Development for LHC@home

A new app version has been released (v45). This is a FAT image (548MB compressed) that contains many of the files were downloaded via CVMFS. It should reduce the amount of network traffic and hence make everything a little bit more efficient.

Please let us know how things are going by posting on the message boards.

Thanks,

Laurence

Hi Laurence,

I listened very carefully and tried only once. I got a download error. Result:

<core_client_version>7.4.36</core_client_version>
<![CDATA[
<message>
app_version download error: couldn't get input files:
<file_xfer_error>
<file_name>CMS_2015_03_13.vdi</file_name>
<error_code>-224 (permanent HTTP error)</error_code>
<error_message>permanent HTTP error</error_message>
</file_xfer_error>

</message>

Shall I try a second time or do you have to fix something?

CP

Same here.

13/03/2015 15:26:39 | CMS-dev | Started download of CMS_2015_03_13.vdi
13/03/2015 15:26:40 | CMS-dev | Giving up on download of CMS_2015_03_13.vdi: permanent HTTP error

---

Sorry, the file permissions were not correct. Should work now.

Sorry, the file permissions were not correct. Should work now.

Yes, works now.
A word of warning to those of us on slow connections, the new image is 547 MB. I'll have to make sure I don't want to watch iPlayer when I re-enable the project at home tonight... :-)

---

Sorry, the file permissions were not correct. Should work now.

Confirmed! Nice file to test my Internet speed => 9600 Kbps

The first job ended fast in the same exception message, we had before:

I'm having the same thing at home:

[37m[13/03/15 22:05:38] ----- Begin Fatal Exception 13-Mar-2015 22:05:38 CET-----------------------
[13/03/15 22:05:38] An exception of category 'Incomplete configuration' occurred while
[13/03/15 22:05:38]    [0] Constructing the EventProcessor
[13/03/15 22:05:38]    [1] Constructing ESSource: class=PoolDBESSource label='GlobalTag'
[13/03/15 22:05:38] Exception Message:
[13/03/15 22:05:38] Valid site-local-config not found at /cvmfs/cms.cern.ch/SITECONF/local/JobConfig/site-local-config.xml
[13/03/15 22:05:38] ----- End Fatal Exception -------------------------------------------------

Now, I'm sure it was working at work today, but was that before or after the latest VM? SITECONF/local is again pointing to an empty string. Guess I'll have to stroll down to the office again tomorrow. :-/

---

Now, I'm sure it was working at work today, but was that before or after the latest VM? SITECONF/local is again pointing to an empty string. Guess I'll have to stroll down to the office again tomorrow. :-/

This is, as the late Sir Terry Pratchett would say, an embuggerment. The SITECONF/local problem has definitely re-emerged; at work it's pointing to our Grid site rather than the dummy config area.

---

What should a normal run time be? In other words, if it goes longer than x minutes, we should about it?

---

Reno, NV
Team: SETI.USA

What should a normal run time be? In other words, if it goes longer than x minutes, we should about it?

24 hours runtime is the limit again now after being 1 hour for test purposes.
You can see it on one of your 'unlimited' hosts ;)

http://boincai05.cern.ch/CMS-dev/results.php?hostid=175

Now, I'm sure it was working at work today, but was that before or after the latest VM? SITECONF/local is again pointing to an empty string. Guess I'll have to stroll down to the office again tomorrow. :-/

This is, as the late Sir Terry Pratchett would say, an embuggerment. The SITECONF/local problem has definitely re-emerged; at work it's pointing to our Grid site rather than the dummy config area.

I have a suggestion from Laurence as to how to fix this. However, as a novice git/git novice I'm struggling to work out how to push the file to the repository that I supposedly have authorised access to. This may take an hour or two...

---

That sounds not so good:

[37m[15/03/15 19:11:12] ERROR:root:No message received! Nothing to do!
[15/03/15 19:12:07] ERROR:root:No message received! Nothing to do!
[15/03/15 19:13:08] ERROR:root:No message received! Nothing to do!
[15/03/15 19:14:07] ERROR:root:No message received! Nothing to do!
[15/03/15 19:15:07] ERROR:root:No message received! Nothing to do!
[15/03/15 19:16:07] ERROR:root:No message received! Nothing to do!
[15/03/15 19:17:07] ERROR:root:No message received! Nothing to do!
[15/03/15 19:18:07] ERROR:root:No message received! Nothing to do!

That sounds not so good:

[37m[15/03/15 19:11:12] ERROR:root:No message received! Nothing to do!
[15/03/15 19:12:07] ERROR:root:No message received! Nothing to do!
[15/03/15 19:13:08] ERROR:root:No message received! Nothing to do!
[15/03/15 19:14:07] ERROR:root:No message received! Nothing to do!
[15/03/15 19:15:07] ERROR:root:No message received! Nothing to do!
[15/03/15 19:16:07] ERROR:root:No message received! Nothing to do!
[15/03/15 19:17:07] ERROR:root:No message received! Nothing to do!
[15/03/15 19:18:07] ERROR:root:No message received! Nothing to do!

Hi newman,
this error pops up, when our job queue is empty. We have just filled it up again, so it should work now :)

This problem has now been fixed again. It is not clear to me how CernVM contextualization tricks and CVMFS configuration magic play together. In the meantime I put in place a workaround until this is understood.

This problem has now been fixed again. It is not clear to me how CernVM contextualization tricks and CVMFS configuration magic play together. In the meantime I put in place a workaround until this is understood.

Thanks Laurence, that looks good at home again. For my part, I wish I got on better with these verdammt versioning systems, I always seem to have trouble driving them.

---

What should a normal run time be? In other words, if it goes longer than x minutes, we should about it?

24 hours runtime is the limit again now after being 1 hour for test purposes.

I had to abort this task after running over 48 hours. Maybe the limit is not working?:

http://boincai05.cern.ch/CMS-dev/result.php?resultid=24939

---

Reno, NV
Team: SETI.USA

What should a normal run time be? In other words, if it goes longer than x minutes, we should about it?

24 hours runtime is the limit again now after being 1 hour for test purposes.

I had to abort this task after running over 48 hours. Maybe the limit is not working?:

http://boincai05.cern.ch/CMS-dev/result.php?resultid=24939

Hi,
from looking at the task it looks like some communication problem between the vboxwrapper and boinc. The stderr log is just completely empty, which it normally isn't. The task was running, right? Was the VM itself running as well? Did you have a look into the VM?

The VM is doing real work this morning ;)

Can CERN confirm that they get results back for their CMS-database?

At least I see that a connection is made to input1-boinc-cms.cs3.cern.ch

Can CERN confirm that they get results back for their CMS-database?

Jep, we are getting your results back :)