I was just looking through my single core machine's task list and saw an interesting WU.

wuid=1384280749

_0 timed out
_1 completed, validation inconclusive (me)
_2 completed, marked as invalid
_3 in progress


That's.... odd.

The Stderr for the invalid has no autocorr result. The guy is still trying to run v7 tasks through a v6 app. Take a look at Invalid tasks for computer 5940769. Lots of resources going to waste.

There are a few Linux hosts doing this and it's a shame that the people running these rigs obviously arn't monitoring their output. :-(

Cheers.

The Stderr for the invalid has no autocorr result. The guy is still trying to run v7 tasks through a v6 app. Take a look at Invalid tasks for computer 5940769. Lots of resources going to waste.

ack.. I got nosy enough to notice that seemed weird, but not nosy enough to take it one click further. I've been doing that more and more lately... odd.

---

Linux laptop:
record uptime: 1511d 20h 19m (ended due to the power brick giving-up)

Have we got another stuck tape?
Ready-to-send buffer had been slowly but steadily growing, till a short while ago. Now it's falling again & the splitter output appears to be spluttering.

---

Grant
Darwin NT

Combination of demand and a shorty storm

---

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

Combination of demand and a shorty storm

And splitters that struggle even at the best of times.

---

Grant
Darwin NT

There's also an issue with the upload server that is, as far as I know, unresolved since BOINC started, in that it will accept more connections than it can service when it is overloaded. Uploads then time out and need to be restarted and all that bandwidth is thus wasted. So, once it starts getting overloaded the overload snowballs from the wasted bandwidth and retries.

You can see this easily in the BOINC client transfers tab by noting all the failed uploads with a significant percentage of completion. All of that is wasted bandwidth as they don't resume, unlike the downloads which do.

---

The servers hate me; over 1/4 million WU ready to be sent and I got a total of eight in the last hour.

---

As has been noted by others it is the bug.

---

The servers hate me; over 1/4 million WU ready to be sent and I got a total of eight in the last hour.

Your BOINC version 7.2.39 has been noted as buggy either upgrade to 7.2.42 or downgrade to an older version could help you out.

Cheers.

...BOINC version 7.2.39 has been noted as buggy either upgrade to 7.2.42 or downgrade to an older version could help you out...

The problem did start after I "upgraded" to 7.2.39. So it is NOT my bug.

---

...BOINC version 7.2.39 has been noted as buggy either upgrade to 7.2.42 or downgrade to an older version could help you out...

The problem did start after I "upgraded" to 7.2.39. So it is NOT my bug.

That's what Wiggo is trying to tell you. You upgraded to a buggy version. Upgrade again or downgrade to where you were before.

---

David
Sitting on my butt while others boldly go,
Waiting for a message from a small furry creature from Alpha Centauri.

There's also an issue with the upload server that is, as far as I know, unresolved since BOINC started, in that it will accept more connections than it can service when it is overloaded. Uploads then time out and need to be restarted and all that bandwidth is thus wasted. So, once it starts getting overloaded the overload snowballs from the wasted bandwidth and retries.

You can see this easily in the BOINC client transfers tab by noting all the failed uploads with a significant percentage of completion. All of that is wasted bandwidth as they don't resume, unlike the downloads which do.

But at least with the bandwidth available at the Colo, it can eventually work itself out.

---

David
Sitting on my butt while others boldly go,
Waiting for a message from a small furry creature from Alpha Centauri.

Finally finished those 4 AP tasks, appr 150 hours each

RAC went up with almost 400 points woohoo!

---

rOZZ
Music
Pictures

But at least with the bandwidth available at the Colo, it can eventually work itself out.

Agreed... happens far less often than it used to (remember when it was almost daily?) CoLo move was quite the improvement.

---

Finally finished those 4 AP tasks, appr 150 hours each

RAC went up with almost 400 points woohoo!

GG Julie :)

---

It only took a little over 52 hours for my Pentium (E5200)

http://setiathome.berkeley.edu/show_host_detail.php?hostid=7225420

to complete an AP task.

http://setiathome.berkeley.edu/result.php?resultid=3413643091

I thought that was slow. But, then again, I was comparing it to one of my GTX780ti's.

That is Slow. However, anyone running the Stock AP App deserves what they get. Now, if you ran the Optimized AP APP on your slightly OCed E5200 you would complete one in about 10.5 hours...like mine, http://setiathome.berkeley.edu/result.php?resultid=3410271666
Run time: 37,704.22
Windows x86 rev 557, Don't Panic!, by Lunatics SSE2

Back when this host was running, http://setiathome.berkeley.edu/show_host_detail.php?hostid=6864181 it was completing an AP in about 8 hours using AstroPulse_v6, v6.01 SSE3 (r546) on a very similar processor. I'm thinking about reviving that host as soon as APs are back as it was completing APs faster than in Win XP.

Well.. my 10-day cache of APs finally ran dry this morning. Oh well. More will arrive eventually. Might get an odd resend here and there in the meantime.

---

Linux laptop:
record uptime: 1511d 20h 19m (ended due to the power brick giving-up)

It only took a little over 52 hours for my Pentium (E5200)

http://setiathome.berkeley.edu/show_host_detail.php?hostid=7225420

to complete an AP task.

http://setiathome.berkeley.edu/result.php?resultid=3413643091

I thought that was slow. But, then again, I was comparing it to one of my GTX780ti's.

That is Slow. However, anyone running the Stock AP App deserves what they get. Now, if you ran the Optimized AP APP on your slightly OCed E5200 you would complete one in about 10.5 hours...like mine, http://setiathome.berkeley.edu/result.php?resultid=3410271666
Run time: 37,704.22
Windows x86 rev 557, Don't Panic!, by Lunatics SSE2

Back when this host was running, http://setiathome.berkeley.edu/show_host_detail.php?hostid=6864181 it was completing an AP in about 8 hours using AstroPulse_v6, v6.01 SSE3 (r546) on a very similar processor. I'm thinking about reviving that host as soon as APs are back as it was completing APs faster than in Win XP.

Slow would be the 7.27 android app on beta currently.

---