I recently checked on a node that I assumed was running fine, and it had a very old stuck work unit.

I've removed it, but when I issue an update request via boinccmd, it logs
"[climateprediction.net] Not accepting requests from this host"

How do I get this fixed ? Forum posts mention some sort of email address I should use but I can't locate one.

http://climateapps2.oerc.ox.ac.uk/cpdnboinc/show_host_detail.php?hostid=879914 if it's any help...

---

Hello Falken

That computer was blocked because it was crashing a lot of models.
You would have been sent an email about it at the time.

Check the error messages, fix the problem, and then post here about it.
We'll then ask for it to be unblocked.

I never got an email I noticed, sorry.

The last work unit was a while ago, as you can see, but it is just listed as "Timed out - no response", the one before that was "completed" though.

I'm running the latest BOINC version now, so it doesn't need to be blocked any more ?

---

Error messages are listed inside the page for each model.However, as it was some time ago, with a different BOINC version, I'll ask for it to be unblocked, and we'll see what happens.

Cheers, do you know how long that will take so I know when to check the logs again ?
At least as of now, it's still blocked:

28-Feb-2015 10:50:55 [climateprediction.net] update requested by user
28-Feb-2015 10:50:56 [climateprediction.net] Sending scheduler request: Requested by user.
28-Feb-2015 10:50:56 [climateprediction.net] Requesting new tasks for CPU
28-Feb-2015 10:50:59 [climateprediction.net] Scheduler request completed: got 0 new tasks
28-Feb-2015 10:50:59 [climateprediction.net] Not accepting requests from this host

[/code]

---

Certainly won't happen over the weekend.

That machine won't receive new work until the boinc installation is changed. It has v.7.2.39, a dodgy version. Post back when that is done and we'll ask staff to reset the "-1" value for number of allowed daily tasks (check information for the machine in 'Your account').

By the way, that machine's last task, two years ago, was allowed to languish until it timed-out and, much later, the task was issued to and completed on another machine.

---

"We have met the enemy and he is us." -- Pogo
Greetings from coastal Washington state, the scenic US Pacific Northwest.

It's now running

# rpm -q boinc-client
boinc-client-7.2.42-1.gitdd0d630.fc20.x86_64

it was probably upgraded some time ago :-)

---

Good. I'll notify Andy.

Please check that your 64-bit Linux installation has the 32-bit libraries installed.

---

"We have met the enemy and he is us." -- Pogo
Greetings from coastal Washington state, the scenic US Pacific Northwest.

Thanks.

What exactly should I be checking ?
If you mean those listed towards the end of
http://boinc.berkeley.edu/wiki/Installing_on_Linux#Fedora
then this is done, apart from compat-libstdc++-33.i386 which appears to be ....i686 these days.

I assume I can just run 'ldd' on the modal once it's downloaded one to check ?

---

There's a sticky post at the top of this Linux section about the 64 bit problem.

If there's "the usual" missing lib, then the model will crash in about 6 seconds.

Ah right, ta, yes, those are installed.

Hopefully the mysterious Andy will plug me back in :-)

---

You are plugged-back-in and there is a new release of European regional tasks for Windows and Linux.

Andy is one of the too-short staff at Oxford we rely on to implement new releases and keep the home-fires burning.

---

"We have met the enemy and he is us." -- Pogo
Greetings from coastal Washington state, the scenic US Pacific Northwest.

Tell Andy thanks, tasks running.

Back to saving the future...

---

Done.

Good luck with helping save the future . . . Every bit helps. eh?

---

"We have met the enemy and he is us." -- Pogo
Greetings from coastal Washington state, the scenic US Pacific Northwest.