For users and passwords, see
SysAdmin.
Machines
Data Analysis
Need analysis PCs for analysis rooms in B510.
DAQ
DAQ PCs for
BL4S
Readout for
BL4S DAQ crate
- Model
- CTC VP 717 SBC
WebcamDetector and
TimePix DAQ PC
Main DAQ PC
Monitoring shift PC
HV crate controller
Pilot test machine, also to be used for the first camera.
- Supermicro
- XEON E5440 @ 2.83 Ghz
- 4GB system RAM
- Supermicro
- XEON E5440 @ 2.83 Ghz
- 4GB system RAM
Computer Setup
Operating system
CC7
- CERN CentOS7
- http://linux.web.cern.ch/linux/centos7/
SLC6
- Scientific Linux CERN
- http://linux.web.cern.ch/linux/scientific6/
Windows
- Windows 7
- https://espace2013.cern.ch/winservices-help/NICEEnvironment/NICEInstallation/Pages/InstallationOfWindowsAtCERN.aspx
Kerberos
- Setup information
- http://linux.web.cern.ch/linux/docs/kerberos-access.shtml
Service accounts
See
AFSAndServiceAccount for writing out a keyfile for a service account.
There is also a
mk-keytab
script in the bl4sextras repo:
https://svnweb.cern.ch/cern/wsvn/bl4sextras/scripts/mk-keytab
Services
virt-who
This seems to be on by default in SLC6 but throws this error into
/var/log/rhsm/rhsm.log
every minute on every system we have:
2015-04-28 18:05:49,885 [ERROR] @virtwho.py:133 - Error in communication with subscription manager, trying to recover:
Traceback (most recent call last):
File "/usr/share/virt-who/virtwho.py", line 125, in _send
self._sendGuests(config, virtualGuests)
File "/usr/share/virt-who/virtwho.py", line 150, in _sendGuests
manager = Manager.fromOptions(self.logger, self.options)
File "/usr/share/virt-who/manager/manager.py", line 41, in fromOptions
return subcls(logger, options)
File "/usr/share/virt-who/manager/subscriptionmanager/subscriptionmanager.py", line 48, in __init__
self.readConfig()
File "/usr/share/virt-who/manager/subscriptionmanager/subscriptionmanager.py", line 62, in readConfig
raise SubscriptionManagerError("Unable to read certificate, system is not registered or you are not root")
SubscriptionManagerError: Unable to read certificate, system is not registered or you are not root
2015-04-28 18:05:49,897 [ERROR] @virtwho.py:136 - Unable to recover, retry in 60 seconds.
It can be disabled with
sudo chkconfig virt-who off
CCM
This is a cron task in the CERN openstack CC7 image. The task is defined in
/etc/cron.d/ccm-fetch.cron
as:
0 * * * * root (perl -e 'sleep rand 3600'; /usr/sbin/ccm-fetch) >>/var/log/ccm-fetch.log 2>&1
This dumps the following into
/var/log/ccm-fetch.log
:
LWP::Protocol::http::Socket: Bad hostname 'host.mydomain.org' at /usr/share/perl5/LWP/Protocol/http.pm line 51.
2015/04/30-17:57:39 [WARN] Retrieve: <http://host.mydomain.org/profiles/git-test.xml>: Can't connect to host.mydomain.org:80 (Bad hostname)
- CERN service ticket
- https://cern.service-now.com/service-portal/view-incident.do?n=INC0780289
Storage
Workspaces
- Dedicated AFS workspace for datafiles/backups is at
-
/afs/cern.ch/project/bl4s
- Administration is explained at
- http://information-technology.web.cern.ch/book/afs-administrator-guide
Automation
cfg
- Host
- bl4sdaq.cern.ch
- Period
- 1 min
- Command
- /afs/cern.ch/user/d/daquser/public/daq_extras/scripts/autogit -vf ~/public/daq_extras/BL4S_analysis/cfg &>> /home/daquser/logs/autogit.log
Version all files in the
BL4S_analysis configuration folder. Changes can be viewed at:
https://gitlab.cern.ch/BL4S/config/commits/master
Changes pushed to that repo will be pulled down by the script and merged if possible.
afs-sync
- Host
- lnxpool41.cern.ch
- Period
- 1 min
- Command
- /afs/cern.ch/user/d/daquser/public/afs_sync.sh 1>> /home/daquser/logs/afs_sync.out 2>> /home/daquser/logs/afs_sync.err
Upload RAW data files from SBC hard drive into AFS project space. Uses a lock file to ensure only one copy at a time. Uses rsync for copy. Only syncs
*.data
files, not
*.writing
.
--
TimBrooks - 2015-04-29