This machine

http://setiathome.berkeley.edu/show_host_detail.php?hostid=1154305

was my top producer. Now its RAC is below anything else I own. Why? I can guarantee the machine is running as it runs my 13 diskless machines farm. But its RAC seems to have been at best zeroed?

I think the original host record disappeared, as I had to merge two "ghost" machines to make this one. Database error?

---

This machine

http://setiathome.berkeley.edu/show_host_detail.php?hostid=1154305

was my top producer. Now its RAC is below anything else I own. Why? I can guarantee the machine is running as it runs my 13 diskless machines farm. But its RAC seems to have been at best zeroed?

Hello!

I think by merging you keep RAC only if new machine has not returned any WU.
This machine was created "7 Jul 2005 20:01:34 UTC", but you merged and so
it shows all WUs crunched before.

Wait two to three weeks.

BR
Iztok

You had many Computation Errors lately on that one. Maybe there is a hw-problem with this mashine.

It gets:
"Maximum disk usage exceeded"-errors. Maybe you should alter your disk settings for this one.

---

_\|/_
U r s

Thanks guys but look again. This machine has crunched 60000 cobblestones, it is not a new addition....

---

It gets:
"Maximum disk usage exceeded"-errors. Maybe you should alter your disk settings for this one.

Preferences:

Use no more than 100 GB disk space
Leave at least 1 GB disk space free
Use no more than 75% of total disk space


Ouput from "df -H"

/dev/hda2 79G 6.7G 68G 9%

In other words, I have 79Gb space, 6.7Gb used (which includes 13 other BOINC installations), 68GB free, 9% used.

So although BOINC errors with "Max disk space exceeded", how can that be? And that STILL doesn't explain why a machine running 365/7/24 has an RAC of 3 and a total of 63k+ cobblestones?

---

Thanks guys but look again. This machine has crunched 60000 cobblestones, it is not a new addition....

Already said:
- created today
- rest is merged
- problems reported to Berkeley:
Can't acquire lockfile - exiting
or
process exited with code 251 (0xfb)


SETI@home error -5 Can't open file
(work_unit.sah) in read_wu_state() errno=2

Something wrong with disks, mounting? Maybe you try to run two copies?
Maybe you try to run is as different user (not owner of temporary files)?
File permissions?
Have you merged yourself or it was done automagically?

BR
Iztok

The Wiki has some stuff on the lockfile problem.

BOINC Wiki

---

It also has some stuff on "process_exited_with_code 251".



<a> BOINC Wiki[/url]

---

Ok I'll make this real simple.

I know this machine has issues, I think it is memory but I am still tracking it down. That isn't my point, it has crunched 60k+ CS successfully so it isn't really that bad. Plus I have 13 machines which rely on this one to function and they are still running and producing results. So the machine runs.

This machine has a little brother, it shows an RAC of 206.75, hostID 898337
This machine's host record has disappeared completely, RAC should be 220+
I manually merged two ghost copies earlier today, all live WU's are there, but RAC is 3+ not 220+

Where has my machine's host record gone?

---

Ok I'll make this real simple.

I know this machine has issues, I think it is memory but I am still tracking it down. That isn't my point, it has crunched 60k+ CS successfully so it isn't really that bad. Plus I have 13 machines which rely on this one to function and they are still running and producing results. So the machine runs.

This machine has a little brother, it shows an RAC of 206.75, hostID 898337
This machine's host record has disappeared completely, RAC should be 220+
I manually merged two ghost copies earlier today, all live WU's are there, but RAC is 3+ not 220+

Where has my machine's host record gone?

When a machine stops producing good work, the RAC will drop. This is to be expected.

---

BOINC WIKI

If I click on the link you posted;

http://setiathome.berkeley.edu/show_host_detail.php?hostid=1154305

Then click on the number next to the results at the bottom.

Then if I click on the numbers all the way to the left I get linked to a page that shows a result.

http://setiathome.berkeley.edu/result.php?resultid=81953270

Maybe if you look through them you can figure out when and why it started coming up with invalid work units being reported.

---