Page tree
Skip to end of metadata
Go to start of metadata

Attendees

Apologies

  • None this time.

Agenda

  • Persistence of logs between VM start-ups and VM versions.
  • Metis and light-curve product granularity.
  • VM status - Engineering.
  • VM product status.
  • Need for Splinter - or maybe telecon.
  • AOB

Discussion items

ItemNotes
Persistence of logs

RC explained that he raised this point after noting in discussion with TF that the SPICE log file was being cleared on every restart. RC discussed the advantages he saw in having the log contents preserved over a restart. He asked if any other instruments' VMs were currently clearing the log on start-up. No one present indicated that their VM did this. And TF clarified (RC had not previously understood this) that with the already implemented log-rotation the SPICE VM also no longer did this. He asked RC to try this out.

The topic moved on to whether (archived, already 'rotated') logs should be removed after a certain time, and if so, then after how long. No one expressed strong feeling about this. LH explained that the Solo-HI VM is currently rotating the log every day. On Isaias' suggestion, RC agreed to make some recommendations and circulate them with these minutes for later inclusion in the next SEGU version.

RC took the chance to stress again the need for timestamps. Without these, diagnosis from the log is much more difficult.

Metis and light-curve product granularityDM thanked MP for his answers to Dave's earlier comments, was glad to have the clarification of the granularity of the products, and indicated that he was in the process of putting together a response.
VM Status - engineering.

Explanatory points not added to table:

PHI:

  • MK asked what PHI might most easily do which would help the SOC. One option is to provide representative FITS files. AdeG said this would be useful, but asked MK about the very slow but to some extent functional VM version that was mentioned a few month ago, and suggested that might be useful for the SOC despite slowness. MK indicated that he suspected that such a VM as he could provide quickly probably would not provide output very different from that coming from the last version 01.03.00. It emerged that everyone's memories of the status of the output of that VM were very blurry. The SOC indicated that representative fits files would indeed be very useful, and they would review the status of the 01.03.00 outputs.
  • RC mentioned the separate log files for each request produced by the PHI VM, and MK explained the reason for this (separate tasks forked for each request).
  • RC mentioned also the 'odd' characters noted in the log files. But he was not able to remember the details. He will review and let MK know

MAG/Telemetry for testing:

  • Isaias asked if the SOC would be feeding E2E-1 test data into the VMs. CW answered that no, as in fact there would be no science data produced as a result of these tests. However he was supportive of feeding data from other tests. It was understood that in the case of MAG, FM data was particularly needed. CW indicated that the SOC will try to get hold of science telemetry data where possible - not just for MAG. AW noted that the best sources were very probably the thermal vacuum tests.

STIX:

  • Shane confirmed that work is progressing on the next STIX VM version, but had found he had to first pull together all the FITS processing software and had just returned to working on the LLVM specifics. He plans to deliver a new VM in 1 month.
VM product statusThere were no specific new product files that the SOC had examined since the last telecon.
AoB

It was decided that instead of a splinter at the SOWG we would have a further telecon on the 24th of July.

Preparatory Actions

New Actions

None.