Been running R@H now for over a fortnight and always had a full queue.
Today however, I'm running out of WU's on my two machines at home.
One of them has work for only 12 half a day left.
Anyone knows what the problem can be and perhaps has a solution ?

Connect to network about every
(determines size of work cache; maximum 10 days) 5 days

---

Any other projects attached? You may check Long Term Debt then...
Does "running out" mean you are empty of that the queue is close to empty (e.g some WUs are still left).

I'm running Rosetta on AMD X2 (second core in on CPDN SpinUp) and have Connect to network about every 0.1 days. Only one WUs sits in queue but Rosetta is so stable, that I never got problem immediatelly getting new one when it is finished with the ongiong one...

---

Any other projects attached? You may check Long Term Debt then...
Does "running out" mean you are empty of that the queue is close to empty (e.g some WUs are still left).

I'm running Rosetta on AMD X2 (second core in on CPDN SpinUp) and have Connect to network about every 0.1 days. Only one WUs sits in queue but Rosetta is so stable, that I never got problem immediatelly getting new one when it is finished with the ongiong one...

This machine is running one (30% done) and 4 left to crunch on.
After that the queue is empty.
My other machine at home has one running(80% done) and 3 left.
Before I always had enough for 2 days.
No other projects attached.

---

in general preferences ............have you changed this setting to less time ??

Network usage
Connect to network about every
(determines size of work cache; maximum 10 days) 2 days

---

46 years dc so far

join team FaDbeens
join us

in general preferences ............have you changed this setting to less time ??

Network usage
Connect to network about every
(determines size of work cache; maximum 10 days) 2 days

As I stated below it was 2 days.

Before I always had enough for 2 days.

---

in general preferences ............have you changed this setting to less time ??

Network usage
Connect to network about every
(determines size of work cache; maximum 10 days) 2 days

As I stated below it was 2 days.

Before I always had enough for 2 days.

there are seperate preferences for......... school - home - work

have you checked all of them ?

---

46 years dc so far

join team FaDbeens
join us

---

---

Checked it and he first red message is "Checksum of signature error for 1di2.pdb.gz" and than a lot of "unrecoverable errors for result 1di2_abrelax_rand_len10_jit02_omega_sim_XXXXX"
And this I've had for several times, but now it seems it's ok.
We'll see if it stays this way.
At least, my machines can keep crunching.

---

there was a serie of wu's that had a bad checksum due to some changes earlier this week, you propably saw some of those. the new ones are ok again, so you should not be too worried

---

--
Florian
www.domplatz1.de

there was a serie of wu's that had a bad checksum due to some changes earlier this week, you propably saw some of those. the new ones are ok again, so you should not be too worried

Yup. I had a flock of those guys show up on one of my systems here. Spooked me for a minute, till I remembered about this issue. It seems to have worked out now, there's an appropriate number of WU's in the queue, ready to rock and roll.

---

I've noticed that my linux box seems to run out of work. The log says:

2005-11-25 22:20:55 [rosetta@home] Sending scheduler request to https://boinc.bakerlab.org/rosetta_cgi/cgi
2005-11-25 22:20:55 [rosetta@home] Reason: To fetch work
2005-11-25 22:20:55 [rosetta@home] Requesting 8640 seconds of new work
2005-11-25 22:21:00 [rosetta@home] Scheduler request to https://boinc.bakerlab.org/rosetta_cgi/cgi succeeded
2005-11-25 22:21:00 [rosetta@home] Message from server: No work sent
2005-11-25 22:21:00 [rosetta@home] Message from server: No work sent
2005-11-25 22:21:00 [rosetta@home] Message from server: (there was work for other platforms)
2005-11-25 22:21:00 [rosetta@home] Message from server: (there was work for other platforms)
2005-11-25 22:21:00 [rosetta@home] No work from project
2005-11-25 22:21:00 [rosetta@home] No work from project
2005-11-25 22:28:34 [---] request_reschedule_cpus: project op

---

I've noticed that my linux box seems to run out of work. The log says:

2005-11-25 22:20:55 [rosetta@home] Sending scheduler request to https://boinc.bakerlab.org/rosetta_cgi/cgi
2005-11-25 22:20:55 [rosetta@home] Reason: To fetch work
2005-11-25 22:20:55 [rosetta@home] Requesting 8640 seconds of new work
2005-11-25 22:21:00 [rosetta@home] Scheduler request to https://boinc.bakerlab.org/rosetta_cgi/cgi succeeded
2005-11-25 22:21:00 [rosetta@home] Message from server: No work sent
2005-11-25 22:21:00 [rosetta@home] Message from server: No work sent
2005-11-25 22:21:00 [rosetta@home] Message from server: (there was work for other platforms)
2005-11-25 22:21:00 [rosetta@home] Message from server: (there was work for other platforms)
2005-11-25 22:21:00 [rosetta@home] No work from project
2005-11-25 22:21:00 [rosetta@home] No work from project
2005-11-25 22:28:34 [---] request_reschedule_cpus: project op

That's weird, as far as I know that (there was work for other platforms) only happens when they use redundancy (test on the same CPU & OS system) here ?

It may be somthing to do with the roll out of the graphics or maybe they are testing the redundacy option?

edit: fix formatting and added a bit

---

Something was up with the linux work Q last night.
A lot of my PC's had run out of work and had the message about there being work for other platforms.
The few Windows boxes i have were all fine.
A really bad effect of there failure to get any work was that the clients had all backed off for 24 hrs. :)

I am going to increase the connect to setting a bit to create more of a buffer
incase this happens again.

---

Something was up with the linux work Q last night.
A lot of my PC's had run out of work and had the message about there being work for other platforms.
The few Windows boxes i have were all fine.
A really bad effect of there failure to get any work was that the clients had all backed off for 24 hrs. :)

I am going to increase the connect to setting a bit to create more of a buffer
incase this happens again.

Great minds think alike. :)

I notioced that Rosetta has a wonderfully long completion deadline, so I increased the connect time to once every seven days. I generally do a manual update once a day, but it does mean that I have a much lower risk of running out of work.

---