Historical Log of Parkes System Changes

• 2023-10-12 :

   MASER drift adjustment, -115x10^-15 @ 11:10 AEDT
  70505 -> 70390 



• 2023-09-13 :

   MASER drift adjustment, -120x10^-15 @ 09:05 AEST
  70625 -> 70505 



• 2023-09-11 :

   UWL streams offline today (11th September) from 12:35 to approx 13:15 AEST, when they were reset.  



• 2023-09-07 :

   UWL streams offline today (7th September) from 15:40 to approx 17:30 AEST, when they were reset 



• 2023-09-05 :

   UWL streams went offline today (5th Sept.) from 13:45 through to 16:10 localtime. Any data taken within this period though be discarded. 



• 2023-09-05 :

   Reprogrammed FPGA after power outage roughly 3pm (working in racks) and resync all streams (3:50 pm) 



• 2023-08-14 :

   QSFP replaced to address loss of PolA data from subbands 0-4 https://jira.csiro.au/browse/PKSSUP-5826
  Note data taken after 01:56 UTC today should be checked for any issues. 



• 2023-08-11 :

   In an attempt to address loss of Pol A in subbands 0-4, the UWL streams were reset at ~ 4pm AEST today. 



• 2023-08-10 :

   UWL tower and digitiser streams restarted ~ 8am and again at 8:30am AEST 10th August. 



• 2023-08-10 :

   UWL digitiser streams restarted ~ 4pm AEST 9th August, tower digitisers offline. 



• 2023-07-18 :

   MASER drift adjustment, -100x10^-15 @ 09:50est
  70800 -> 70700 



• 2023-06-28 :

   MASER drift rate adjustment, -60x10^-15 @ 09:38est
  70860->70800 



• 2023-06-21 :

   MASER drift rate adjustment, -40x10^-15 @ 08:55est
  70900->70860 



• 2023-05-24 :

   MASER drift rate adjustment, +140x10^-15 @ 09:40est
  70760->70900 



• 2023-05-17 :

   Post tune MASER drift adjustment, +900x10^-15 @ 09:40
  69860->70760 



• 2023-05-17 :

   MASER auto tune started at 09:08 on 16/05/23, stopped at 21:15 on 16/05/23. 



• 2023-05-10 :

   UWL streams restarted/resync at 23:54 UTC. 



• 2023-05-05 :

   UWL streams restarted and resync at 2023-05-01 23:16:52 UTC. 



• 2023-05-04 :

   Large MASER drift rate adjustment, -100x10^-15 @ 12:50 AEST
  69960->69860 



• 2023-04-24 :

   All TOS scanning templates updated to allow fast scans (> 3 deg/min.) 



• 2023-04-18 :

   MASER drift rate adjustment, -30x10^-15 @ 08:43est
  69990->69960 



• 2023-04-17 :

   UWL streams restarted again at 12:44 localtime after focus cabin streams died at 11:50am local. 



• 2023-04-17 :

   Restarted UWL streams (and resync) at 10:00am localtime to bring the tower digitisers back online. 



• 2023-04-05 :

   SDHDF pipeline and definition updated to v3.0
  tag v3.0
  Tagger: Lawrence Toomey 
  Date:   Wed Apr 5 09:31:19 2023 +1000
  
  Release v3.0: Updates for frequency to have 2 dimensions (ndumps * nchans); Replace deprecated DataFrame.append with concat; Add missing dimension labels to frequency datasets; Update cal dataset names; Remove J2000 from coordinates; Add EQUINOX, SOURCE_LABEL, CUSTOM_, NBINS, WEIGHT_TYPE, FLAG_TYPE fields; Update datasets data_weights to weights and data_flags to flags; Update FRAME description; Update all to v3.0 



• 2023-03-30 :

   UWL focus cabin streams went offline at 16:05 localtime. Restarted at 18:05 localtime which included a resync. 



• 2023-03-28 :

   MASER drift rate adjustment, -20x10^-15, @ 14:30est
  70060->70040 



• 2023-03-24 :

   AOC cable between switch and medusa-gpu3:eth2 replaced, expecting to fix data loss issues on UWL  sub-bands 9 and 10 



• 2023-03-21 :

   MASER drift rate adjustment, -75x10^-15, @ 09:52est
  70135->70060 



• 2023-03-18 :

   The UWL streams were resynced due to dropouts and a need to restart the streams give the default setup does not automatically restart the tower digitisers. Times are Parkes localtime.
  
  ```
  uwb_server_pksobs_23_03_09.log:17:41:21) Sync: Waiting for event and JESD sync... (~10s)
  uwb_server_pksobs_23_03_09.log:17:41:31) [SUCCESS] Sync SUCCESS - Reset BAT = 5185060923500000 (0x126bc9127be5e0)
  uwb_server_pksobs_23_03_09.log:16:02:33) Sync: Waiting for event and JESD sync... (~10s)
  uwb_server_pksobs_23_03_09.log:16:02:43) [SUCCESS] Sync SUCCESS - Reset BAT = 5185573395500000 (0x126c4064320be0)
  uwb_server_pksobs_23_03_15.log:22:16:10) Sync: Waiting for event and JESD sync... (~10s)
  uwb_server_pksobs_23_03_15.log:22:16:20) [SUCCESS] Sync SUCCESS - Reset BAT = 5185595812500000 (0x126c459c5a5220)
  uwb_server_pksobs_23_03_15.log:22:53:28) Sync: Waiting for event and JESD sync... (~10s)
  uwb_server_pksobs_23_03_15.log:22:53:38) [SUCCESS] Sync SUCCESS - Reset BAT = 5185598050500000 (0x126c4621bf7da0)
  uwb_server_pksobs_23_03_15.log:23:22:46) Sync: Waiting for event and JESD sync... (~10s)
  uwb_server_pksobs_23_03_15.log:23:22:56) [SUCCESS] Sync SUCCESS - Reset BAT = 5185599808500000 (0x126c468a887120)
  uwb_server_pksobs_23_03_15.log:23:40:05) Sync: Waiting for event and JESD sync... (~10s)
  uwb_server_pksobs_23_03_15.log:23:40:15) [SUCCESS] Sync SUCCESS - Reset BAT = 5185600847500000 (0x126c46c87652e0)
  uwb_server_pksobs_23_03_15.log:23:54:02) Sync: Waiting for event and JESD sync... (~10s)
  uwb_server_pksobs_23_03_15.log:23:54:12) [SUCCESS] Sync SUCCESS - Reset BAT = 5185601684500000 (0x126c46fa59ee20)
  uwb_server_pksobs_23_03_15.log:23:54:20) Sync: Waiting for event and JESD sync... (~10s)
  uwb_server_pksobs_23_03_15.log:23:54:30) [SUCCESS] Sync SUCCESS - Reset BAT = 5185601702500000 (0x126c46fb6c96a0)
  uwb_server_pksobs_23_03_15.log:00:25:54) Sync: Waiting for event and JESD sync... (~10s)
  uwb_server_pksobs_23_03_15.log:00:26:04) [SUCCESS] Sync SUCCESS - Reset BAT = 5185603596500000 (0x126c476c50bc20)
  uwb_server_pksobs_23_03_15.log:00:55:14) [ERROR] Unknown command: resync
  uwb_server_pksobs_23_03_15.log:00:55:17) Sync: Waiting for event and JESD sync... (~10s)
  uwb_server_pksobs_23_03_15.log:00:55:27) [SUCCESS] Sync SUCCESS - Reset BAT = 5185605359500000 (0x126c47d565fae0)
  uwb_server_pksobs_23_03_16.log:01:10:11) Sync: Waiting for event and JESD sync... (~10s)
  uwb_server_pksobs_23_03_16.log:01:10:21) [SUCCESS] Sync SUCCESS - Reset BAT = 5185606253500000 (0x126c480aaf5660)
  uwb_server_pksobs_23_03_16.log:01:17:43) Sync: Waiting for event and JESD sync... (~10s)
  uwb_server_pksobs_23_03_16.log:01:17:53) [SUCCESS] Sync SUCCESS - Reset BAT = 5185606705500000 (0x126c4825a04f60)
  uwb_server_pksobs_23_03_16.log:16:29:09) Sync: Waiting for event and JESD sync... (~10s)
  uwb_server_pksobs_23_03_16.log:16:29:19) [SUCCESS] Sync SUCCESS - Reset BAT = 5185747791500000 (0x126c68ff01d2e0)
  uwb_server_pksobs_23_03_18.log:12:04:18) Sync: Waiting for event and JESD sync... (~10s)
  uwb_server_pksobs_23_03_18.log:12:04:29) [SUCCESS] Sync SUCCESS - Reset BAT = 5185818300500000 (0x126c7969abc820)
  ``` 



• 2023-03-09 :

   At 17:40pm localtime, the UWL digitisers were resynched to look at packet loss in subbands 9 and 10.
  
   



• 2023-02-28 :

   At 9:41am localtime, the UWL digitisers were resynched after we lost the streams around 9:15am localtime. 



• 2023-02-22 :

   - SSL certificates updated on tommen & medusa-srv0 



• 2023-02-07 :

   * Updated TOS Observation Procedures for Scan, Track and Drift; now using an AstroPy implementation.
  * Tested epics-ioc-metadata=3.1.2-2~rc2; changes to Metadata IOC for scanning (see https://jira.csiro.au/browse/PKSSUP-5723) 



• 2023-02-07 :

   The UWL digitisers were resynched today at 15:14 and 15:20 localtime (see https://jira.csiro.au/browse/PKSSUP-5733) 



• 2023-01-31 :

   ME switched from PDP11 based controller to new Galil controller today. 



• 2023-01-31 :

   The UWL digitisers were resynched multiple times today.
  ```
  uwb_server_pksobs_23_01_31.log:09:07:29) [SUCCESS] Sync SUCCESS - Reset BAT = 5181833281500000 (0x1268d994077360)
  uwb_server_pksobs_23_01_31.log:09:15:34) [SUCCESS] Sync SUCCESS - Reset BAT = 5181833766500000 (0x1268d9b0eff6a0)
  uwb_server_pksobs_23_01_31.log:09:27:05) [SUCCESS] Sync SUCCESS - Reset BAT = 5181834457500000 (0x1268d9da1fc960)
  uwb_server_pksobs_23_01_31.log:09:31:09) [SUCCESS] Sync SUCCESS - Reset BAT = 5181834701500000 (0x1268d9e8aaee60)
  uwb_server_pksobs_23_01_31.log:11:03:24) [SUCCESS] Sync SUCCESS - Reset BAT = 5181840236500000 (0x1268db32945420)
  uwb_server_pksobs_23_01_31.log:12:10:29) [SUCCESS] Sync SUCCESS - Reset BAT = 5181844261500000 (0x1268dc227cf460)
  ``` 



• 2023-01-31 :

   MASER drift rate adjustment of -30 in 10^15 at 1319est.
  70250->70220 



• 2023-01-30 :

   UWL/tower digitisers reset at 05:25:27 UTC today, but this proved futile given underlying fibre connection issues. 



• 2023-01-25 :

   The UWL digitisers were resynched twice today, first just prior to 2023-01-25 00:00:39 UT and then again at 2023-01-25 00:18:09 UT. Medusa was also restarted each time. 



• 2023-01-18 :

   MASER drift rate adjustment, -30x10^-15
  70280->70250 @ 0725 est 



• 2023-01-09 :

   Drift rate adjustment, -20x10^-15,
  70300->70280, @ 07:27est 



• 2023-01-01 :

   MASER drift rate adjustment, -20x10^-15,
  70320->70300, @ 1044est 



• 2022-12-22 :

   Tower digitisers were not initialised. Appears streams were turned off 19th December, 13:20 localtime. A restart was done today at 17:50 localtime with potential resync. Medusa also restarted.
  
   



• 2022-12-19 :

   Resynchronised the digitisers and turned Medusa on and off to deal with synchronisation issue as mentioned in PKSSUP-5708 (19th Dec 2022, local time 13:15 to 13:30) 



• 2022-12-13 :

   Another Drift rate adjustment (I'm trying to coax the line back towards zero)
  -30x10^-15, 70380->70350 @ 11:45 est. 



• 2022-12-07 :

   Another Drift rate adjustment (I'm trying to coax the line back towards zero)
  -20x10^-15, 70400->70380 @ 10:02 est. 



• 2022-12-05 :

   MASER drift rate adjustment of -20x10^-15, 70420->70400, @ 07:31est. 



• 2022-12-02 :

   As part of P737 on Dec 2nd morning (starting UTC: 221201_222540), we carried out tests in which the UWL digitisers were resynchronised 2 times.  At each resync, the Medusa cluster was restarted.  The tests completed at UTC 221201_231516. 



• 2022-11-29 :

   MASER drift rate adjustment of -30x10^-15, 70450->70420, @ 0943est. 



• 2022-11-23 :

   MASER drift rate adjustment of -80x10^-15, 70530->70450. @ 12:37est 



• 2022-10-19 :

   12/10/22 10:58am - (John Tuthill) reconfigure & resync UWL digitisers which stopped sampling and streaming data (possibly due to work in the tower server cabinets interrupting the clocks)
  19/10/22 10:45am - Resync of all UWB (UWL receiver and tower digitisers together) was done for VLBI to ensure alignment (tower digitisers were running 'independently' for many months) 



• 2022-10-13 :

   DHAGU v2.8 20221013
  * PKSSUP-5689: Modified the Medusa Mars configuration to mitigate a DC spike occurring on the HC5N spectral-line at the same frequency. 



• 2022-10-13 :

   MASER drift adjustment: -60x10^-15:  70662->70602, done at 10:53 est 



• 2022-09-27 :

   9/9/2022:
  A new antenna was installed on the Geoscience Australia IGS Monument.  This is also the source of the Common View timing data.
  27/9/2022:
  MASER drift adjustment of -30 in 10^15, 70692->70662 



• 2022-09-20 :

   MASER drift adjustment, -30x10^-15, 70722->70692.  Done at 0952 EST. 



• 2022-09-06 :

   MASER drift rate adjustment of -50x10^-15
  70772->70722, done at 11:10est. 



• 2022-08-31 :

   I've been advised by Geoscience Australia that yesterday (30/8/2022) they turned off data sharing from the Parkes** IGS/Common-View GPS system**.  This is due to issues with the data which they are currently investigating.  They were unaware of the CV aspect and I've advised my contact to contact Michael Wouters at NMI for discussions.  The expectation is that a new antenna will be required and they will send a team out to correct the situation.  I gained the impression that this process will take ~6 weeks to complete. 



