LOFAR Newsletters August 2015
Distributed on 26 August 2015.
Array status:
----------------------
- 38 stations operational in the Netherlands: 24 core stations and 14 remote stations.
- 9 international stations operational: DE601, DE602, DE603, DE604, DE605, FR606, SE607, UK608, DE609.
- 3 new international stations are being built in Poland and should become part of the array by the end of 2015.
- CS013 is not available for production observations due to problems with software installed on the new LCU - the fix is in progress.
- Various Dutch stations were switched off during the warmest summer days because the temperatures within the cabinets reached critical levels. The Cycle observing schedule has been adapted accordingly.
- Maintenance was performed at various stations throughout the summer and it is still in progress.
- The overview of non-operational antenna elements for LBA and HBA is available here.
- Station calibration: mode 7 calibration tables were updated on August 24^th for all international stations except UK608 and DE605, for which new data must be collected. New mode 3 tables will be produced for all Dutch stations. With a few exceptions, the calibration tables in all modes were
introduced no earlier than Feb 2015. The station calibration status is shown here.
Observing System Status:
------------------------------------
- The observing system was mostly stable during the period July - August 2015. However, a few failures at COBALT nodes, as well as (not announced) outages at the international station lines, warm temperatures, and the instability of processing and ingest systems caused delays in the
exploitation of the Cycle 3&4 observing programmes and in digesting data out of CEP2.
- Low ingest speeds towards the Juelich LTA location are experienced during ILT runs involving the German stations - investigations are still in progress at Juelich to try to fix this issue.
Software Development Status (H. Holties):
------------------------------------------
Over the summer period, the LOFAR Software Development team has continued work on reliability and responsiveness improvements. Rolling out of the message broker infrastructure has continued and information from the pipelines running on CEP2 is now communicated via this infrastructure to the
rest of the system. This development will make handling of pipelines more robust against network hiccups which are likely to occur during long running processing jobs. To further improve robustness, and provide relief to LOFAR operations, communication between the observatory management
software and the LTA ingest system has been simplified, taking out software components that were no longer maintained. Interaction with the TBB community has resulted in a work plan for the integration of TBB operation into the LOFAR system. Several aspects of this will be reusable for other
responsive modes of operation. Adaptations of station software to support the 160 MHz mode have been carried out and are being thoroughly tested. The above software developments, as well as many smaller changes and fixes, will be rolled out in the upcoming software release (2.12) on September
7^th. Full details of all the changes will be available on the LOFAR wiki. Finally, the team is getting involved in the preparation for the CEP4 cluster which will be configured and commissioned in the coming months.
CITT Status (T. J. Dijkema):
-----------------
The Calibration & Imaging Tiger Team (CITT) is wrapping up its activities. Among other things, we have a pipeline version of the Facet Calibration technique, delivering great HBA images. That does not mean we're done: work will continue in CITT2. We will present the outcomes of the CITT and
present plans for CITT2 in a meeting on September 17 and 18 at Astron - see the invitation that was sent to the lofar-news mailing list, or ask dijkema [at] astron [dot] nl for a copy of that announcement.
MSSS status (G. Heald):
------------------
- The first MSSS paper presenting the survey design along with LBA and HBA data from the first 100 square degrees (the MSSS Verification Field) has now been accepted for publication in A&A and will soon be posted on arxiv.
- A major current effort is in finalizing the flux scale across the full survey area.
Observing Programmes and miscellaneous items:
--------------------------------------------------------------------
- Cycle 4 observing programme: 52% complete. The rest is scheduled for completion by November 14^th 2015. The Cycle 4 observing schedule is available here.
- Cycle 3 observing programme: 91 % complete - Remaining 9% is being observed with second priority during Cycle 4.
- The Cycle 5 proposal call was advertised on July 15^th 2015. The text of the call can be found here. In preparation for the call, all LOFAR web pages have been updated - in particular, find here important information about what functionality will be available during the next Cycle. The
proposal submission deadline is September 9^th 2015, 12 UT. The PC meeting will take place on October 29^th - 29^th 2015. Cycle 5 will start on November 15^th 2015.
- Characterization of the current automatic self-calibration pipeline is not yet warranted - for details, see e-mail distributed on August 21^th. In the period before standard CEP2/CEP4 self-calibration becomes operational, users can request processing time on ILT computing resources (CEP3) to
perform self-calibration using the most recent tested version of the standalone pipeline, if they do not have the requisite resources available themselves.
- A new frequency-to-SB converter which also presents bandpass plots for most modes in now available here.
- From March 3^rd until the May 13^th 2015, a bug was inadvertently introduced in the source finder software PyBDSM. For more details, consult the full information given online.
- The Long Baseline working group found that the Station Adder step of NDPPP performed an incorrect calculation of the visibilities for the super station, resulting in an increase in image noise of about 30%. The PIs of all affected projects have been notified. A corrected version of NDPPP has
been rolled out on July 28^th. A full description of the issue is available here.
- On August 19^th, a software change has been applied to COBALT to correct a TAB pointing offset when calculating it from a TAB ring specification. This bug was already present since initial TAB ring support for beamforming also in the BlueGene.
Note: the TAB RA/DEC metadata is and was in all cases accurate with respect to what was actually observed. Just not with what was implied in the TAB ring specification.
- The syntax `[CR]S*&' in NDPPP should be instead replaced by `CS*,RS*&'. This is because the use of `[]' triggers a bug in the code. This info will be reported in the Lofar Imaging Cookbook.
- At the software roll out of June 29^th, a switch to casacore v2 was made in the production system, unaware that a backward incompatible change was then introduced. In particular, users started experiencing problems when trying to read with CASA Measurement Sets created after the
aforementioned date. For new data, a fix has been rolled out in production on August 25^th. On August 26^th, the Radio Observatory distributed a fix script for the data created between June 29^th - August 24^th 2015.
CEP news:
----------------
- CEP2:
* Disk space situation is often critical, especially in case of instability of the processing and ingest systems and after week-ends.
* Locus100 fully cannibalized for spare parts to repair other failing nodes. Locus095-099 are being emptied for the same purpose.
- CEP3:
* Cluster info and schedule available here.
* Monitoring of node usage is active now. These statistics will be used to evaluate requests for extensions.
Calendar next LOFAR activities:
-------------------------------------------------------------------
- Next LSM's : 02/09, 16/09, 30/09, 14/10, 28/10, 11/11
(note: all presentations given at the LSM are available here).
- Next Stop days: 01/09, 06/10, 10/11
- Next software roll outs: 07/09, 02/11
- Cycle 5 proposal submission deadline: September 9^th, 12 UT
- Next LOFAR bulletin: October 2015.