when it has no tasks listed anymore.


Claggy
ID: 1230147 · Report as offensive
Richard Haselgrove Project Donor
Volunteer tester

Send message
Joined: 4 Jul 99
Posts: 14650
Credit: 200,643,578
RAC: 874
United Kingdom
Message 1230197 - Posted: 10 May 2012, 21:01:54 UTC - in response to Message 1230147.  

As far as I know it's possible to get the new OS installation to have the same ID as the old one. If you did this then all the tasks would be resent to the new system. I don't know how to do it, but if you stop the new system getting new tasks by the time it's processed what you have I am sure someone will tell you how to do this.


From the BoincFAQ: How to revert to an older HostID number?

You've nurtured your low hostID number for a couple of years and all of a sudden BOINC decides it's time you got a newer one. Is it possible to revert to the older number? Well yes, here's how.

1. Make sure you have no tasks in your queue (Finish & Upload, abort, report them all)
2. Go to the web site of the project, your list of computers and find your old Host ID number.
3. Stop and Close BOINC.
4. Edit client_state.xml, find the <hostid> tags and replace the number here with the old ID.
5. Find the <rpc_seqno> tag and increment the number here with one or two.
6. Save client_state.xml
7. Restart BOINC.

Now whatever you do, do not merge the hostIDs on the web site, or else you'll be stuck with the higher ID number. Just wait for all the tasks in the new ID's number to time out, then delete the ID when it has no tasks listed anymore.


Claggy

Those sound like the instructions for when a working machine is spontaneously allocated a new HostID. Len has the slightly different task, of getting a new machine to mimic an old machine - so he'll have to adapt the instructions somewhat.

At step (2), on the project website, open the computer details, and:
2a. Note the "Number of times client has contacted server"

At step (5), set the <rpc_seqno> tag to a number one or two above the 'number of times client has contacted server' noted above at step 2a.
ID: 1230197 · Report as offensive
tbret
Volunteer tester
Avatar

Send message
Joined: 28 May 99
Posts: 3380
Credit: 296,162,071
RAC: 40
United States
Message 1232358 - Posted: 15 May 2012, 6:04:28 UTC

Now I get to apologize, too.

My newest combination of things, the Phenom II 965 and AMD 6770, has died an ugly death.

I walked in this afternoon to find it running, but no video. A reboot brought no video. Exchanging video cards brought no video. Pulling everything but the video card - no video.

I had to reach behind the radiator to check the motherboard 12v connector and "crunch," the PSU plug crumbled into my hand.

I pulled everything apart to examine it. The connector had melted into the socket, then become brittle. The power supply wire leading into the plug has melted a little and flowed just a little.

The motherboard socket is irrecoverable. There's melted plug all down in the socket.

It did occur to me that I could unsolder the socket, clean it out, and re-install it, but it also occurred to me that *something* was pulling way, way too much juice and I really don't want a fire.

Since the CPU (although new) is on a new Antec Kuhler and since when I checked the CPU was running very cool whenever I checked it, I suppose I could have a bad CPU. I also suppose the motherboard could be the culprit, or for that matter the 6770 could have "gone south" and taken everything with it.

Of course, the power supply is / was a pretty new Corsair GS700, the video card is an almost new 6770, the CPU cooler is new, two sticks of RAM are new to this computer, the CPU is about a week old... it could be *anything*. But the MSI motherboard is a couple of years old, at least. It's been a champ.

There are no obvious problems like bulging capacitors or burned spots on the board and the 6770's contacts are bright and shiny all-around.

...oh, who knows?

But I wish I could diagnose it rather than rebuilding it and killing a new motherboard if that wasn't the problem.
ID: 1232358 · Report as offensive
Profile skildude
Avatar

Send message
Joined: 4 Oct 00
Posts: 9541
Credit: 50,759,529
RAC: 60
Yemen
Message 1232496 - Posted: 15 May 2012, 14:23:40 UTC - in response to Message 1232358.  
Last modified: 15 May 2012, 14:25:09 UTC

HEHEHe I got lucky this weekend. FOr whatever reason I had set BOINC to NNT at some point last week. So my cache was fairly small.
Mothersday was a lot of work and when I came home I found my Main machine with a BSOD. A restart gave me a chkdsk... dundundunnnnn I let the checkdisk run and when I woke on Monday it was still running. Uh Oh, What the heck let it run until I get home from work.

I get home and its piling up lost files into folders. section 3 and only 36% done. I have to assume my HDD is crap. I shut it down and installed WIndows on a new HDD. I was able to save my BOINC folders. Most likely I'll be able to save most of my files since it appears that most of the corrupted files were the OS itself. Here's to a week of moving and reinstalling everything


In a rich man's house there is no place to spit but his face.
Diogenes Of Sinope
ID: 1232496 · Report as offensive

Message boards : Number crunching : Apologies to wingmen: Hard Drive Failure all work lost


 
©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.