• 2022-08-29 :

   MASER drift adjustment today at 15:10hrs, -40x10^-15, 70812->70772
  Also a catch-up from last week:
  23/8/2022, at 0929est, The Qtrs Symmetricom GPS clock was decommissioned, and a Meinberg GPS Rx'r installed in MASER Hut is now used. This machine also supplies the disciplinary 1pps for the Rb, now also installed in the MASER Hut and supplying 5MHz to the standby clock (pkclk01) via the secondary RFoF link from the MASER hut to the Upstairs Control Room.
  A second Meinberg GPs rx'r was installed today, replacing the Symmetricom GPS receiver acting as the reference to the primary Station clock.  The plots of the raw tick phase indicate the transition to the new reference occurred at around 1219hrs today (29/8/2022).  Internal delays were adjusted to approximate the tick phase of the previous clock. 



• 2022-08-28 :

   UWL attenuators were reset on the 20th August to 15, 19, 7 dB, but should have been 21, 13, 10 dB. They were reset to 21, 13, 10 dB today, 2022-08-27 03:41:22 UTC. 



• 2022-08-19 :

   SDHDF pipeline updated to v2.8
  tag v2.8
  Tagger: Lawrence Toomey 
  Date:   Thu Aug 18 12:21:08 2022 +1000
  
  Release v2.8: Fix interpolation for parameters that wrap, specifically RA_DEG and GL; Enable processing queue priority for PID P960; Improve scanning modes parameter checks 



• 2022-08-18 :

   Updated the CAL_EPOCH for the UWL cal waveform to a more recent date/time, i.e.
  Bat:  0x12561d59894340  which is ...
  UTC:  2022-06-06 12:00:00+00:00
  The new value should pass to medusa via TOS automatically. 



• 2022-08-03 :

   Digitisers re-synchronised just before UTC 2022-08-03-06:51:28 (3rd August 17:00 local time). This was required to fix a timing offset (and a noise source synchronisation failure) that occurred sometime between 220707_174036 and 220719_061012 and is described in Jira ticket PKSSUP-5644. It is possible (need to check) that the event was related to our clock free-running when it was disconnected to swap the MASER inputs to the clock, this happened at 14:28 on 19/7/2022 



• 2022-08-02 :

   29/7/2022:
  MASER Drift adjustment of +355x10^-15, 70520->70875 done at 0725est.
  2/8/2022:
  MASER drift adjustment of -90x10^-15, 70875->70785, done at 0936est 



• 2022-07-21 :

   On the 11th, 13th, 15th, and 18th, the Geoscience Australia IGS/**Common View** GPS receiver made brief excursions from its usual 260ns offset from our station clock, so Common View (one-point-per-day) data may be questionable on those days.  When the MASERs were changed over at 1424hrs on the 19th, it dropped its bundle completely, returning to normal operation at 1344hrs on 20/7/2022 after manual intervention (re-synching from Geoscience Australia) 



• 2022-07-21 :

   19/7/2022:
  The clock wasn't syncing reliably to the pkMASERb's output and it was found the 5MHz from the MASER was too low/intermittent for the RFoF system (the signal transport to the clock in the Upstairs Control Room).   The MASER was found to have a loose SMA connector at the rear of the Electronics Package.  This was rectified but an alternate 5MHz signal was sourced from the MASER.
  At 1428hrs pkMASERb was swapped onto the station clock (pkclock00) and pkMASERa was relegated to the secondary clock ('01).  At the time of change-over the old MASER had reached a tick phase of 11.31807us!
  **The resulting step change at change-over was 11.304us, from a tick phase of 11380.7ns, to 76.43333ns**.  Note that these figures are from the logged mean tick phases and the precision is a result of the calculations to obtain the mean figure. 



• 2022-07-21 :

   18/7/2022:
  With pkMASERb still acclimatising to its new surroundings, it was connected to pkclock01 and drift rate monitoring commenced 



• 2022-07-21 :

   15/7/2022, Another drift rate adjustment of the same magnitude was applied (@1526), again with no apparent change, ***the MASER is now officially out of control***. 



• 2022-07-21 :

   13/7/2022: Although the drift rate had been rained-in by controlling the temperature in the room, it had settled to a still-unacceptable rate of 130ns/day, so a coarse drift rate adjustment of 1000x10^-15 (at 0831) was attempted with no apparent change to the drift rate! 



• 2022-07-21 :

   12/7/2022 (Tuesday)
  Hydronic Heater flow pump was replaced in response to the recurring 7H fault code on the Daikin system.  Heat load removed from the hut. 



• 2022-07-21 :

   At ~0030 the following morning (Saturday) the hydronic heater failed again and the rate of frequency drift ramped up severely and was arrested (by adding a heat load in the room) by around 0914 shortly after it was discovered on Monday morning (11/7/2022).  By this time the Tick phase had grown to +1.26us! 



• 2022-07-21 :

   08/07/2022 (Friday) (catch-up)
  ~0200: Climate control in the MASER hut fails on the heating side which triggers a large change in drift rate of the MASER.  The temperature was stabilised by fitting new batteries in the two temperature sensors and resetting the hydronic heater.  MASER drift rate returned to normal. 



• 2022-07-19 :

   ```
  PORTAL/FROG/DHAGU updates:
  * PKSSUP-5631
  TOS updates:
  * python3-askap.epics_3.14.2-rc1_all.deb 
   ** Increase stow timeout from 300 to 400s
  * python3-askap.opl.script_4.6.2-rc2_all.deb
   ** PKSSUP-5623 scanning mode triggered early by other states.
   ** New translator control for Parkes.
  ``` 



• 2022-07-15 :

   The above actions were required due to a UPS outage in the cabin on 13/7/2022.  Daniel George restarted things in the cabin at the time, but history shows that the re-synch in the Upstairs Control Room was also required. 



• 2022-07-15 :

   From John Sarkissian (14/7/2022):
  This afternoon, Daniel George, performed an re-sync of the medusa boards. 
  
  Willem van Straten was observing, so I took a bit of his time so that we could do the re-sync and test observation.
  
  The attached plots show the before and after phase differential plots. The first was the last CAL Willem observed on J1141-6545_R, before the re-sync. The second plot was my test CAL on Vela, J0835-4510_R, after the re-sync. All good, now.
   



• 2022-07-11 :

   At around 1800EST on Thursday 7/7/2022 the MASER hut air-conditioner lost its heating capability and the temperature in the Hut began to fall.  At around 5am the following morning, the MASER drift rate began a positive excursion.  We got the heaters working again that morning and the temperature was again stable by around 11am at which time the MASER drift rate also stabilised back to its relatively flat trajectory.
  Unfortunately, the MASER hut heating failed again at the end of the day (Friday) and the MASER drift rate again ran away in the positive direction.  On Monday morning (11/7/2022) we discovered the situation and installed external heating in the MASER hut and called the Air-conditioning maintenance contractors for what seems likely to be a circulation pump replacement.  Temperatures were back to normal by around 10:00.  MASER drift rate appears to have settled again but is now sitting at +1400ns! 



• 2022-06-24 :

   On 2022-06-22 around 01:32:55 (UTC) the version of dspsr running on Medusa was updated.  Prior to the fix strong RFI (in particular in the low band) produced a broadband, single sample, strong signal across the sub-band on a time-scale linked with the FFT length. This was fixed by extra padding in the FFT. This would have only affected zero-DM observations (not coherently-dedispersed data) which includes most search mode data sets and folded-observations of the noise source 



• 2022-06-20 :

   TOS updates:
  ```
  python3-askap.epics/buster 3.14.1 all [upgradable from: 3.13.2]
  python3-askap.opl.script/buster 4.6.1 all [upgradable from: 4.6.0]
  tos-epics-config/buster 3.3.1 amd64 [upgradable from: 3.2.3]
  tos-ice-config/buster 3.2.0 amd64 [upgradable from: 3.1.0]
  tos-site-config/buster 3.1.0 amd64 [upgradable from: 3.0.0]
  ``` 



• 2022-06-14 :

   tag v2.7
  Tagger: Lawrence Toomey 
  Date:   Tue Jun 14 10:10:49 2022 +1000
  
  Release v2.7: Fix appending to header datasets; implement function to query schedule block metadata; add source coordinates to BeamHeaderTbl definition; SDHDF version updated to 2.7
  
  commit 9631808b42a7e607edbed3bc53c36b4d29993a3b (HEAD -> master, tag: v2.7, origin/master, fix_append_hdr)
  Author: Lawrence Toomey 
  Date:   Tue Jun 14 10:06:22 2022 +1000
  
      Update to v2.7 



• 2022-06-10 :

   DHAGU v2.1 20220610
  * SPOT template added for receiver pointing, only continuum mode is supported.
  * Removed ability to view a Parameter Set from the Scheduler tab- please use the Editor tab to view.
  * Other minor bugs fixed.
   



• 2022-06-07 :

   Drift rate adjustment made to arrest the new erratic behavior of the MASER:
  +100x10^-15, 56590->56690.  Carried out at 08:19est. 



• 2022-05-31 :

   Rearranged Euryale incoming NVME drives into mountpoints that correctly correspond with their position in the system:
   * incoming0 => incoming2
   * incoming1 => incoming3
   * incoming2 => incoming1
   * incoming3 => incoming0
  Also moved files on all filesystems so that they remain in the appropriate place. 



• 2022-05-24 :

   SDHDF pipeline updated to v2.6 



• 2022-05-17 :

   DHAGU v2.0 
  * This is a major version update to enable observing with MARS/13MM; see PKSSUP-5588.
    * Please refer to the User's Guide on observing with DHAGU and MARS/13MM.
    * All UWL Parameter Sets have been converted to a new format; please report any issues to parkes.support@csiro.au
    * It is no longer possible to import TOS Parameter Sets. CSV import remains available.
  * Fixed a bug where spectral kurtosis and the adaptive filter options where set to true; see PKSSUP-5600.
  * NoDrive observing: removed coordinate/epoch inputs. When scheduling, the current telescope position is used, with epoch of date positions provided in the data product metadata.
  * Medusa tab: for fold-mode, in addition to the integrated plots, subband plots are now displayed; see PKSSUP-5590.
  * Medusa tab: if any Medusa disk exceeds 75% capacity, the "GPU nodes" table will display a red indicator.
  * Updated Bootstrap CSS to v4.6.1
  * Other small bug fixes and cosmetics. 



• 2022-05-16 :

   Minor change to 'desk' code to correct the telescope status message when tracking in Az-El mode. See PKSSUP-5584 



• 2022-05-12 :

   A bug in DPSR was fixed, to allow for Medusa to process/display AA,BB polarisation data. This was triggered by using ```medusa.fold.output_stokes = 2``` (see PKSSUP-5598.) 



• 2022-05-06 :

   Round of system updates and reboots applied the all medusa GPU servers (Wed/Thu) 



• 2022-05-03 :

   Historical Note for the Logs: Issue identified where cross-pol data for UWL sub-bands 16, 18, 19, 20 and 21 were corrupt. Problem occurred between ~13th Feb and 26th April. 



• 2022-04-29 :

   MASER drift adjustment of -20x10^-15 made at 1642est.
  56610->56590. 



• 2022-04-29 :

   SRS Rb time standard installed in Upstairs Control Room.  Reference/steering 1pps is from the tower Symmetricom (pks ntp-2).
  The Rb based 5MHz output is now supplying the standby clock pkclk01 



• 2022-04-12 :

   TOS updates:
  ```
  askap-css-opi-parkes/buster 3.8.21 amd64 [upgradable from: 3.8.20]
  libparkes-uwb/buster 3.0.1 amd64 [upgradable from: 3.0.0-7]
  parkes-ioc-uwbrx/buster 3.1.0-3 amd64 [upgradable from: 3.1.0-2]
  python3-askap.epics/buster 3.13.2 all [upgradable from: 3.13.1]
  ``` 



• 2022-04-12 :

   UWL low/mid/high attenuators changed from 15,19, 7 dBm to 21,13,10 dBm. 



• 2022-04-08 :

   MASER drift rate adjustment, -20x10^-15
  56630->56610 



• 2022-03-23 :

   DHAGU v1.29 20220317
  * Added the NoDrive template. This allows you to observe when wind parked or stowed, with the sky passing through beam at the sidereal rate. Whilst primarily to be used for local development and testing, it could also be used for FRB monitoring.
  * Updated Users' Guide with information on NoDrive template. 



• 2022-03-23 :

   A part of the conditional test for the telescope status: "DISH HAS REACHED HORIZON LIMIT", has changed for consistency and greater reliability. See PKSSUP-5568 



• 2022-03-17 :

   Hi all, A new info alert has been submitted: On 17th March 2022 during the working day updates were made to the UWL system: (1) notch filters have been implemented which reduces the persistent transmission tower signals, (2) this had led to a change in the attenuation levels in the three RF bands, (3) the noise source has been replaced (S_cal and T_cal is similar, but not identical to the previous noise source) and (4) the reduction in strong persistent RFI in the low band implies that our monitoring plots are now more affected by transient RFI. Please submit fault reports if you notice anything unexpected. -- George Hobbs 



• 2022-03-10 :

   On 3rd March, the observatory site elevation value used in TCS and Desk was increased to 414.666m 
  See PKSSUP-5438 



• 2022-03-10 :

   DHAGU v1.28 20220310
  * PKSSUP-5558: clicking an executive (START, STOP, NEXT, ABORT) or antenna control (PARK, STOW, UNSTOW) button will now disable all buttons in the group for 5 seconds.
  * PKSSUP-5560: During a wind park, the source queue will be disabled to avoid the antenna repeatedly slewing to a source without the weather being accounted for by the observer. 



• 2022-03-03 :

   TOS update to allow scanning to continue with "ME SLEWING" message (see PKSSUP-5478.) 



• 2022-03-03 :

   tag v2.5
  Tagger: Lawrence Toomey 
  Date:   Thu Mar 3 11:55:07 2022 +1100
  
  Release v2.5: Implement parsing of observations from multiple receivers; small bug fixes and improvements 



• 2022-02-23 :

   Removed a few lines of desk code from 2007 that fudged the ME status to deal with ME tracking problem:- "ME incorrectly reports SLEWING when scanning > 7deg/m" 



• 2022-02-21 :

   MASER drift adjustment: -130x10^-15
  56760->56630, at 14:33EST during a BL slew 



• 2022-02-09 :

   On the 8th February it was discovered that polarisation 0 on sub-bands 22 and 23 were not being received by Medusa, with implication being the output from the UWL FPGAs. It was identified to have started at 2022-02-01-03:37:39 Local Time. A fix is being pursued. 



