Документ взят из кэша поисковой машины. Адрес оригинального документа : http://www.eso.org/~qc/admin/QC_Guide_for_Paranal_Missions.pdf Дата изменения: Tue May 22 11:54:31 2007 Дата индексирования: Tue Oct 2 09:44:01 2012 Кодировка: Поисковые слова: п п п п п п п п п п п п п п п п п п

QC Guide for Paranal Missions
Reinhard Hanuschik. Version 1.1 (2007-05-22) Quality Control at the VLT is a shared process. A major part of it is done by QC Garching, another part is done as on-line QC by Paranal staff. Therefore it is important to maintain and improve the contacts to the Paranal Instrument Scientist and other key persons there in the QC process. QC scientists are therefore expected to travel to Paranal at least once, better twice a year. There is no fixed length for the stay. A reasonable rule of thumb is probably 3-6 days, depending on the number of issues and the number of previous visits. As a first-time Paranal visitor, a longer stay is preferable, with a not so dense daily schedule. Mission goals: · see the process · understand the process · improve the process Prepare: · list of items to be discussed · list of questions you have · list of proposals · cross-check with supervisor if in doubt · ask your colleagues if you are on a first-time mission Come: · in a positive spirit; usually your visit will be much appreciated · don't come as `controller'; we want to understand and help · be proactive, come with new proposals, help to improve the system · listen; react positively to proposals, think about how to implement them · make sure that you share with your colleagues on the mountain the same vision about the instrument and its data Possible topics of your collaboration: · Health Checks, trending plots o explain what you have, what their purpose is o discuss what could be improved · QC1 parameters: o are they complete? o after a couple of years: do you need all of them? pipeline: o new recipes needed? o improve existing recipes? o performance ok?

·


·

you are expected to participate in the daily operational meeting (usually at 16:00 in the control room)

Upon arrival · know and follow all rules and regulations on the Observatory · if you do not comply, there is probably nothing we can do for you here in Garching · if in doubt about the rules, always ask your contact person · don't forget: this is a site at 2600m and one of the driest places in the world. Be careful and apply common sense. Sociology · if you have earlier succeeded in building a good relationship, you will now capitalize on that · if not yet, this is your chance to set it up Watch the process: · understand how the people work, and what their role is o see what the daytime / nighttime astronomers do; o see their check lists o see the data handling administrators and ask them about their work · see the data flow system working: o see the pipeline running on the pipeline workstation; best time for this is when the daytime calibs are taken o understand the differences to the Garching system o see and understand the DataOrganizer and the ReductionBlockScheduler working o see the Appendix for technical details check the ops log files check the autrep tool arrange for a tour of the telescopes and instruments; try to see as much as possible; it may be useful for another assignment in two years from now ...

· · ·

Keep in touch · read your emails (use webmail) · ask your supervisor or your colleagues if in doubt After the tour · prepare a short end-of-mission report, to be submitted to the group head Work · you are on daytime schedule; nevertheless it may be useful to see for a few hours the nighttime astronomers doing their job


· ·

do overtime only if necessary; get it signed by the contact person or by head of sciops usually you can avoid doing overtime: you are not part of a commissioning team, you don't have a time-critical mission; so don't exhaust yourself; better enjoy after a long day the swimming pool

Technical appendix Pipelines
Pipeline workstations: wu1pl, wu2pl etc. These machines run the operational pipelines. Feel free to watch them, but don't interact with them! Log in as `pipeline'. The pipeline workstations are part of the Data Flow System: raw data are delivered from the instrument workstations, pipeline products are created. · · · · · raw data: /data/raw/ pipeline products: /data/reduced/ master calibrations: in the calibDB (see below) there are no ABs; reduction blocks are in $DFS_REDBLOCK (/data/redblock) pipeline logs: $DFS_LOG (/data/msg)

Contrary to the dfo machines in Garching, these workstations are organized per unit telescope. They collect all data for all instruments on that telescope. E.g., wu2pl will have a mixture of FORS1, UVES, and GIRAFFE files, logs etc. Calibration database: · under /cal///cal (some instruments: /cal//cal) find the calibration files, both the static ones and the `dynamic' ones (where dynamic means the one provided by QC Garching and refreshed every few months) · /cal//rul: here are the configuration files for the DO · /cal//rec: the pipeline recipes · /cal///cal Data Organizer (DO): this is the equivalent of the OCA association tool in Garching. Its main difference to ABbuilder is that it works strictly sequentially. Find its log as $DFS_LOG/DataOrganiser.log . Reduction Block Scheduler (RBS) is the interface to the pipelines. This tool also works strictly sequential. Find its log as RBS. under $DFS_LOG. Both tools run in automatic mode and in the background. Don't touch them ­ they are operational!


Link to Garching
All machines on Paranal are firewall-protected. You cannot outside world, nor can you connect to them from Garching. exception, odyssey4.pl.eso.org. Login is possible from both or rcp. The account name is par2gar. You can e.g. transfer connect to the Garching operational machines. connect to the There is one sides with telnet files or use it to