Hi all,

We just released native version 2.87. This version uses an "unpacked" Singularity image file instead of a sif file. This means tasks start quicker because there is no need to unpack the image (which can take up to a minute), and also there is no longer any need to install squashfs-tools, which was only used for this unpacking step.

As usual let us know if there are any problems!

(in case you need to debug any problems, the image changed from /cvmfs/atlas.cern.ch/repo/containers/images/singularity/x86_64-centos7.img to /cvmfs/atlas.cern.ch/repo/containers/fs/singularity/x86_64-centos7)

Thank you David and your Team,
all native are running well.
squashfs is now history, this is a good idea.

ATLAS native has been updated I get lots of messages like:
[pre]Do 26 Aug 2021 09:09:27 CEST | LHC@home | No tasks are available for ATLAS Simulation[/pre]
although the server status page shows 4346 unsent ATLAS tasks.


In the past this error was caused by the fact that LHC's BOINC service runs on 2 (sometimes more) machines, currently:
[pre]lhcathome.cern.ch. 60 IN A 137.138.148.247
lhcathome.cern.ch. 60 IN A 137.138.156.174[/pre]


It looks like only 1 of the BOINC service instances - or the corresponding ATLAS queue behind the service - is correctly configured to send out work for the recent ATLAS version v2.87.
The other service/queue may require a restart.

Thanks for pointing this out. Hopefully the power outage this afternoon in the CERN data centre took care of restarting the other machine.

s still a mix of work requests on all clients that succeed and others that don't.
Merged from different clients:
[pre]27-Aug-2021 00:32:31 [LHC@home] Scheduler request completed: got 1 new tasks
27-Aug-2021 01:09:53 [LHC@home] Scheduler request completed: got 1 new tasks
27-Aug-2021 01:28:52 [LHC@home] Scheduler request completed: got 1 new tasks
27-Aug-2021 01:43:06 [LHC@home] Scheduler request completed: got 1 new tasks
27-Aug-2021 01:59:17 [LHC@home] No tasks are available for ATLAS Simulation
27-Aug-2021 01:59:28 [LHC@home] No tasks are available for ATLAS Simulation
27-Aug-2021 02:02:26 [LHC@home] Scheduler request completed: got 1 new tasks
27-Aug-2021 02:13:01 [LHC@home] Scheduler request completed: got 1 new tasks
27-Aug-2021 02:27:11 [LHC@home] Scheduler request completed: got 1 new tasks
27-Aug-2021 02:37:32 [LHC@home] No tasks are available for ATLAS Simulation
27-Aug-2021 02:37:43 [LHC@home] No tasks are available for ATLAS Simulation
27-Aug-2021 02:40:19 [LHC@home] Scheduler request completed: got 1 new tasks
27-Aug-2021 03:13:04 [LHC@home] No tasks are available for ATLAS Simulation
27-Aug-2021 03:15:12 [LHC@home] Scheduler request completed: got 1 new tasks
27-Aug-2021 03:24:54 [LHC@home] Scheduler request completed: got 1 new tasks
27-Aug-2021 03:50:53 [LHC@home] Scheduler request completed: got 1 new tasks
27-Aug-2021 05:05:13 [LHC@home] Scheduler request completed: got 1 new tasks
27-Aug-2021 05:14:59 [LHC@home] No tasks are available for ATLAS Simulation
27-Aug-2021 05:15:11 [LHC@home] No tasks are available for ATLAS Simulation
27-Aug-2021 05:17:47 [LHC@home] Scheduler request completed: got 1 new tasks
27-Aug-2021 05:37:13 [LHC@home] Scheduler request completed: got 1 new tasks
27-Aug-2021 05:47:02 [LHC@home] No tasks are available for ATLAS Simulation
27-Aug-2021 05:47:13 [LHC@home] No tasks are available for ATLAS Simulation
27-Aug-2021 05:49:49 [LHC@home] No tasks are available for ATLAS Simulation
27-Aug-2021 05:50:00 [LHC@home] No tasks are available for ATLAS Simulation
27-Aug-2021 05:52:22 [LHC@home] No tasks are available for ATLAS Simulation
27-Aug-2021 05:52:33 [LHC@home] No tasks are available for ATLAS Simulation
27-Aug-2021 05:55:55 [LHC@home] Scheduler request completed: got 1 new tasks
27-Aug-2021 06:22:23 [LHC@home] No tasks are available for ATLAS Simulation
27-Aug-2021 06:22:44 [LHC@home] No tasks are available for ATLAS Simulation[/pre]


May be the cleaning lady/guy pulled the wrong plug yesterday.

XXX.YYY 3128 - - [27/Aug/2021:09:59:13 +0200] "GET http://s1nikhef-cvmfs.openhtc.io/cvmfs/singularity.opensciencegrid.org/.cvmfspublished HTTP/1.1" 200 1830 "-" "cvmfs Fuse 2.4.4 49b2fac1-df25-48d2-a4ee-4612ca6a31f8" TCP_REFRESH_MODIFIED:HIER_DIRECT
Is this a normal entry for squid?

May be the cleaning lady/guy pulled the wrong plug yesterday.

Actually it was a planned power test which went slightly wrong...

I think I found the other host and restarted the services there.

... Is this a normal entry for squid?

Yes.
Have them too.


Within 1 week:
Destination #requests #bytes
s1nikhef-cvmfs.openhtc.io 132506 2.53 GB

I think I found the other host and restarted the services there.

Thanks.
Since then work requests are running fine.

[2021-08-27 13:59:51] CVMFS is ok
[2021-08-27 13:59:51] Using singularity image /cvmfs/atlas.cern.ch/repo/containers/fs/singularity/x86_64-centos7
[2021-08-27 13:59:51] Checking for singularity binary...
[2021-08-27 13:59:51] Using singularity found in PATH at /usr/bin/singularity
[2021-08-27 13:59:51] Running /usr/bin/singularity --version
[2021-08-27 13:59:51] singularity version 3.8.1-1.el8
[2021-08-27 13:59:51] Checking singularity works with /usr/bin/singularity exec -B /cvmfs /cvmfs/atlas.cern.ch/repo/containers/fs/singularity/x86_64-centos7 hostname
[2021-08-27 13:59:52] WARNING: While bind mounting '/cvmfs:/cvmfs': destination is already in the mount point list fx8s
[2021-08-27 13:59:52] Singularity works
Need this Info a correction for Singularity installation of the CentOS8-VM?

WARNING: While bind mounting '/cvmfs:/cvmfs': destination is already in the mount point list.
CentOS8-VM with Singularity version 3.8.0-1.el8 don't have this warning,
only CentOS8-VM with Singularity version 3.8.1-1.el8.

When four native-Tasks for a maximum is reached, Boinc (7.16.11) get this News text every time:

LHC@Home: Notice from server
VirtualBox is not installed

Is it possible to stop this News text in Boinc, because this is only for News.