• 2022-02-07 :

   Post package-change drift adjustment, done today at 0831EST.  Full amount required is likely to be ~-185x10^-15.  Adjusted by -100 to start, i.e. 56860->56760 



• 2022-02-03 :

   The site MASER was shut down for 27 minutes yesterday  Feb 2, 2022 from 12:52 (EST).  Electronics bin was replaced, but initially at least, the IF level appears unchanged.  The site clocks were "slid" as far as possible leaving a **residual step change of -77ns** 
  before change: tickphase = -0.039us, 
  after change: tickphase = -0.116us
  These details were also added to the file here:
  /nfs/online/timing/pksMASER_clock_events.txt 



• 2022-01-31 :

   All TOS Fold and Search mode templates now have 11.1231 Hz as the default cal waveform frequency. The (prior) 62.5 Hz cal introduces a saw-tooth function to both the Stokes I and the polarization properties of the cal which adversely then affects the polarization calibration of the pulsar target.  This is likely due to an interaction with the very strong interference at 1800 MHz. The 11.1231 Hz cal does not suffer from these issues and should be preferred for pulsar calibration. Please note commensal templates such as ContinuumFold and ContinuumSearch also have 11.1231 Hz as the default. 



• 2022-01-27 :

   Catch-up from 25/1/2022: Drift adjustment of -20x10^-15
  56880->56860, made at 07:17 est 



• 2022-01-04 :

   MASER drift tweak, 0738 EST:
  -20x10^-15, 56900->56880 



• 2021-12-24 :

   *****Catch-up***** 20/12/2021, ~10:10 EST:
  MASER drift tweak of a further -10x10^-15.
  56910->56900 



• 2021-12-17 :

   Interrupted GA and CV timing again today between 0700 and 0900 EST.  Work involved disconnecting 10MHz reference to measure levels and installing pads to GA system.  Final configuration was a !0dB pad installed at the rear of the GA (Septentrio) GPS rx'r 



• 2021-12-15 :

   DHAGU v1.26 20211215
  * Removed dependency on Log4j v1.X as primary logging, moved to v2.15.0. 



• 2021-12-15 :

   Interrupted the Common-View timing system during Maintenance on 14/12/2021.  Large transients will be present in the CV data but these may be disregarded, the station clock is unchanged. 



• 2021-12-14 :

   FROG 20211214 v3.19:
  * Removed dependency of Log4J v1.X as main logging artifact; upgraded to v2.15.0.
  * Upgraded to Ice/IceStorm v3.7.2 inline with buster/python3 upgrade. 



• 2021-12-14 :

   New package for the uwbrx ioc, v3.0.1-2, released.  This is a rebuild to overcome an issue with the previous version (3.0.1-1) terminating due to a segmentation fault 



• 2021-12-14 :

   PORTAL: 20211214 (v2.27)
  * Removed Log4J v1.2.17 dependency, upgraded to v2.15.0; whilst v1.X is not directly impacted by the CVE-2021-44228 vulnerability, CVE-2019-17571 and CVE-2020-9488 are still relevant (even to the patched RedHat version), and v1.X is no longer maintained. 



• 2021-12-14 :

   Updated [ParkesLive](https://parkes-ops.atnf.csiro.au/ParkesLive/) to the Spring Boot framework and now use InfluxDB instead of MoniCA. The Atmosphere Websocket library is now replaced with Spring Server-side events. 



• 2021-12-14 :

   SDHDF pipeline updated to v2.4 



• 2021-12-13 :

   MASER drift rate tweak: 



• 2021-11-04 :

   DHAGU v1.24 20211104
  * Added the Drift scanning mode; refer to the User's Guide for more details.
  * Avoid submitting new observations to the source queue if SERVO state is not "DISH IS STATIONARY" or "STOPPED". See PKSSUP-5484. 



• 2021-10-27 :

   SDHDF pipeline updated to v2.3
  tag v2.3
  Tagger: Lawrence Toomey 
  Date:   Wed Oct 27 15:00:29 2021 +1100
  
  Release v2.3: Fix npol bug in calibration data - CONTINUUM_OUTSTOKES is incorrectly set in cal data file to 1 when NPOL=4; Fix empty metadata query logging and append to history dataset; Fix integration time out by half a dump; Fix PULSE@Parkes data processing for spectral line modules 



• 2021-10-13 :

   Minor addition to desk program relating to generation of 'DISH HAS REACHED HORIZON LIMIT' status message. See PKSSUP-5468 



• 2021-10-13 :

   SDHDF pipeline updated to v2.2
  tag v2.2
  Tagger: Lawrence Toomey 
  Date:   Wed Oct 13 08:43:30 2021 +1100
  Release v2.2: Major updates to parsing of integration timestamps for the interpolation routine; Add missing dimension labels; Add data types as strings to attributes; Replace Parkes specific timezone with LOCAL_TIME; Add extra decimal places to RA_DEG and DEC_DEG output; Add pressure, pressureMSL, relative humidity and temperature to definition and obs_params datasets; Add OSS license; Add PROC_LOG to history dataset; Update logging; Minor bug fixes and updates 



• 2021-10-05 :

   5th Oct in P737 time - updated PSRCAT database and software on joffrey (and pushed the updated database to Medusa and Dhagu).  Also updated the tempo2 software on joffrey (used by tcs). 



• 2021-09-28 :

   NVMe disk on Euryale replaced (thanks Tim and Dan!) 



• 2021-09-22 :

   Minor changes to dish status messages generated by desk program.  The 'SOURCE ... RISE' messages now only occur when servo is locked (and ME tracking). See PKSSUP-5411 and PKSSUP-5459 



• 2021-09-21 :

   PORTAL 20210921 (v2.25)
  * Effort to make PORTAL WAI-ARIA compliant.
  * Rejigged Bootstrap tabs again.
  * Improved websocket handling.
  * Reverted Atmosphere Java/JavaScript libraries to 2.7.0 and 3.1.0 respectively.
  * Removed all JavaScript map files.
  * Created minified JavaScript for home-grown libraries. 



• 2021-09-21 :

   MASER drift adjustment: -20x10^-15, 56960->56940 



• 2021-09-21 :

   DHAGU v1.21 20210921
  * Effort to make DHAGU [WAI-ARIA compliant](https://www.w3.org/TR/wai-aria/).
  * Rejigged Bootstrap tabs again.
  * Improved websocket management. 



• 2021-09-14 :

   DHAGU v1.20 20210914
  
  This release is due to reports of DHAGU being sluggish, or slow. Whilst the exact nature of the issue(s) are still unknown, some improvements have been achieved in debugging the issues reported. See PKSSUP-5441 and PKSSUP-5461.
  
  * Added JavaScript/CSS library dependancies under the Help tab. This should be included in any ticket on DHAGU.
  * Moved to a new notification scheme. Instead of a Bootstrap modal popup, notifications are done via a Bootstrap alert to the right of screen. The colour of the alert indicates severity of the issue; info, success, warning, danger. Note that danger alerts are sticky, though other alerts autohide after 5.5 seconds.
  * Removed all JavaScript map files.
  * Reverted Atmosphere Java and JavaScript libraries from v2.7.2 to v2.7.0 and v3.1.2 to v3.1.0 respectively.
  * Moved instances of setInterval()/setTimeout() to d3.interval(), which has in-built support for requestAnimationFrame().
  * Reduced polling antenna pointing and health status from 2/5 to 3/10 seconds respectively.
  * Improved handling of import and selection of project ids.
  * After profiling, reverted mathjs from v9.9.4 back to v5.4.2.
  * Reverted Bootstrap from 4.6.0 to 4.5.3.
  * Some SVG elements were being queried using jQuery, not D3js selectors.
  * Reworked element selection for status tables under the "Status" and "Observation" tabs.
  * Annotated Java schedulers with @Singleton 



• 2021-09-09 :

   Adjusted the MASER drift rate today during a telescope slew, -20x10^-15, 56980->56960.  Done at ~10:00am. 



• 2021-09-08 :

   DHAGU v1.19 20210908
  * Updated D3.js from v6.2.0 to v7.0.1.
  * Updated mathjs from v7.5.1 to v9.4.4.
  * Fixed zoom for the telescope health sunburst.
  * Increased timeout on Ajax calls to 20 seconds (was ten). 



• 2021-08-26 :

   New tacho installed on Zn2, and tacho transferred from Zn2 to Zn1, over the course of the last two days. 



• 2021-08-22 :

   PORTAL: 20210823 (v2.23):
  * For ATCA, each calendar slot now shows the array, receiver(s), and CABB configuration used (ATCASW-82).
  * Australian Public Holidays are now displayed atop the booking calendar. You can toggle visibility via the "HolidayVisibility" button.
  * Implemented schedule upload in JSON format (admin users only.)
  * Changed log4j configuration (see PKSSUP-5441).
  * Upgraded the FullCalendar library from 4.3.1 to 4.4.2 .
  * Updated Atmosphere Java and JavaScript libraries from v2.7.1 to v2.7.2 and v3.1.0 to v3.1.2 respectively.
  * Other bug fixes and cosmetics.
  
  FROG: 20210823 v3.17:
  * Updated Atmosphere Java and JavaScript libraries from v2.7.1 to v2.7.2 and v3.1.0 to v3.1.2 respectively.
  * Changed log4j configuration (see PKSSUP-5441).
  
  DHAGU: v1.17 20210818
  * Manager tab: you can now filter the Scheduling Block table for a specific id.
  * Updated Atmosphere Java and JavaScript libraries from v2.7.1 to v2.7.2 and v3.1.0 to v3.1.2 respectively.
  * Changed log4j configuration (see PKSSUP-5441). 



• 2021-08-20 :

   MASER tuned 13/8/2021, incremental adjustments followed (with the last one being yesterday) to re-acquire a shallow drift rate 



• 2021-08-19 :

   pkdesk upgraded to debian stretch (as of 17/08) 



• 2021-08-13 :

   pkdesk moved to new IP address in the online network. It will still be available on its old IP address until I can confirm nothing is using it anymore. 



• 2021-08-13 :

   Updated medusa backend switch configuration, removing the hipsr data VLAN and also some legacy configuration which was required for unicast FPGA streaming. 



• 2021-08-12 :

   Medusa GPU nodes and server updated to debian buster, and to a newer release (5.0) of the mellanox ethernet driver 



• 2021-08-12 :

   Restarted 12m drives IOC 



• 2021-08-09 :

   Y1 Ball-screw and Nut replaced, Translator initialised (using current z position) 



• 2021-08-09 :

   TOS templates updated: For UWL observing, the default cal frequency of 11.123 Hz was chosen for historical reasons from the pulsar community. In recent times, this caused problems because it was not an integral number of samples and phase issues were occurring.  The default frequency for all DHAGU-related templates is now 62.5 Hz. Please note that 11.123 Hz does not contain an integral number of microseconds, and as such long tracks will introduce phasing issues.
  
  Note that existing Parameter Sets in DHAGU have **not been changed**. 



• 2021-08-04 :

   MASER drift adjustment, -20x10^-15
  56650->56630, carried out at 08:17est 



• 2021-07-31 :

   Euryale: SDHDF pipeline updated 



• 2021-07-30 :

   SDHDF update on the 30/07/21 has changed the structure of the SDHDF files for the source and calibrator data sets to remove 1 dimension (from 5 to 4). This may impact pre-existing processing pipelines and code, with affected files SDHDF v2.1. 



• 2021-07-30 :

   27-7-2021, 08:19 EST, 
  MASER drift adjustment or -20x10^-15
  56670->56650 



• 2021-07-29 :

   Euryale: SDHDF pipeline updated 



• 2021-07-29 :

   Updated User Guide to include new 13MM gain-el curve at 23916/900 MHz. 



• 2021-07-23 :

   Euryale: SDHDF pipeline updated 



• 2021-07-21 :

   Big update: Andrew Jameson updated Medusa to normalise the spectral line/continuum data sets by the number of "hits" into each spectral channel.  The first observation in this mode is uwl_210721_020001.hdf, but there was a mistake in the code at this point. The first observation in which we believe everything is working correctly is uwl_210721_224956.hdf.   The system is now running in this mode. 



• 2021-07-13 :

   Updated User Guide to include new 13MM gain-el curve at 22387/900 MHz. 



• 2021-07-13 :

   12/6/2021, 11:00-13:00 possible elevated risk of RFI from a turbine powered cropduster operating within 500m of the 64m antenna 



• 2021-07-08 :

   A hard limit of 100 Hz as the UWL cal waveform frequency is now set in all TOS templates (was 3000 Hz.) 



• 2021-07-06 :

   Replaced QSFP2, of FPGA1, Bin #2 from the UWL FPGA Processor/Data Distributors (Control room) this morning during maintenance, to address the sub-band 11/12 problem. 



• 2021-07-02 :

   From 26th June 2021 at 00:30 (AEST)  through to the present, sub-bands 11 and 12 are missing a single pol due to a likely failure of a QFSP on the UWL - the intention is to replace this during the maintenance session on the 6th July 2021. 



• 2021-07-02 :

   MASER drift rate adjustment of -20x10^-15
  56690->56670, done at 14:51 local time, yesterday. 



• 2021-06-21 :

   Multibeam receiver removed from Parkes 2nd November 2020 



• 2021-06-17 :

   DHAGU v1.16 (hotfix)
  * Bootstrap modals cannot be dismissed by clicking the backdrop or, using the keyboard (see PKSSUP-5408). To dismiss, click the "X" button, top-right of the modal. The exception is the logger modal, which can still be dismissed by clicking on the backdrop; it will auto-dismiss after 3 seconds for non-error messages, where it will auto-dismiss after 30 seconds. 



