Incorrect. Not confusing at all. :-)

Café
Number Crunching
Science (non-Seti)
Politics

I find this one useful though, especially as you won't get me into the Cafe, not even if it was the last forum on Earth, or I was paid a million credit per viewing.

A great BIG +1.

I don't read anything there, but I do go just for nostalgia. Many, many years ago I lived in the Cafe with a bunch of friends that are no longer around. Now I spend all my time here in Number Crunching. Much more interesting reads here. :)

Have a great day! :)

Siran

---

CAPT Siran d'Vel'nahr - L L & P _\\//
Winders 11 OS? "What a piece of junk!" - L. Skywalker
"Logic is the cement of our civilization with which we ascend from chaos using reason as our guide." - T'Plana-hath

Hi all, hope It's OK to ask this here. I haven't been crunching for a few years. Some personal and real life issues got in the way of my hobbies, but I'm trying to set that straight now :-)

As Murphy's Law would have it, my main computer fried the video card and corrupted the hard drive just about the time I was getting ready to get it crunching again. Had to reinstall everything from scratch and the only BOINC backups I have are from June 2013. Is there any way I can salvage the computer ID from there, or do I have to start over?

Thanks.

Your host ID is in the client_state.xml file in your BOINC backup. If you can't just restart your BOINC backup then you will have to reinstall BOINC. What that would normally cause is the generation of a new host ID because of likely computer hardware or OS changes and possibly a new login ID and or a new email address and the fact that the BOINC servers have never seen that computer before.

If you know your previous login name and password it will be much easier to get back your old account. Also you can edit the client_state.xml file that will be created in a new host and replace the host ID with our old one. You also need to edit the rpc_seqno in the client_state.xml file to be one greater than the number of times your old host contacted the servers. If you don't know that number you can just put a value like 10000 in its place if you think you never contacted the servers in the past that many times. That should get back your old host ID and history.

---

Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)

Thank you for that piece of knowledge. I have another old system that successfully reconnected so I have all my account info and can look around at the file structure. I'll have time to play around with it in the next couple of days and let you know how it turns out.

My preliminary plan is to install BOINC and copy the client_state.xml over before I connect to the project. I'll also edit out all the file data for the work units that were pending when that backup was done and increase the rpc_seqno accordingly before connecting. If you have any comments or suggestions all feedback would be appreciated.

If you can just copy over the old client_state with all the work unit information removed at the end of the file and increase the rpc_seqno to any value higher than what the servers reported for the "number of times client has connected server" shown on the website for the old 2013 host, you will just pick up where you left off with new work.

Use the Site Search in the Seti main page menu and copy your host ID from the old client_state file into the search field. First hit should be the old 2013 host Details page where the number of times the client contacted the server is shown. Increment the rpc_seqno by one in the client_state file and save it. You should be ready to go at that point

---

Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)

Thank you! I'm at work right now, but you're getting me all motivated to try it when I get home tonight :-)

Well, I got the last gpu upgraded in my dual cpu e5-2590v2. Lets stop at 3 Gtx 1070Ti's :) This MB won't run 4 gpus.
The gpus are sitting as far a part as I can manage it. The only other thing I could do to cool them down further would be throw the fan directly on them instead of blowing through the front of the chassis through another fan. They all seem to be pegged at 80+ C and seem to be running at "full tilt". At least under BL14 they were doing 90 seconds or so at 80+ C.

Maybe I can crack 180,000 RAC and stay there :)

Tom

---

A proud member of the OFA (Old Farts Association).

If you can just copy over the old client_state with all the work unit information removed <snip>

I followed your instructions and it worked like a charm. The old system that hadn't seen SETI for a couple of years is now back home and crunching away. I think I'll let the stock apps run for a couple of weeks then start looking at getting optimized apps back on there. Thanks again for all the help.

Glad it was fairly simple. Sometimes a blown up computer is a total rebuild from scratch. The stock gpu apps are the same as the Lunatics currently. The Lunatics cpu apps are still faster than the stock cpu apps for the most part. Since the gpu produce more output than the cpu, no hurries. At least for MB. The Lunatics tuning on the AP apps is much faster than stock. A lot more improvement possible there especially with the AstroPulse WU party we've been having for the past month.

---

Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)

Just curious: a work unit twice replicated is the norm, thrice more rarely, but why has this WU been replicated quadrice ? Will it be replicated quince, sice or sevence until one of the many client varieties can validate it conclusively ?

Screen capture: https://i.imgur.com/HOD9vME.jpg

---

Apr 3, 1999 - May 3, 2020

Will it be replicated quince, sice or sevence until one of the many client varieties can validate it conclusively ?

Up to 10 times.
It's problem has been the Applications that have been processing it. The Intel iGPU (or it's drivers) does tend to produce poor results, as do some applications on some Apple hardware/ OS combinations.

---

Grant
Darwin NT

Just curious: a work unit twice replicated is the norm, thrice more rarely, but why has this WU been replicated quadrice ? Will it be replicated quince, sice or sevence until one of the many client varieties can validate it conclusively ?

Screen capture: https://i.imgur.com/HOD9vME.jpg

This happens much more often than you think and this could go on for up to 8 resends (10 tasks in total) until someone comes along with a result that concurs with 1 or more of the previously returned results. ATM all 3 returned results are different.

Cheers.

50-50% chance it will go to another host since the current host is a 1060 running the old CUDA50 app.

---

Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)

Actually much more likely to be sent out again due the current "deciding" host timing out - it hasn't contacted the servers since 4th Jan and has an incredibly low RAC.

---

Bob Smith
Member of Seti PIPPS (Pluto is a Planet Protest Society)
Somewhere in the (un)known Universe?

Actually much more likely to be sent out again due the current "deciding" host timing out - it hasn't contacted the servers since 4th Jan and has an incredibly low RAC.

And this is the only WU downloaded by the only computer registered by this new (Nov. 2018) user...

Either it is a low priority WU (*), either there is a Pity module (*) hidden in the scheduling entrails' server (which would also explain why I get any WUs at all when there are so many other more powerful and efficient crunchers around, some of them who've had the curtesy of answering my original question here - which I thank for their informative inputs =:)
___
* No, I don't really believe that; who knows which WU migh contain part of the next Wow signal ? =:)

---

Apr 3, 1999 - May 3, 2020

What time does the weekly outage start in UTC time ?

When the right person gets into the lab - anytime between 13:00 and 19:00 UTC.....

---

Bob Smith
Member of Seti PIPPS (Pluto is a Planet Protest Society)
Somewhere in the (un)known Universe?

When the right person gets into the lab - anytime between 13:00 and 19:00 UTC.....

Thanks, I noticed that last week it was around 16:20 UTC, but I thought it used to be earlier when I was a regular here.

Anyone know what board manufacturers (ASUS in particular) charge for replacing ATX Connectors?
I have one with a burnt 24pin, and other with burnt 8pin connector. Both backsides look fine.

ASUS just said, We don't have a fixed rate, so send them in and we will tell you later.

---