SW&IT Meeting Minutes 2019-06-10
Attendees: Andrew, David, Igor, Matthieu, Petr, Stephen, Xianyu, Yang
Report auto-generated every, Do not edit this table, use entries in smartsheets.
| LR20190603 - Bec, Matthieu
| LR20190603 - Colson, Andrew
| LR20190603 - Edmeades, David
| LR20190603 - Sola, Igor
| LR20190603 - Kubanek, Petr
| LR20190603 - Hooper, Stephen
| LR20190603 - Zhang, Yang
| LR20190603 - Zhang, Xianyu
News from management
Previous Action Items
- TCS leap seconds (?)
- use of calendar for recuring tasks
- OC, WPC, management
- started review group PE2018, prepared my own
- dx-hp1 closer look in telemetry
- AHS project discussion with CV, LF
- Jenny's induction to Github - may need someone follow up
- WPC 'state of the s/w address' and obs1
- second meet up with Ketiv, looked with David at their script
- setuo DDS on catalog.moutain, clean some rpm along the way
- lookig at obs1 next steps, some rpm perl/python2 generation
- 4D mount
- Centos 6 doesn't play nice with SMB >1.0, so I had to re-enable it on the 4D machines. This is technically a security risk, but we should be insulated from exploitation due to its place on the network
- Mapped drive from Aquisition to Analysis host, created simlink to mimic the path.
- 4D remote desktop
- VNC works, but is subject to the recent bug/quirk/classic Redmond wherein VNC fails when no monitor is attached
- Decision to switch to RDP. Single connection only, but works with no monitor.
- 4D software crashes
- Unplugging jumper cable had no effect
- Changing IP config to mismatch with the 4D software's config file caused crash with "unexpected error". This is extremely unhelpful and I can't think of a reason why it would crash the software given that it works fine with no cable plugged in. It seems like something that could easily crop up during operations.
- Xianyu is setting up a call with 4D regarding this and other questions.
- Matlab licensing
- Jet: Decision to keep it on its yearly license. Heard back from the license administrator: they have submitted the refresh PO and expect to have the new one in place by 30 June. The current licenses expire on 30 July, so we should have plenty of overlap.
- Test stand: Similar to jet, generate license file from the Mathworks web site and sneakernet it over. It is similarly tied to the expiration of the campus license.
- Octave: I left it installed on the test stand PC and Jonathan has expressed interest in trying it. He is currently struggling with getting the code running on native Matlab, so I expect progress to be slow on this front.
- Autodesk licensing
- Limited-license v. 2018 products are good until 2021
- I put notes on the calendar with reminders 120 and 60 days ahead.
- Google outage
- After initial claims that it was caused by external network congestion, they have finally posted a full post-mortem, which found that it was a confluence of small misconfigurations and bugs that led to a major cascade failure.
- Possible problems with IMAP auth at the end of last week (Xianyu, how's that going?)
- Safety training
- JP/Notification script: I emailed about it and she doesn't seem like she needs a lot of help. She deferred the conversation until "July", at which point I will render what assistance she needs.
- Vault replication take 2
- Opened a ticket with Autodesk, fobbed off to forums. No reply to my question
- Emailed Ketiv as another angle. Jose was out of the office last week; I will follow up today
- Rack diagram and network
- Images for LBTO site mockup
- 4D call
- Docker, x11docker
- system packages
- Ice3.7 tests looks good
- worked with Yang's PCS SHARK patch
- Ice3.7, fix PEPSI problem
- UMACs, AOC,..
- OVMS FFT plots
- web-dms x vs y plots
- Verified refresh rate of all-sky and radar images in wda.sh
- safety training
- TCS work:
- SHARK authorization nearly completed (github pull request to be merged).
- Updates to ApplyWaffleAO ready and to be followed on github.
- Discussions about the instrument hardware alarm bits assignment for SHARK.
- LBC flat wiki:
- Updated with spec and systemd settings, first pull request ready to go.
- Working on some refinements and updates.
- LBC work:
- Several LBC issues came up.
- Will focus on IT#7723, understanding how the LBC focus mechanism works.
- Investigate IT#7731, change how LBC stores data (saving directly to a remote filesystem).
- work on supportscripts (porting), packaging, etc; work on Steve A's scripts the same.
- kickstart John's Godzillux to obs1 base
- kickstart catalogs.mountain
- cleaning out 580F with Robert's help
- mounts for Steward -- needs more discussion and planning
- Python packaging into RPMs -- got to go through how to do this.
- Kickstart display NUC
- Second round of mods to "supportscripts".
- github Steve's perl
- build computer for AUX control room.
- Once LUCI ships, computers in 507 for TSIP/Eng?
- Started working on OT issue list.
- Working on getting a common configuration file for each of the LUCI GUIS, for remote and mountain obs machines. It is easier to maintain if we create templates and a config file generator that adds the machine name and the missing parts of the xml. The generator will create a single file, that will be different for each of the LUCI GUIs and will be the same for obs machines in Tucson or mountain.
- Finish configuration files and do the stow of LUCI code and environment to the /usr/local
- Look at what's needed to save SRR mode fits cube. (ongoing)
- get all LUCI scripts in git (ongoing)
- Keep working on OT issue list.
- Make sure all scripts are in git
- clean up Italian spreadsheet script generation.
- Work with David for the 4D IT issues
- Procedures to determ jumping actutor with Guido
- DX AdSec Shell bump issue and dicussion with Guido, John
- M2 reflectivity dicussion and test
- 4D meeting and demontration
- Contiue working on 4D
- Continue jumping actutor issue with Guido
- Improve the magnet detach prodcedure?