• 2021-06-15 :

   MASER drift adjustment, -20x10^-15.
  56710->56690 



• 2021-06-15 :

   EURYALE: SDHDF updated to v2.0 



• 2021-06-15 :

   DHAGU v1.16
  * The internal PSRCAT used by DHAGU can be updated using the makePSRCAT_backends script on joffrey. Note: a logout and login to DHAGU is needed for the update to be picked up (see PUWL-205).
  * When displaying information on a scanned observation, the Azimuth and Elevation reflect the starting position of the scan. LST rise time is determined by which end-point of the scan rises last, with the LST set time determined by which end-point sets first.
  * DHAGU will now preserve input coordinate systems when viewing and saving Parameter Sets. Previously, GALACTIC coordinates were transformed to the J2000 frame as TOS currently supports J2000 and AZEL only (though see AXE-530). When submitted to TOS, GALACTIC coordinates are still transformed to J2000. A suggestion is to incorporate GALACTIC (l,b) coordinates into the source name, i.e., g0.0+0.0
  * Removed the option to select a great or small-circle scan. Scanning behavior defaults to a small circle if (and only if) lat1 === lat2, otherwise a great circle scan is performed. This is a reversion of functionality introduced in v1.13.
  * Fixed a bug where user-selected subbands were not labeled correctly under the Medusa  tab.
  * Code cleanup, refactoring of JavaScript classes, moved non getter/setter methods into prototype declarations.
  * Outside of the main ephemeris JavaScript library, removed support for B1950. 



• 2021-06-01 :

   All TOS Track templates updated.
  * If the source rises in the next 10 minutes:
    * If slew time < rise time: the antenna will slew to its approximate Azimuth of rising, and wait at the horizon for the source to rise.
    * If slew time > rise time; the antenna slews to the approximate position the source will be at slew completion.
  * If the slew time exceeds the time a source remains visible (i.e., it sets during the slew), the observation will abort. 



• 2021-05-26 :

   Parkes Users' Guide: updated 13MM Moon Tsys/Tcal plots. 



• 2021-05-25 :

   UWL ADC setup + resync following Brett's work (2021-05-25 01:50 UTC) 



• 2021-05-19 :

   EURYALE: updates to SDHDF pipeline 



• 2021-05-18 :

   DHAGU minor release:
  * Fixed a bug in the calculation of Continuum data rates.
  * Fixed a bug where CSV import with GALATIC coordinates did not generate LST rise/set information. 



• 2021-05-18 :

   Parkes Users' Guide updated: Improved Moon Tsys/Tcal plots, plus removed H-OH (superseded by UWL.) 



• 2021-05-11 :

   Added a check in the desk code for single spurious 'wswait' values sent from servo to avoid setting the WSTOW flag in error. 



• 2021-05-11 :

   DHAGU v1.15
  * Updated PSRCAT database. Modified PSRCAT and ICRF source imports for the editor table. Typical calibration sources such as J0437-4715, B1934-638 now now prefetched at initialisation into the Bloodhound (suggestion) engine. The combined ICRF/PSRCAT catalogue is treated as remote data and will only be called if the source is not in the prefetched data.
  * Upgrade to jQuery 3.6.0 (security fix), required alterations for jQuery upgrade.
  * Alterations to the editor table, better use of tooltips to display errors.
  * Removed ```window.stop()``` in browser detection; this stopped mobile devices loading the application, even if they are recent versions.
  * Other minor cosmetics and bug fixes. 



• 2021-05-11 :

   MASER drift adjustment: -30x10^-15, 56740->56710 



• 2021-05-03 :

   30/4/2021, 06:45 EST:  Damage to fibre-optic cables caused by rodents interrupted the signal path between the MASER and the Station Clock (pkclk00).  It took around 5 hours to recognise there was a problem, locate it, and affect a repair (refer pkssup-5393).  While unlocked, the station clock drifted ~15ms (rendering it I suspect pretty much useless during that time), but once repaired John Reynolds was able to slide the clock, returning it to the same 200ns cycle as before.
  Special thanks to Tim and John for their quick and effective response. 



• 2021-04-14 :

   Adjusted MASER drift rate by -20x10^-15, 56780->56760 (at 15:40est) 



• 2021-04-14 :

   13/4/2021:
  Turned off most of the equipment in rack i, primarily the roach boards used for the now decommissioned Multibeam receiver. 
   The CASS Network switch, Swinbourne switch, and "hipsr-server0" machines were all left running for the time being. 



• 2021-04-07 :

   DHAGU v1.13
  * For ScanMap templates, DHAGU will only allow editing of the scan mode (great/small) via the "Edit template" button (see PKSSUP-5384).
  * Addition of scanMode parameter to all ScanMap* templates (PKSSUP-5384).
  * Python indentation issue fixed for ScanContinuum templates (PKSSUP-5384).
  * Fixed an issue where it was not possible to abort the observing queue (PKSSUP-5385). 



• 2021-04-06 :

   DHAGU v1.12
  * For scanning, it is now possible to set the scan as a small or great-circle scan. Previously, any scan where lat1 === lat2 (exact) would implement a small-circle scan (consistent with the TCS implementation), otherwise a great-circle scan is performed. Refer to the Users' Guide for more information on which mode to select.
  * If the antenna is executing a DESK related action (i.e., PARK/STOW/UNSTOW), pressing the "Abort" button will inform the observer the DESK operation cannot be aborted (PUWL-281).
  * DHAGU can now handle MEDUSA plots from non-UWL receivers.
  * Added extra information to errors- the HTTP status code and text, plus a message relating to the action being attempted.
  * Updated JSONEditor JavaScript library from 2.4.0 to 2.5.3
  * Updated Atmosphere JavaScript/Java libraries from 3.0.6/2.6.4 to 3.1.0/2.7.0.
  * Other minor bug fixes. 



• 2021-03-12 :

   11/3/2021, 02:28 UT:
  Station secondary clock reference standard (Symmetricom GPS clock 
  ntp3-pks) was reset causing an apparent blip in the backup station timing (pkclk01), however timing continuity was unaffected. 



• 2021-03-12 :

   8/3/2021, 01:56 UT: 
  Station clock reference standard (Symmetricom GPS clock 
  ntp2-pks) was reset causing an apparent blip in station timing, however timing continuity was unaffected.
  **Weirdly though, power cycling the Symmetricom GPS time reference seems to have introduced an -7ns offset!** 



• 2021-03-01 :

   Parkes/ATCA SOC machines pyxis/ara both run Debian stretch, with v78esr versions of Firefox installed. Both PORTAL and DHAGU now recognise this as a supported version, and no longer provide a popup at login. Previously, only v82+ was supported. The Users' Guide has been updated to reflect this. 



• 2021-02-25 :

   DHAGU v1.11
  
  * Changed DHAGU queue dead-zone Zenith limits (PKSSUP-5343) to (0.986,1.12) => (0.95,1.5).
  * Fixed CSV/TOS import logic to setup DataTable for selected template selection.
  * Updated DataTables JavaScript library to v1.10.23
  * Minor bug fixes and cosmetics. 



• 2021-02-24 :

   24th Feb: Andrew Jameson and George Hobbs updated the automatic flag commands used when making the Medusa online plots.  This should make the online plots less affected by RFI and should not have affected the output data files in any way. 



• 2021-02-19 :

   Removed unused devices: MB Rx and MB Cable Equaliser, from pkmc gui. 



• 2021-02-18 :

   MASER drift rate adjusted, 56800->56780, at 12:04 summer time. 



• 2021-02-18 :

   DHAGU v1.10 20210218
  * Previously, if a source was below the horizon, the Scheduling Block would transition to the ERRORED state. If a source (or starting position for a scan) is due to rise within the next 10 minutes, the antenna will wait at its current position, then slew to the source once it is above the antenna horizon. A solution where the antenna slews to the Azimuth of rising whilst the source is still below the horizon, is being worked on.
  * Editor tab: When importing a TOS template, the import tool now checks user-supplied parameters against the TOS template (TrackFold, ScanMapContinuum, ...; PUWL-243). Deprecated or invalid keywords are displayed if present.
  * Editor tab: added ability to add comments to CSV files.
  * Editor tab: better reporting when entering sources manually. A tooltip will appear above an input indicating an issue (i.e., a scan end point exceeds antenna limits, etc.).
  * Fixed track and scanning in AZEL mode. Note the ScanMap templates cannot be done in AZEL mode.
  * Fixed an issue in calculating data rates, plus other minor bug fixes and cosmetics. 



• 2021-02-10 :

   EURYALE: updates to SDHDF processing pipeline 



• 2021-02-09 :

   MASER drift rate adjustment, -20x10^-15, 56820->56800, done at 12:02 EDT 



• 2021-02-05 :

   EURYALE: bug fixes to pulsar search-mode pipeline 



• 2021-02-03 :

   DHAGU v1.10 20210203
  * Fixed track and scanning in AZEL mode. Note the ScanMap templates cannot be done in AZEL mode.
   



• 2021-01-29 :

   On 28th January system and TOS packages were updated on Octomore. 



• 2021-01-28 :

   DHAGU v1.9:
  
  * There is now an Unstow-ExLim button in DHAGU. This unstows the antenna to a Zenith Angle = 1.5 degrees (PUWL-276). If you enter an Azimuth CW or CCW limit, this should also drive the antenna out of the hardware limit.
  * As per the current UWL/Medusa [data rate policy](https://www.parkes.atnf.csiro.au/observing/documentation/users_guide/html/pkug.html#Data-rates-and-volume-policies),  if you exceed a rate of 0.5 TB per hour, an email is sent to data management for review (PUWL-207).
  * The table under the "Manager" tab will show the executive duration if present (i.e., the Scheduling Block was executed), otherwise show the source duration calculated from src%d parameters. 



• 2021-01-28 :

   MASER drift rate tweak by -20x10^-15, 56840->56820, done at 0710 est. 



• 2021-01-25 :

   In relation to the previous issue of dropped packets reported for MEDUSA (see update from 26/12/2020), on Friday 22/01/2021 the issue was investigated and 3 QFSPs were replaced in chassis number 1 - this resulted in full restoration of package transfer. 



• 2021-01-22 :

   Improved Medusa histograms on the Stats monitoring page to zoom in and show more useful details 



• 2021-01-22 :

   Improved the Min/Mean/Max plotting display for the Medusa timeseries monitoring plots. Fixed bug with Medusa bandpass monitoring plots for Pol AA. 



• 2021-01-19 :

   Updated all TOS ScanMap* templates to include the scanmap.* parameters in ObsUserParameters, which are used to generate individual scans in a map and could be useful metadata when processing. 



• 2021-01-19 :

   DHAGU v1.8
  * Fixed a bug where ScanMap templates were not run due to unknown keys in the Parameter Set (PKSSUP-5363). The unknown keys were part of the schema used to generate the start/end/separation and other components for each scan in the map. As a result, all parameters used to define maps are now included in ObsUserParmeters.
  * To reduce forced downloads at login, the ephemeris, sat-TLE and atnf-specific JavaScript libraries are moved into a separate directory; the Spring filter will enable a force-reload for these and not other JavaScript  3rd party libraries which are (mostly) static. 



• 2020-12-26 :

   Dropped packets have been identified into MEDUSA:   
  
  4.096 Gb/s from sub-band #13 (2432 MHz) This seemed to start at 2020-12-05-06:13:52 AEDT.
   ~1.6 Gb/s from sub-band #20 (3328 MHz) This seemed to start at 2020-12-02-01:54:15 AEDT
   ~1.68 Gb/s from sub-band #21 (3456 MHz) This seemed to start at 2020-12-02-01:55:07 AEDT
  
  In all of these cases, the data drops started as a trickle, then transitioned up to the rates quoted above. #13 is likely to be missing a polarisation, not clear for bands #20, #21.  



• 2020-12-23 :

   MASER drift rate adjustment, -20x10^-15, 56860->56840.  Done at 10:38 local summer time. 



• 2020-12-16 :

   We have identified a few issues relating to two-bit pulsar search mode data sets.  We have not fixed any of the issues yet, but for notice:
  
  1) The threshold levels are set in a linear fashion (0, 1, 2, 3).  However, dspsr reads them assuming that they are nonlinear (-2, -0.5, 0.5, 2)
  
  2) The 2-bit values are set based on samples in a specific sub-integration block.  We know that broad-band "RFI" is present (coming from digitizer artifacts relating to the 1800 MHz transmission tower RFI).  As the levels are set independently per block there are low level baseline variations in the output time series, which significantly reduces the data set sensitivity to long-period pulsars.
  
  Separately, we found that Medusa baseband recording can silently fail and lose packets. A log is available that indicates such failures, but that log is not accessible by general observers.
  
  We are working on solutions to these issues.
   



• 2020-12-16 :

   DHAGU v1.7 20201215
  * Fixed a bug in calculating TB/hr in the editor, displayed when you click on "Show parameters".
  * Upgraded Gradle build system from 2.10 to 6.7.1
  
  Users' Guide:
  * Added info on the "NEXT" button in DHAGU- allows the observer to skip to the next source in the observing queue (if any.) Will also transition Scheduling Block from EXECUTING to OBSERVED (Section 3.2.2.1: Control).
  * Clarified DHAGU/TOS will unstow the telescope if it is already stowed (Section 3.3. (Un)Stowing and parking.) 



• 2020-12-09 :

   MASER drift rate tweaked to promote a downward trend, -20x10^-15: 56880->56860. Completed at 08:28EST. 



• 2020-12-01 :

   MASER drift rate adjusted by -30 x10^-15, 56910->56880.  Done at 11:28 EST 



• 2020-12-01 :

   DHAGU v1.6 20201201
  * Observation tab: a new button "Next" is added to the Queue control (PUWL-215). When pressed, the currently executing Scheduling Block will be aborted, and then automatically transitioned to OBSERVED. You may want to use this feature if your source has reached sufficient S/N, and continue on to the next source in the queue.
  * Other minor cosmetics and bugs fixed.
   



• 2020-11-24 :

   TOS package update:
  ```
  mar637@octomore:~$ sudo apt-get install askap-css-opi-parkes askapioc-tools epics-ioc-metadata libaskapioc parkes-ioc-pksccu python-askap.cli python-askap.epics python-askap.opl python-askap.opl.script python-askap.slice tos-epics-config  parkes-ioc-pksmon parkes-ioc-uwbrx
  Reading package lists... Done
  Building dependency tree
  Reading state information... Done
  The following held packages will be changed:
    askap-css-opi-parkes askapioc-tools epics-ioc-metadata libaskapioc parkes-ioc-pksccu python-askap.cli python-askap.epics python-askap.opl python-askap.opl.script
    python-askap.slice tos-epics-config
  The following packages will be upgraded:
    askap-css-opi-parkes askapioc-tools epics-ioc-metadata libaskapioc parkes-ioc-pksccu parkes-ioc-pksmon parkes-ioc-uwbrx python-askap.cli python-askap.epics python-askap.opl
    python-askap.opl.script python-askap.slice tos-epics-config
  13 upgraded, 0 newly installed, 0 to remove and 13 not upgraded.
  Need to get 723 kB of archives.
  After this operation, 11.3 kB of additional disk space will be used.
  Do you want to continue? [Y/n]
  Get:1 http://debian-archive.atnf.csiro.au/aptly/TOS stretch/main amd64 askap-css-opi-parkes amd64 3.7.13 [25.5 kB]
  Get:2 http://debian-archive.atnf.csiro.au/aptly/TOS stretch/main amd64 askapioc-tools amd64 4.0.0 [8,998 B]
  Get:3 http://debian-archive.atnf.csiro.au/aptly/TOS stretch/main amd64 tos-epics-config amd64 3.2.2 [3,056 B]
  Get:4 http://debian-archive.atnf.csiro.au/aptly/TOS stretch/main amd64 libaskapioc amd64 4.0.0 [129 kB]
  Get:5 http://debian-archive.atnf.csiro.au/aptly/TOS stretch/main amd64 epics-ioc-metadata amd64 3.1.2-2 [159 kB]
  Get:6 http://debian-archive.atnf.csiro.au/aptly/TOS stretch/main amd64 parkes-ioc-pksccu amd64 3.0.1-6 [52.9 kB]
  Get:7 http://debian-archive.atnf.csiro.au/aptly/TOS stretch/main amd64 parkes-ioc-pksmon amd64 3.0.2-4 [54.7 kB]
  Get:8 http://debian-archive.atnf.csiro.au/aptly/TOS stretch/main amd64 parkes-ioc-uwbrx amd64 3.1.0-1 [104 kB]
  Get:9 http://debian-archive.atnf.csiro.au/aptly/TOS stretch/main amd64 python-askap.cli all 3.2.6 [19.1 kB]
  Get:10 http://debian-archive.atnf.csiro.au/aptly/TOS stretch/main amd64 python-askap.epics all 3.7.0 [78.5 kB]
  Get:11 http://debian-archive.atnf.csiro.au/aptly/TOS stretch/main amd64 python-askap.slice all 4.2.1 [51.6 kB]
  Get:12 http://debian-archive.atnf.csiro.au/aptly/TOS stretch/main amd64 python-askap.opl all 4.1.0 [12.5 kB]
  Get:13 http://debian-archive.atnf.csiro.au/aptly/TOS stretch/main amd64 python-askap.opl.script all 4.3.0 [24.1 kB]
  Fetched 723 kB in 0s (8,416 kB/s)
  Reading changelogs... Done
  (Reading database ... 151950 files and directories currently installed.)
  Preparing to unpack .../00-askap-css-opi-parkes_3.7.13_amd64.deb ...
  Unpacking askap-css-opi-parkes (3.7.13) over (3.7.10) ...
  Preparing to unpack .../01-askapioc-tools_4.0.0_amd64.deb ...
  Unpacking askapioc-tools (4.0.0) over (3.3.2) ...
  Preparing to unpack .../02-tos-epics-config_3.2.2_amd64.deb ...
  Unpacking tos-epics-config (3.2.2) over (3.2.1) ...
  Preparing to unpack .../03-libaskapioc_4.0.0_amd64.deb ...
  Unpacking libaskapioc (4.0.0) over (3.3.2) ...
  Preparing to unpack .../04-epics-ioc-metadata_3.1.2-2_amd64.deb ...
  Unpacking epics-ioc-metadata (3.1.2-2) over (3.1.2-1) ...
  Preparing to unpack .../05-parkes-ioc-pksccu_3.0.1-6_amd64.deb ...
  Unpacking parkes-ioc-pksccu (3.0.1-6) over (3.0.1-4) ...
  Preparing to unpack .../06-parkes-ioc-pksmon_3.0.2-4_amd64.deb ...
  Unpacking parkes-ioc-pksmon (3.0.2-4) over (3.0.2-3) ...
  Preparing to unpack .../07-parkes-ioc-uwbrx_3.1.0-1_amd64.deb ...
  Unpacking parkes-ioc-uwbrx (3.1.0-1) over (3.1.0) ...
  Preparing to unpack .../08-python-askap.cli_3.2.6_all.deb ...
  Unpacking python-askap.cli (3.2.6) over (3.2.2) ...
  Preparing to unpack .../09-python-askap.epics_3.7.0_all.deb ...
  E: namespace:121: cannot remove /usr/lib/python2.7/dist-packages/askap/__init__.py
  Unpacking python-askap.epics (3.7.0) over (3.6.7~rc1) ...
  Preparing to unpack .../10-python-askap.slice_4.2.1_all.deb ...
  E: namespace:121: cannot remove /usr/lib/python2.7/dist-packages/askap/__init__.py
  Unpacking python-askap.slice (4.2.1) over (4.1.3) ...
  Preparing to unpack .../11-python-askap.opl_4.1.0_all.deb ...
  E: namespace:121: cannot remove /usr/lib/python2.7/dist-packages/askap/__init__.py
  Unpacking python-askap.opl (4.1.0) over (4.0.4) ...
  Preparing to unpack .../12-python-askap.opl.script_4.3.0_all.deb ...
  E: namespace:121: cannot remove /usr/lib/python2.7/dist-packages/askap/opl/__init__.py
  E: namespace:121: cannot remove /usr/lib/python2.7/dist-packages/askap/__init__.py
  Unpacking python-askap.opl.script (4.3.0) over (4.1.3) ...
  Setting up python-askap.slice (4.2.1) ...
  Setting up python-askap.epics (3.7.0) ...
  Setting up askap-css-opi-parkes (3.7.13) ...
  Setting up askapioc-tools (4.0.0) ...
  Setting up tos-epics-config (3.2.2) ...
  Installing new version of config file /etc/askap/sites/vm/epicsenv.sh ...
  Processing triggers for libc-bin (2.24-11+deb9u4) ...
  Setting up libaskapioc (4.0.0) ...
  Setting up python-askap.cli (3.2.6) ...
  Setting up python-askap.opl (4.1.0) ...
  Setting up parkes-ioc-pksccu (3.0.1-6) ...
  Setting up python-askap.opl.script (4.3.0) ...
  Setting up parkes-ioc-uwbrx (3.1.0-1) ...
  Setting up parkes-ioc-pksmon (3.0.2-4) ...
  Setting up epics-ioc-metadata (3.1.2-2) ...
  Processing triggers for libc-bin (2.24-11+deb9u4) ...
  [ Rootkit Hunter version 1.4.2 ]
  File updated: searched for 179 files, found 145
  ``` 



• 2020-11-24 :

   DHAGU v1.5 20201124
  * Editor: It is now possible to add a comment relating to a source in the editor table. When the Parameter Set is loaded into the scheduler, you can toggle comment visibility using the "Hide/Show comment" button above the table. Note that sources with a comment will have a green "plus" indicator, which goes to a red "minus" when the source comment is visible.
  * Other small bug fixes and cosmetics. 



• 2020-11-18 :

   EURYALE: all pulsar processing software updated during the maintenance period 



• 2020-11-17 :

   DHAGU v1.4, main changes (see the DHAGU "Help" tab for all):
  
  * Unsupported browsers will now invoke a popup at login. See 'Web browser support' in the Users' Guide for more details.
  * All server requests now have a 10 second timeout (previous versions had unlimited timeout); a "timeout" popup will ensue if the request is not completed within the timeout. If you see repeated timeout popups, there may be an issue with your connection to the Parkes servers. 



• 2020-11-11 :

   Connected the new Geoscience Australia GPS receiver (replacing their Tate Rx'r) today and transferred the common View receiver to use the same antenna but using a new splitter in the new GA installation.  An additional delay of approx 40ns was noted.  Refer entry from 9/10/2020. 



• 2020-11-11 :

   Cal Control Unit control/monitor software moved to host pkdesk. It now runs within process 'pkmcServer00' under systemctl. 



• 2020-11-10 :

   pkclk00 updated to debian stretch 



• 2020-11-06 :

   Euryale updated to debian buster from debian stretch 



• 2020-11-06 :

   pkclk01 updated to debian stretch version of AT clock software + driver 



• 2020-10-26 :

   SSD failure on medusa-gpu2 has meant that Medusa has been reconfigured to process 4 sub-bands per server (skipping medusa-gpu2) whilst we wait for a hard disk replacement 



• 2020-10-23 :

   DHAGU v1.3.0 20201021
  * Editor: fix a bug which stopped saving to an existing Parameter Set.
  * Editor: overhauled scanning implementation; small/great circle scans now determined internally for both Scan* and ScanMap* templates.
  * Optimizations to D3js rendering.
  * Minor JavaScript cleanup, gradual move to ES6.
   



• 2020-10-20 :

   PORTAL v2.20:
  * ATCASW-79: DA should see on-call info.
  * ATCASW-80: Cannot certify people as observers or DAs
  * Fixed a bug in parsing dates when dealing with Green Time requests.
  * Fixed a bug with batch booking, where you could book for a slot multiple times.
  * Fixed text colour for PTF calendar events. 



• 2020-10-20 :

   DHAGU v1.2.0:
  * Editor: removed option to select small/great scans; DHAGU will set small-circle scans if lat1 = lat2 (the TCS convention.) The option to select great or small-circle scans remains for ScanMap templates.
  * Editor: saving Parameter Sets is more robust; a check is now made to see if the provided alias/name already exists and if so, prompt to continue (overwrite) or cancel (allow to rename).
  * Disabled server logging. When multiple users were logged on, excessive disk logging may have caused performance/responsiveness issues.
  * Drive time incorrect: PKSSUP-5332
  * Fixed login issue where incorrect username/password message was not generated. 



• 2020-10-09 :

   56880->56910 



• 2020-10-01 :

   1/10/2020, drift improving, try -230x10^-15
  57110->56880 (adjusted 08:56 est) 



• 2020-09-30 :

   Better leave the copy on joffrey for now 



• 2020-09-30 :

   The pkmc M/C software for the SwitchMatrix has been moved to pkdesk.  It runs under atnf-pkmcserver.service, (together with the legacy receivers and MBCE), and can be restarted with  sudo systemctl restart atnf-pkmcserver.service.  The SwitchMatrix host has been updated in the TCS ini file. 



• 2020-09-30 :

   Various old 12m PAF related infrastructure powered down and removed from 12m pedestal cabinet. 



• 2020-09-28 :

   * FROG v3.15
   * Moved the synoptic options into two dropdowns, as satellites were causing scroll issues.
   * Atmosphere Java now v2.6.3 and JavaScript v3.0.6; primary connection is now via long-polling, with fallback to websocket.
   * Updated jQuery to v3.5.1 .
   * Other minor cosmetics and code cleanup.
  
  * PORTAL v2.18.0
   * Primary connection is now via long-polling, with fallback to websocket. 
   * Updated jQuery to v3.5.1 
   * Other minor cosmetics and code cleanup.
  
  * DHAGU v1.0.1
   * When saving, prompt the user to continue (overwrite) or abort (rename) if the Parameter Set being edited    
  already exists. 
   * Added current OIC check for removing/adding/reordering source to the Pending list.
   * Changed from websocket to long-polling as the default transport mechanism.
   * Atmosphere no longer removes session timeout, long-polling should keep sessions active.
   



• 2020-09-25 :

   SDHDF definition updated to v1.9.3 



• 2020-09-23 :

   * DHAGU updated to v1.0.0 (see Help tab for changes.) Parameter Sets for all projects are in the process of being migrated over.
  * Parkes Users' Guide updated . 



• 2020-09-23 :

   TOS and system package update (minus kernel)
  ```
  The following packages have been kept back:
    askap-css-opi-parkes askapioc-tools carepeater epics-ioc-metadata epics7-bin
    libaskapioc libasyn42 libautosave libcaclient libepics7 libiocadmin
    libstreamdevice linux-image-amd64 parkes-ioc-pksccu parkes-ioc-pksmon
    parkes-ioc-uwbrx procserv python-askap.cli python-askap.epics
    python-askap.executive python-askap.iceutils python-askap.logarchiver
    python-askap.opl python-askap.opl.script python-askap.slice tos-epics-config
  The following packages will be upgraded:
    base-files ca-certificates cups-bsd cups-client cups-common dbus
    dbus-user-session dbus-x11 debian-security-support glib-networking
    glib-networking-common glib-networking-services gnutls-bin
    libaskap-epics-logging libcups2 libcupsimage2 libcurl3 libcurl3-gnutls
    libcurl4-openssl-dev libdbus-1-3 libdns-export162 libexif12 libgnutls-dane0
    libgnutls30 libirs-export141 libisc-export160 libisccc-export140 libisccc140
    libisccfg-export140 libjpeg-dev libjpeg62-turbo libjpeg62-turbo-dev
    libmariadbclient18 libperl5.24 libproxy1v5 libpython2.7 libpython2.7-dev
    libpython2.7-minimal libpython2.7-stdlib librsvg2-2 librsvg2-common
    libruby2.3 libsnmp-base libsnmp-dev libsnmp30 libxml2 linux-libc-dev
    linux-perf linux-perf-4.9 linux-tools mariadb-client-10.1
    mariadb-client-core-10.1 mariadb-common mariadb-server-10.1
    mariadb-server-core-10.1 nfs-common nfs-kernel-server openjdk-8-jdk
    openjdk-8-jdk-headless openjdk-8-jre openjdk-8-jre-headless perl perl-base
    perl-modules-5.24 python-pip-whl python2.7 python2.7-dev python2.7-minimal
    rake ruby2.3 tzdata xserver-common xserver-xorg-core
  ``` 



• 2020-09-09 :

   MASER drift adjustment: -20x10^-15
  56630->56610.  Occurred at 12:43 EST. 



• 2020-08-31 :

   MASER drift adjustment of -20x10^-15  56650->56630.
  change made at 0835 local time (EST) 



• 2020-08-18 :

   https://confluence.csiro.au/display/CASSPKSDATA/SDHDF+-+Single+Dish+Hierarchical+Data+Format 



• 2020-08-18 :

   SDHDF updated to v1.9.2. Updates include the implementation of storage of the Medusa 32-bin calibrator signal. Affective from and including UTC 2020-08-10-01:28:44 



• 2020-07-21 :

   Routine MASER drift rate adjustment of -20x10^-15, 56690->56670
  completed at 08:26 EST 



• 2020-07-16 :

   MB Cable Equaliser controller removed from host pkicc01 



• 2020-07-07 :

   ATCS version 'next' (1.142) updated. TCS connection to MB Cable Equaliser needs to be checked at some stage.  There is also a new warning message about TOS and optional bail-out when tcs gui starts.  This is for PKSSUP-5274 'Unknown observer started TCS whilst DHAGU was in operation'. 



• 2020-07-07 :

   New instance of MB Cable Equaliser controller added to PKMC server on pkdesk (pkmcServer00). Old instance on pkicc01 is still running pending a re-build and install of ATCS. 



• 2020-07-06 :

   Parkes PORTAL updated to v2.12.
  
  * Upgraded Twitter Bootstrap from v3.3.7 to v4.5.0, also using Font Awesome icon library.
  * Upgraded Atmosphere Java runtime from v2.6.0 to v2.6.1 and JavaScript client from v3.0.4 to v3.0.5. 



• 2020-07-03 :

   3/7/2020, Weather station auxiliary charger turned off (back to solar only) 



• 2020-07-03 :

   Catch-up:
  22/6/2020, Weather station auxiliary charger turned on for the first time 



• 2020-07-01 :

   New version of CSS gui has been installed on joffrey now. @stacym it just needs to be restarted 



• 2020-07-01 :

   Extra information outputs added to the UWL cal waveform generator (EVT IOC) ... Calculated period, pulseWidth and phase offset; Actual waveform start time as hex BAT and UT.  Also, epoch can now be entered manually via the CSS gui as hex BAT or UT string.
   



• 2020-06-23 :

   MASER drift rate adjustment, -20x10^-15, 56710->56690.  Adjustment made at 13:51 est, during maintenance. 



• 2020-06-19 :

   On 2020-06-16 @mal_smith_1719 and @sean.severs installed an improved bias card (V1-1-UWL) for the UWL PolA LNA1. It fixes the long term UWL 'bias good' issue (PKSSUP-5244) and has better temperature stability than the old design. The rest of the bias cards will be updated soon (July?). 



• 2020-06-10 :

   Fixed a bug in the Parkes Online Sensitivity Calculator where the beam for the UWL was fixed across the band. The beam (for all receivers) is now calculated based on input frequency. 



• 2020-06-09 :

   Upgraded OS of tommen (online web VM) from debian jessie to debian stretch 



• 2020-06-09 :

   Small tweak of MASER drift rate, -20x10^-15, 56730->56710, done at ~0930 EST. 



• 2020-05-19 :

   Security updates installed on:
   - medusa-srv0
   - medusa-gpu[0-8] 



• 2020-05-19 :

   HA Cluster restart, all VMS restarted (& software updates installed):
   - joffrey
   - myrcella
   - dalwhinnie
   - tyrion
   - tommen
   - octomore
   - dalmore 



• 2020-05-13 :

   updated metadata IOC with pre-release to address PKSSUP-5289
  `dpkg -i epics-ioc-metadata_3.1.1~pre1_amd64.deb` 



• 2020-05-13 :

   MASER drift adjustment of -30x10^-15, 5678->56750, executed at 10:33 on 12 May 2020 local time. 



• 2020-05-12 :

   Updated kernel on pkdesk from 3.16.x to 4.9.x and rebuilt eventgen kernel driver to match 



• 2020-05-04 :

   DHAGU v0.9.2 2020504
  
  * Fixed a bug where coherent dedispersion and DM were not populated for ScanMap Search-related templates; these parameters live under "Parameters" on the ScanMap modal.
  * Re-jigged the "Table Global Change" section under the Editor table. Added coherent dedispersion and DM options, plus only one frequency selection for all options.
  * Other minor bugs and cosmetics. 



• 2020-05-01 :

   DHAGU v0.9.1 2020501
  
  * When the link to a Scheduling Block is clicked (i.e., from the Manager tab), there is now an Errors tab. This tab displays WARN/ERROR/FATAL messages associated with a Scheduling Block. It also includes errors in ObsVariables, which may not be displayed in real-time logs. 



• 2020-04-30 :

   DHAGU v0.9.0 20200430
  
  * Search mode parameters custom_dm (dispersion measure) and co_dedisp (coherent dedispersion) are now configurable at the source level. Prior to this release, these parameters were only configurable at the subband level. The change in parameters is as follows:
  
  ```medusa.search.subband%d.custom_dm => common.target.src%d.search_custom_dm```
  
  ``` medusa.search.subband%d.co_dedisp => common.target.src%d.search_co_dedisp```
  
  * Updated Atmosphere Framework runtime jar 2.5.12 -> 2.5.14 and JavaScript 2.3.8 -> 3.0.0. 



• 2020-04-28 :

   DHAGU v0.8.3 2020427
  
  * Fixed a bug with TOS import; if a Parameter Set contains a Medusa mode that is not in the selected template, it is removed (with warning) from the import.
  * Removed source visibility from the Editor table; relevant only when source was loaded in table.
  * Modified classes extending the base Star() class; in calculation of ephemeris, assume the default time of calculation is the current UTC. 



• 2020-04-22 :

   Yesterday's error corrected, final (overall, including yesterday) adjustment was -20x10^-15, 56800->55870->56870.  Adjustment made between observing programmes, during a first slew, at 10:00est. 



• 2020-04-22 :

   DHAGU v0.8.0 20200422
  
  * Added Scan. It is now possible to define individual scans using a central position, scan length and rotation (position) angle. Check the User Guide for details.
  * ScanMap: fixed a bug where the map size exceeded defined limits. Also fixed a bug where Tobs was not populated.
  * Other minor bug fixes and cosmetics.
  
  Updated User's Guide: now includes receiver parameters, if available (Tsys, Tcal, gain-elevation curves, ...). Also updated sections on creating ScanMap and Scan parameter sets. 



• 2020-04-15 :

   DHAGU releases
  
  v0.7.0 2020415
  * **Removed positional offsets**. This has implications for existing parameter sets and importation of CSV/TOS files.
  * For scanning, the scheduler table now displays both start and end positions; horizontal scrolling is also enabled to display content.
  * Upgraded bootstrap-select from v1.3.9 to v1.3.12.
  * Minor bug fixes and cosmetics.
  
  v0.6.0 20200408:
  * The ScanMap scheduler has been reworked. It is now possible to specify both square and rectangular regions, plus apply a rotation angle to the map. The functionality to start scanning at a cardinal point (NE, NW, SE or SW), has been removed, but this can be compensated with the rotation functionality.
   



• 2020-03-30 :

    All current TOS templates have been updated with a maximum cal frequency of 3kHz. 



• 2020-03-27 :

   MASER drift rate adjustment: -24x10^-15, 56824->56800, 
  carried out 09:47 26/3/2020 EST. 



• 2020-03-24 :

   Cal waveform start/stop controls have been added to TOS and cal waveform generator (EVT IOC) restarted in a mode that requires them.  UWL verification tests should include a check that cal waveform is operating. 



• 2020-03-24 :

   Update various TOS package to the latest version. Still needs verification observation when the system is available. 



• 2020-03-15 :

   FROG: fixed a bug where selecting calibrator sources on the synoptic view cause the app to logout. 



• 2020-03-09 :

   Updated TOS templates to allow cal frequencies from 0.01 to 1kHz (was previously 0.01 to 100 Hz).  



• 2020-02-25 :

   Antenna pointing offsets: The pointing offsets OF2 (Ax) and OF3 (El) were found to be empty values in the InfluxDB and in the Medusa header files. This affects all historic UWL observations where a pointing offset was applied/requested via the DHAGU Az/El offset switch, and/or where offsets were queried in InfluxDB.    



• 2020-02-10 :

   MASER drift adjustment, -100x10^-15, 56944->56844, done at 12:03 summer time. 



• 2020-02-07 :

   MASER drift adjustment +200x10^-15, 56744->56944 @ 14:39 local summer time.
  A further adjustment will be required next Monday. 



• 2020-02-07 :

   MASER tuned during 3-day observatory shut-down.  Tuning commenced at 10:01 on Feb 4th 2020, and ceased at 10:55 on Feb 5, local (daylight savings) time.
  During this shutdown it was noted that the station clock (the computer pkclk00, not the MASER) was critically ill with its solid state disk drive 100% full and bursting with errors.  The clock was powered down between 13:59 and 15:34 (local, as above) on the 5th, to install two new SSD drives and re-build the system (thanks to Dan Craig).
  The MASER was then left to accumulate a new Tick-phase trend in anticipation of a drift adjustment. 



• 2020-02-05 :

   The following changes have been made to the pulsar fold-mode processing pipeline: the only data product is now the full band (no longer the 3 RF bands individually), and the custom zapping of the data product is only available for diagnostics and will no longer be archived. 



• 2020-01-08 :

   MASER frequency adjustment, -50x10^-15 (56794->56744), done at 0747 est 



• 2019-12-05 :

   Code added to desk program to guard against bad wind-stow-wait value received from servo (value outside valid range 0 to 600 seconds) 



• 2019-12-05 :

   SWEO room air-conditioner significantly upgraded to higher capacity Inverter system.  Job commenced 3/12/2019 and completed today. 



• 2019-12-04 :

   MoniCA - added alarm point for FC hoist powered-on (PKSSUP-5223); added a few antenna points:- RA_J2000, DEC_J2000 and ME_ERR that were present in desk stream but missing from points db. Re-started moniCA server on dalwhinnie ... all good. 



• 2019-11-26 :

   MASER freq adjustment, -10x10^-15 (56824->56814), done at 0724 EST. 



• 2019-11-20 :

   Medusa updated to support running 2 observing modes per sub-band at the same time (e.g. fold + search, search + continuum). Only possible via the manual web interface for now.  



• 2019-11-19 :

   MASER frequency adjustment, -20x10^-15 (56844->56824), carried out  at 08:23 local time 



• 2019-11-19 :

   * DHAGU URL updated to https://parkes-ops.atnf.csiro.au/DHAGU/
  * Parkes homepage link also updated.
  * DHAGU updated to v0.3 (see Help tab for updates.) 



• 2019-11-12 :

   TOS software update on _octomore_
  ```
  Setting up python-askap.epics (3.4.3) ...
  Setting up python-askap.opl.script (3.1.7) ...
  Setting up askap-css-opi-parkes (3.5.0) ...
  Setting up askap-css-config (3.1.0) ...
  Installing new version of config file /etc/askap/sites/parkes/css.ini ...
  Setting up libmariadbclient18:amd64 (10.1.41-0+deb9u1) ...
  Setting up askapioc-tools (3.3.0) ...
  Setting up libaskap-ioclog (3.0.0-2) ...
  Setting up python-askap.ds (3.0.3) ...
  Setting up python-askap.executive (3.0.4) ...
  Setting up libaskapioc (3.3.0) ...
  Setting up libaskap-epics-logging (3.0.0-2) ...
  Setting up parkes-ioc-uwbrx (3.0.1-4) ...
  Setting up parkes-ioc-pksmon (3.0.2-2) ...
  Setting up libcaclient (3.0.0-3) ...
  Setting up epics-ioc-metadata (3.1.0) ...
  ``` 



• 2019-11-12 :

   Updated the Parkes MoniCA server program (open-monica) to v1.2.4 



• 2019-11-08 :

   * Updated the Search template to only include nbit options of 1, 2, 4 and 8 (PUWL-216) 



• 2019-11-04 :

   * Updated TOS templates FOLD/SEARCH/CONTINUUM- if all sources in the Scheduling Block are not visible, it will now transition to ERRORED, not OBSERVED as before. 



• 2019-10-28 :

   FROG v3.8 update:
  * Updated Atmosphere websocket library from 2.5.2 to 2.5.4. 



• 2019-10-28 :

   PORTAL v2.3 update:
  * The booking calendar again prompts the users with a modal popup when booking.
  * Made bookings (i.e., person's name) more readable in crowded situations.
  * Updated Atmosphere websocket  library from 2.5.2 to 2.5.4.
  * Other minor cosmetics and bugs squashed. 



• 2019-10-25 :

   Medusa DSPSR updated to git hash 289dcdad, SPIP to e0d9103 



• 2019-10-22 :

   Parkes MASER adjusted -20 in 10^15 (56864->56844) 



• 2019-10-15 :

   ```PSRDADA_GIT_HASH        347c851
  PSRCHIVE_GIT_HASH       9cc6f3597
  DSPSR_GIT_HASH          fa0de11e
  SPIP_GIT_HASH           6cd55a8``` 



• 2019-10-15 :

   online NFS server upgraded from debian wheezy -> debian stretch 



• 2019-10-08 :

   PORTAL update v2.2 (to be released tomorrow):
  
  * Updated FullCalendar library from 3.4.0 to 4.3.2. All calendars completely overhauled as a result.
  * All calendars (Bookings, PTF and Admin) have the time zone configurable from the settings cog.
  * It is now possible to select a time zone, and book observing slots in that time zone and translate to any time zone (user request).
  * Fixed a bug where requests for unscheduled time did not show up for the user on the booking calendar- it is now possible to cancel requests.
  * Fixed a bug where the batch booking interface did not match date/times with the main calendar for the selected time zone.
  * Other cosmetics and bugs squashed. 



• 2019-10-08 :

   Parkes MASER drift rate tweaked today at 08:29 EST by -20x10^-15 



• 2019-10-08 :

   Updated pulsar search-mode combination script to define OBSFREQ key as type TDOUBLE, replacing type TFLOAT. Affects all search-mode observations after and including UTC 2019-10-01. 



• 2019-10-08 :

   Updated SDHDF definition v1.7 to reflect AJ's changes for Medusa header key 'CONTINUUM_OUTNPOL' to 'CONTINUUM_OUTSTOKES'  



• 2019-10-04 :

   DHAGU v0.2 update:
  Satellites now have a tooltip so that mouse hover reveals individual identifications instead of just constellation association.
  Added ATCA flux calibration sources for LS (>2Jy), CX (>4Jy) and K (>4Jy) to synoptic options. Mouse hover reveals name.
  Hopefully fixed an issue where submitting multiple sources from a Parameter Set does not result in SUBMITTED states (should be SCHEDULED).
  Added in all Parkes OPAL projects into project selection boxes, plus PUNDEF if there is no appropriate option to choose from.
  Addressed audio not playing on some browsers. 



• 2019-10-04 :

    PORTAL v2.1 update:
  Added an audio test button above chat.
  Addressed audio not playing on some browsers. 



• 2019-10-03 :

   FROG v3.7 update:
  
      Satellites now have a tooltip so that mouse hover reveals individual identifications instead of just constellation association.
      Added ATCA flux calibration sources for LS (>2Jy), CX (>4Jy) and K (>4Jy) to synoptic options. Mouse hover reveals name.
      Added an audio test button on the main page.
      Addressed audio not playing on some browsers. 



• 2019-10-02 :

   Added a monitor point to 'desk' ascii stream and moniCA server. Re-installed and restarted both. 



• 2019-10-01 :

   TOS Continuum/Search templates switched over from ```medusa.continuum.subband%d.output_npol = n
  ``` to ```medusa.continuum.subband%d.output_stokes = n``` where n = 1,2,3. All parameter sets need to be updated. Fold templates are not affected. 



• 2019-10-01 :

   Back on 2019-06-26, George changed the UWL's built-in flexible attenuators for Band 1 (the low band) from 18dB to 15dB. The new value has been used for all subsequent observations. 
  All attenuation changes are recorded in monica/grafana, but might not be obvious to users. It's also worth noting that, in the event of a power failure, the band 1 attenuator will reset to the default setting of 12dB. 



• 2019-09-17 :

   Applied minor Medusa software updates in preparation for Zoom mode. Mostly on high level control scripts with no impact on signal processing functions 



• 2019-09-17 :

   Some minor software updates installed to the following: joffrey, tommen, myrcella, tywin, dalmore, euryale, cersei & jaime 



• 2019-09-11 :

   FROG bugfix release v3.4: updated Atmosphere Framework websocket library (2.5.2 -> 2.5.4) and wind graph issues. 



• 2019-09-11 :

   increased the number of open file descriptors on the uwb account from 1024 to 65536. The previous limit was hit when performing a search mode observation which lasted longer than 4.5 hours 



• 2019-09-06 :

   Another pks MASER catch-up advisory:
  28/8/2019 08:36 (local), Frequency adjustment -30x10^-15 



• 2019-09-06 :

   HP_5 switched over to A_2 in the conversion system. 



• 2019-09-05 :

   DAT_FREQ column in Medusa search mode files should now be correctly ordered 



• 2019-09-05 :

   @dtm upgraded Medusa machines to Debian stretch (version 9.9), which resulted in major changes to the GNU compiler (4.8 -> 6.3), CUDA SDK (7.5 to 9.1) CUDA Driver (?? to 390.116)  and Mellanox OFED (?? to 4.2)  



• 2019-09-04 :

   Moved receiver monitor/control server 'pkicc00.main' to pkdesk ... renamed to 'pkmcServer00'. Changed monitor port number to 6244 to avoid clash. Updated hostname:port in pkmc gui and monica config file 'monitor-sources.txt'. Changed controller hostname from 'pkicc00' to 'pkdesk' in tcs receiver database 'rxdb.dat' and pushed new version to fcc directory on pkdesk. 



• 2019-08-28 :

   Minor change to the pkmc program that runs on pkicc01. Affects pkmc monitor streams from devices other than the receivers. See PKSSUP-5110 for more detail. 



• 2019-08-28 :

   Added Cal Phase monitor/control to UWB observer GUI 



• 2019-08-23 :

   Updated Medusa to read the Cal Phase from TOS and apply in the Calibration data 



• 2019-08-22 :

   Missed notifications:
  Frequency changes to the Parkes MASER:
  19/7/2019, -20 in 10^15
  6/8/2019, -20 in 10^15 



• 2019-08-22 :

   Cal phase (default 0 degrees as per MM discussion) now in templates Standard/Continuum/Fold/Search and also now in dhagu :) 



• 2019-08-22 :

   Cal phase (default 0 degrees as per MM discussion) now in templates Standard/Continuum/Fold/Search and also now in dhagu :) 



• 2019-08-22 :

   Updated TOS to the latest release of packages, cal phase control now available pending template updates 



• 2019-08-12 :

   Installed a round of updates today, most notably on euryale + medusa servers (inc reboots) 



• 2019-07-29 :

   Adjusted MASER frequency by -20x10^-15 



• 2019-07-27 :

   UPDATE: Medusa search-mode combination script updated to handle missing sub-bands and fix for incorrect OBSBW values. Affects observations from and including UTC 2019-07-26-17:07:09 



• 2019-07-20 :

   yes, saw it and switched it back to the proper value, haven't had a chance to diagnose the root cause though 



• 2019-07-19 :

   I just manually switched the DUTC value of pkclk00 from 0 to 37 (which is the value that was set on pkclk01) in response to an email from monica. Not sure why it was at zero. 



• 2019-07-18 :

   10dB pads fitted to all 6 UWB outputs to the PCS, translator cables 36, 37, 38, 39, 40, 41.   This is to address Alex Dunning's theory that reflections from the PCS are causing ripple on the signal to the other full signal path. 



• 2019-07-16 :

   TOS templates and UI updated to implement Tsys calibration spectra range 0.001953125 - 1 MHz (not tested though.) 



• 2019-07-15 :

   Medusa UWL preprocessor modified to support production of calibration spectra with higher than 0.125 MHz resolution. Resolution down to 1.953125 kHz tested 



• 2019-07-12 :

   UWL backend was reconfigured/synced following tests at 11/7/19 6:10pm and 12/7/19 4:12pm (AEST). System timing may have changed (as we have little data about stability so far). No impact to legacy users using other backends with the UWL receiver (i.e. VLBI) 



• 2019-07-11 :

   MASER drift rate tweak by -20x10^-15, done at 0927 local.  For future reference, these frequency adjustments don't cause a step change, but rather they produce a change in the rate of drift of the frequency standard.  A gross frequency change will cause a sharp kink in the drift rate, so I'm currently trying more frequent small "tweaks" to keep the changes subtle. 



• 2019-07-09 :

   Please change my update on 2019-07-01 to say:
  - UWL ADC software updated to correct unstable AB phase difference (polarisation alignment) which changed on each power cycle / reconfigure (worst case was +- 63ns, now <1ns)
  - UWL DSP firmware updated to correct absolute delay to be closer to expected value (1005ns). Change from previous setup may be up to 1us. Not yet finalised.
  
  the other change about ADC means can be deleted from this entry because it hasn't been applied yet. 



• 2019-06-27 Prior to today, for MEDUSA, the CAL FFT lengths (for each output channel) were set to 4 x nchan - so for 128 channels/subband, the output channel FFT length was 512. It has been changed to 32 x nchan to address this sort of leakage issue. This means that for 128 channels/sub-band the output channel FFT length is 4096.

• 2019-06-26 Changed UWL Band 1 attenuators to 15 dB.

• 2019-06-26 Changed UWL Band 1 attenuators (both pols) to 12 dB. Also increased length of Medusa FFT sub-band channelizer for fold-mode cal observations from 4xNchan to 32xNchan points, where Nchan is the number of channels/sub-band, to reduce FFT sidebands on strong signals.

• 2019-06-13 Logging cadence has been increased in the InfluxDB points db. Changes will take effect after the next monica restart which will be during maintenance time this Thursday, 13th June.

• 2019-06-06 Medusa resynced

• 2019-06-06 Medusa updated "Rather than discarding the final sub integration in Continuum mode, Medusa will now write out the partial sub-integration in the final file."

• 2019-06-03 re-synched the digitisers/FPGAs as noted that the pols A and B in subbands 21 and 22 (counting from 0) had got out by 16us.

• 2019-05-15 Restart after ~10 days of maintenance - MB receiver installed alongside UWL. Medusa changes: merged changes reflected in meerkat backend development - search-mode - fixes for sub-band time stamp offsets in coherent de-dispersion mode (STT_OFFS are different per sub-band before changes, the same per sub-band after changes) - Medusa now deals correctly with NFCOEF and NTCOEF in obscat.db file (previously this file was ignored) Euryale changes: - fold-mode - implemented file size threshold mode for long observations (combining subints up to a maximum file size ~4GB) - naming scheme for default mode set to 'uwl_YYMMDD_HHMMSS_bN.{r,c}f' - naming scheme for file size threshold mode set to 'uwl_YYMMDD_HHMMSS_bN_N.{r,c}f' - search-mode - naming scheme for default mode set to 'uwl_YYMMDD_HHMMSS.sf' - naming scheme for file size threshold mode set to 'uwl_YYMMDD_HHMMSS_N.sf' - sl-mode - still v1.5

• 2019-05-09 Implemented SDHDF v1.5, incorporating fix cal-on/cal-off dataset ordering and fix missing dimension scales.

• 2019-04-05 Implemented SDHDF v1.4, incorporating fix to correctly order integrations in time and separate datasets for cal-on and cal-off.

• 2019-04-05 Implemented SDHDF v1.3.1, to temporarily fix time integration ordering for P977 data

• 2019-03-31 POL B missing from the UWL-Mid band. P977 observations from UTC_START=2019-03-30-17:33:51 onwards are affected till Sunday midday; FIX: restart FPGA data stream

• 2019-03-30 some of the data in some of the sub-bands (mid and low) identifed as being in the wrong time order.

• 2019-03-29 Baseband tests - DFB4 fold, Medusa baseband for SB5, SB7, SB20, Medusa fold on all other SBs

• 2019-03-28 Over the last couple of days it has come to light that the cal signal for spectral line observations made using the TOS system has not been properly synchronised with the medusa data. Recent testing has shown there was significant merging between the on and off labelled spectra (for example for some recent data the on-off cal spectra were about a factor of 10 smaller than they should have been). Over the course of the last couple of days the issue has been investigated and fixed, so all data from now (28/03) on should be fine. For data taken prior to today with a cal controlled by TOS, the absolute flux scaling will be in error, although we have yet to fully quantify the factor (potentially of the order of a few % TBD). We would also be concerned with the calibration of polarisation data taken prior to today"

• 2019-03-22 AJ implemented 4-pol calibration for Medusa spectral line data - tested in P737 session

• 2019-03-21 Discrepancies with the coherent dedispersion mode data discovered "The short story is that when digifits produces a coherently dedispersed PSRFITS search mode file, it discards a number of samples at the very start of the observation since they become corrupted by the convolution process that removes the intra-channel delays. The number of discarded samples is recorded in the meta-data as a shift in the start time of the observation (STT_OFFS). The number of time samples removed is both DM and frequency dependent. What this means is that for observations with a particular DM, the 26 sub-bands produced by Medusa will all have slightly different start times, that should be larger for lower sub-bands. Where this becomes problematic is when the search mode files are appended on Euryale, since I suspect that the start time (STT_OFFS) of the first search mode file is made common to all the sub-bands within the combined file. This will result in a small temporal shift for each sub-band. If each search mode sub-band file is processed independently (e.g. folding a pulsar with DSPSR), then it honours the start times everything lines up, however it we process the whole file, then these offsets become apparent."

• 2019-03-12 Restart after 8 days of maintenance - MB receiver installed alongside UWL

• 2019-02-20 9am AEDT - band 1 (low band) attenuation changed from 18 to 12dB

• 2019-02-11 Medusa updates, fixed negative bandwidth output

• 2019-01-16 DSPSR patch checked in to resolve start-time ambiguities in DSPSR usage

• 2018-12-17 ~23:45 changes to Medusa chbw from -ve to +ve ; Medusa sub-band data freq, now range from low->high, Source J1644-4559 test

• 2018-12-07 Added metadata changes to config file instead of inline

• 2018-12-05 Fix OFFS_SUB table for PRESTO issues in pfitsUtil_searchmode_combineTime.c (GH)

• 2018-11-17 Second run of search-mode pipeline (freq ranges from low->high in combined sf products)

• 2018-11-16 Changes made to primary header parameters: FD_HAND (empty to -1), FD_SANG (empty to 0); BE_DCC (0 to 1), NRCVR (1 to 2); SUBINT:NSUBOFFS updated to increment NSUBOFFS for observations split in time;

• 2018-11-12 First run of search-mode pipeline

• 2018-11-01 Fold-mode: updated zap list to 3.0

• 2018-10-30 Before this date: Band1 frequency range low->high; Band2 and band3 freq range high->low

• 2018-10-30 After this date: All bands high->low, BE_DCC = 1

• 2018-08-15 Fold-mode: updated zap list to 2.0 - CFITSIO updated to v3410; LD_LIBRARY_PATH updated; updates to pfitsUtil_copyToNew_v2.c (GH); all fold mode data from 14-05-2018 reprocessed;

• 2018-07-11 Fold-mode: zap list 1.0 - band b2 failed due to CFITSIO errors

• 2017-08-31 The version of PSRCAT software updated on joffrey (used by TCS and the DFBs)

• 2017-07-25 The version of PSRCAT software updated on CASPSR

• 2017-01-17 CASPSR was found to not have had the leap second added at the start of the year. Data from 1st Jan until this date will therefore be affected.

• 2016-12-05 Multibeam returns: Initial observations have MB signal through port 4 of the switch matrix versus port 3. Timing properties not predicted to be different.

• 2016-09-13 Maser failed. Check electronic log for details of corrupted observations. A new maser was installed on 16/09 2016.

• 2016-08-16 CASPSR moved to complete backend movement to accommodate SETI equipment. Move took place on 16-17. Pre and post move delay measurements taken on the 15th and 19th (see email from Jimi Green 05 2010 2016).

• 2016-07-09 Translator problems resulted in session with focus disabled during 10/50 observations. Receiver position optimized for 45 degrees elevation.

• 2016-07-06 DFB4 moved to DFB3 rack. Delay changes expected. Bandpass observed to be different before and after move. Delay measurements include observation of J0437-4715 at 10cm with DFB4 and CASPSR (unchanged) before and after, as well as modulated cal scheme (multiple DFB4 configurations, pre-and-post; last configuration before/after pdfb4_2048_1024_1024) and by recording baseband data with CASPSR with the DFB4 polA and polB placed into CASPSR polA.

• 2015-07-20 RMS updated DR1 processing to zap 2670-2690 MHz apparent 4G RFI.

• 2015-02-28 Alectown Optus 4G station powered -- causing serious problems to 50 cm observations.

• 2015-04-13 WvS updated 3100+732.cmd to better configure the attenuators for the power levels required by CASPSR

• 2015-02-10 RMS observed in experimental 50cm modes with variable centre frequency (that may not agree with headers in fits files). Data should be considered unreliable

• 2015-01-16 Updated the tempo2 files (EOPC, pks2gps and gps2utc) at Parkes. They were many years out of date.

• 2014-10-25 "current" at commencement of 20cm session on 25 Oct spectral kurtosis turned off on CASPSR.

• 2014-08-25 "current" versions of pdfb3/4 correlator programmes updated to have CTR_FRQ definition in history table conform to psrfits definition

• 05 2014 to present Major problems with pdfb3. pdfb3 could not sync both FPGA boards making pulsar timing modes inoperable. Search mode (and spectral-line) modes continue to function. PPTA observations continue with CASPSR and PDFB4 only.

• 2014-03-17 "current" versions of pdfb3/4 replaced to correct spurious stt_date and stt_time entries at end of main header. Old "current" versions changed to "previous"

• 2014-02-27 Temporary replacement supply. Clock reset about 3 us off, but still drifting slowly.

• 2014-02-26 During the early hours of the morning, the internal power supply for the maser failed. Brett has fixed it, but there will be a step-change which will be measured

• 2014-01-21 New versions of pdfb3 and pdfb4 installed. New version is labelled "current". Previous version "latest" now labelled "previous".

• 2014-01-14 Updated psrcat so that the ephemeris does not contain "tabs" (only spaces)

• 2014-01-14 Recognised that P456_50cm_DFB3.sch had LEVCAL mode for post-obs cals for J0437, J1744 and J1909. Should be CAL mode so atten not reset. Schedule fixed.

• 2013-09-01 A few days up until this date we were updating the observing software to remove dependency on /psr1. This required updates to psrcat, tempo2 and T2psrcoeff. Various issues were found including 1) pulsars with P0 and P1 instead of F0 and F1 were incorrectly converted for the ephemeris and 2) a bug was introduced into T2psrcoeff so that it did not include the PREDLEN parameter (fixed on the evening of 1/9 2013). The various catalogue files were all moved to a new location and the most up to date versions of tempo2 etc. installed.

• 2013-08-14 Updated pks2gps.clk at Epping (this was very out of date - last update in Oct 2012)

• 2013-05-14 New version of pdfb3 and pdfb4 installed (cccversion latest). Fixes problems with search mode: spurious blocks at EoF for 1-bit mode, first sample zero, channel offset for first block. Now uses PSRFITS V5.1 with nbit=1,2,4,8 all written in B mode.

• 2013-05-09 Maser glitch ~6us at UT 2240 (MJD 56421.93). Reset by Brett P. at approx UT 0245 10/5 2013 (MJD 56422.12)

• 2013-04-12 Mike Keith and George Hobbs updated the catalogue for APSR and CASPSR (noting that J0437-4715 incorrectly had an old A1DOT in the previous ephemeris)

• 2013-03-25 Power failure in tower etc UT 15:45 (MJD 56376). Among other things, clock maser failed. Systems restored by John S and Brett P ~18:00UT. ~600ns offset in Pks clock. Reset to ~20ns on 26/3 2013 at ~22h UT.

• 2012-09-12 CommonView GPS restored to dual-frequency system.

• 2012-08-17 CommonView GPS transferred to old single-frequency GPS.

• 2012-08-14 Maser relocated. GPS system OK, but CommonView failed. Also changes to BAT wiring after clock; effect should be 20ns or less.

• 2012-06-14 PDFB4 removed and sent to Epping to fix connector problem caused by BAT line changes. Reinstalled 3 July.

• 2012-05-30 pdfb3 and pdfb4 cccversion "latest" updated to fix search-mode 8-bit 4-pol offset.

• 2012-05-23 Recurrence of loss of 5 MHz phase lock on PDFB4. Cable length adjusted by John Sarkissian and George Hobbs (5m removed and 20m added) at ~0400 UT.

• 2012-05-16 Brett P. removed 20m of extra cable on PDFB4 5 MHz line (5m left in). Fibre media adaptors in BAT line to PDFB4 removed and replaced by direct copper connection.

• 2012-03-05 Recurrence of loss of 5 MHz phase lock on PDFB4. Cable length adjusted by Tim Ruckley and John Sarkissian on 3/5 2011 (20 metres of cable added)

• 2012-04-06 5 MHz phase lock problems on PDFB4.

• 2011-12-16 New versions of pdfb3 and pdfb4 (latest) with V5.0 of PSRFITS installed ~UT 09:30.

• 2011-12-01 Occasional loss of 5 MHz phase lock on PDFB4. Cable length adjusted by BP on 20 2012 2011.

• 2011-11-18 10cm LO error from UT 06:00 - 17:05 (MJD 55883.25 - 55883.712). Effective RF 3094 MHz.

• 2011-11-18 New par file for J1017-7156 from MJKeith

• 2011-03-28 The A and B cables for the MB receiver were switched in the focus cabin during the shutdown. This was detected and repaired on 28/3 2011. Data from MJD 55644.0 to 55648.2 are affected.

• 2011-03-01 to -24 Major shutdown. Snake cables on MB receiver replaced. Problems at cable-connector interface, causing intermittent bad connections.

• 2011-03-25 10cm data bad from UT0930 to UT1730 due to switch matrix problems related to CASPSR.

• 2011-03-24 (MJD 55644) New ephemerides in ppta.db for all 22 pulsars.

• 2011-03-20 PDFB3 updated control software reinstalled. TCS updated to include CASPSR.

• 2011-02-27 Remaining files affected by 3094 MHz problem fixed on Epping disk

• 2011-02-26 ~06UT PDFB3 reverted to previous version since problems with APSR level control.

• 2011-02-25 0440UT (MJD 55617.2) PDFB3,4 updated control software in pkccc3,4 (WEW). PSRFITS V4.9

• 2011-02-08 Tempo2 updated at Parkes after a bug was found in the use of predictors (the site velocity was sometimes set to infinity when using a tempo2 ephemeris)

• 2011-01-01 J1732-5049 dropped from regular PPTA observing

• 2010-10-13 Switch matrix for IF control installed. CASPSR installed.

• 2010-09-18 10cm files with RF 3094 MHz and 20cm files with wrong sign for bandwidth identified by JER (JER and RNM emails of 2/8 2010). These files fixed on Epping disks using: psredit -m -c ext:obsfreq=3094.0; pam -m -o 3094.5; psredit -m -c bw=-256; pam -m -o 1368.875; (See RNM email of 14/9 2010 for details);

• 2010-09-07 J1017-7156 added to ppta.db and schedule files.

• 2010-08-17 PDFB3,4 firmware restored to V6,V3 at 06UT, 54756.25. J2241-5236 added to ppta.db and schedule files.

• 2010-08-16 New firmware for PDFB3,4 (V7,V4) at 11UT, 54755.46

• 2010-08-13-15 PDFB4 moved to new rack

• 2010-06-30 CPSR2 and PDFB2 decommissioned.

• 2010-05-03 PDFB3 and PDFB4 firmware updated at UT0415 (MJD 55319.18) to remove delays, i.e. to make time-stamp refer to time of signal arrival at digitiser input. In case of PDFB4, the PTU was also changed to remove the two-bin delay. (V6,V3)

• 2010-02-09 J2241-5236 added to PPTA schedule

• 2010-01-07 DM=1 problem which has existed for PDFB3,4 data since 21/09/2009 fixed ~UT0200. Only pulsars with DM1 entries in par file were affected. pkccc3 and pkccc4 updated (~UT1000) to take DM1 into account.

• 2009-12-14 to -24 Pks clock problems. Clock lost lock at MJD 55179.50 (Dec 14), lock restored at MJD 55181.206 (Dec 16) at +46.6 us. Reset to -0.97 us at MJD 55189.173 (Dec 24).

• 2009-11-18 Changed default predictor to tempo2 for psrchive at Epping

• 2009-10-28 Eng Rack distribution amplifier replaced with three-way splitter. 6db pad on inputs maintains levels close to previous values.

• 2009-09-02 APSR/PDFB3 frequency leakage problem fixed by Andrew Brown. Start of good APSR data.

• 2009-09-02 New version of PDFB3 control software installed. Uses Ver 4.0 of PSRFITS and other changes including to level control system. PSREPHEM table replaced by PSRPARAM (block text) table. (V5)

• 2009-08-31 From this date, occasional incorrect DM (= 1) in PDFB3 & PDFB4 file headers. Data are OK.

• 2009-07-29 Entry for J0437-4715 in ppta.db updated to correct drift in PDFB pulse phase during obs

• 2009-07-27 10/50cm receiver reinstalled with 50cm band at 700-764 MHz, reconditioned 10cm preamps.

• 2009-04-26 UT2300 PDFB3 delay compensation activated. (V4) MJD 54947.96 [MTK: NB this means correction for two-bin delay; configuration dependent delays still present.]

• 2009-02-11 UT0600: PDFB4 control software upgraded to use FITS def V3.9.

• 2009-01-01 Following the leap second at 0000UT on 01/01/09 and until 2350UT on 03/01/09 for PDFB3 and until 0000UT on 05/01/09 for PDFB4, recorded start times are late by 1 second, i.e. 1 second should be subtracted from stt_smjd for these files.

• 2008-10-16 PDFB3 software upgraded to be same as PDFB4 (WEW) (V2)

• 2008-10-12 PDFB4 reset bug fixed ~0500UT 54751.30 (AB) (V2, First reliable PDFB4 data)

• 2008-09-29 PDFB4 hardware and firmware changed to fix PTU sync problem.

• 2008-09-22 PDFB4 software and firmware updated (AB). Reset bug introduced so that folding period not updated leading to phase drift during observation. Also intermittent phase jumps introduced due to sync problem in PTU firmware. (V1)

• 2008-09-10 5 MHz phase-lock problem still intermittently present. Another 8m cable added by JR.

• 2008-09-08 Original PDFB3 second board restored since spare unit (installed 22/08/2008) had digitiser problem.

• 2008-09-03 B-channel gain stability problem shown to be in Eng Rack splitter amplifier (Mal Smith). Amplifier bypassed since approx unity gain overall.

• 2008-08-31 PDFB1 decommissioned

• 2008-08-27 Gain stability problem with B channel of PDFB3 and PDFB4 noticed (JR). Gain variations of ~1db on ~minute timescales.

• 2008-08-22 Second PDFB3 board (A2/B2) replaced with spare unit.

• 2008-08-12 At 0545UT loss of 5 MHz phase lock on PDFB3, leading to random changes in pulse phase. Largely fixed by JR inserting 8m cable in 5 MHz line.

• 2008-07-29 New single-board system, PDFB4, commissioned.(V0)

• 2008-07-29 New PDFB3 boards installed by Andrew Brown (MJD 54676). Most recent firmware with PTU on-board. (V1)

• 2008-05-01 PDFB3 schedule files in /psr1/tcs/sched.

• 2008-02-20 First light PDFB3 (V0)

• 2008-01-16 New PSRFITS definition including BE_DELAY

• 2007-12-29 Redefinition of STT_MJD by psrchive removed. Previously middle of 1st subint, now original value.

• 2007-12-17 Pulse phase instabilities in PDFB1 data. PDFB2 appears OK. Possibly fixed by power cycle of whole PDFB1 system 20 2012/07, 03UT.

• 2007-12-02 WEW updates PDFB2 FITS version to 2.17. This includes BE_DELAY - the backend propagation delay from the digitiser input. STT_OFFS unchanged, i.e., still sum of delay to pulse phase zero and BE delay.

• 2007-12-02 John S. to use '-L' option to bin-zap automatically when pre-processing

• 2007-11-19 PDFB2 configurations changed from gain 7 to gain 6; has effect of reducing uncalibrated amplitudes by a factor of 16.

• 2007-10-23 Discovered reason for sudden change in DFB2 digitiser levels on 01/07/2007. RMS levels multiplied by sqrt(8) to restore to correct value. LEVCAL now setting input levels correctly.

• 2007-09 Defaults for TZMJD etc. for T2 predictors

• 2007-08-02 Par files in ppta.db updated (V1.3)

• 2007-07-01 Reported rms levels for PDFB2 inadvertently reduced by sqrt(8). Input power increased by about 10db to compensate, bringing rms levels back to 10.

• 2007-06-21 Applied PDFB2 delays changed to the values appropriate for the 512 MHz system (see 5 June above) - i.e., twice the actual value!

• 2007-06 Implementation of pazi for pre-processing

• 2007-06-05 Commissioning of 1GHz PDFB2 system. The actual d2 delays (see 16 March above) are halved. For 1024 MHz, 1024 chan the actual delay is 4.59 +/- 0.1 us.

• 2007-05-19 Start running PDFB1 in alt mode. Occasional instances of alt-mode TCS giving incorrect sign of bw (e.g., +256 for MULTI). Spectrum remains inverted despite bw sign.

• 2007-05-17 Changed from H-OH to Multibeam - central freq. from 1433 to 1369 MHz

• 2007-05-06 Linearity tests for PDFB2 by RNM and JV. LEVCAL setting of rms to 10.0

• 2007-05 Use of Tempo2 predictor files with PDFB2

• 2007-04-17 - 2007-05-01 Changed FITS HDRVER from 2.5D to 2.10 for VO compatibility. Parallel operation of PDFB1 and PDFB2.

• 2007-03-16 PDFB2 BE delays set to (incorrect) PDFB1 values, i.e., 32 chan: 1.2969 us; 128 chan: 3.5547 us; 512 chan: 12.2813 us; 2048 chan: 46.8829 us. For 256 and 1024 chan, no correction was applied. Actual delays are d1 + d2, where d1 = 14.00 us with an unpredictable fluctuation of up to 60 ns, and d2 is bandwidth and channel dependent as follows: 256 MHz, 512 chan: 13.91 us; 256 MHz, 1024 chan: 17.20 us; 256 MHz, 2048 chan: 43.15 us. The actual delays fluctuate around these values by 100 ns or so. Delays for other bw/chan combinations are currently not determined.

• 2007-03-16 First light for PDFB2, max bandwidth = 512 MHz.

• 2006-12-16 Changed from MULTI_1 to HOH - central frequency from 1369 MHz to 1433 MHz

• 2006-10-03 Moved DIG_STAT table to end of FITS file to avoid CORDAT not keeping up problem.

• 2006-08-07 More linearity tests. LEVCAL operation implemented to reduce SPD levels to 5.0. STT_OFFS modified to allow for nchan-dependent propagation delays through digitiser (indicated by PROC_CMD in the HISTORY table being set to PSRDFB:2006-08-07_14:55). Attenuator settings moved to digstats table. Gal l,b, PA, FA, parang, tel coords added to subint table.

• 2006-06-21 Amplifiers preceding digitiser inputs removed.

• 2006-05-31 PDFB1 linearity tests by RNM, JR, AB.

• 2006-01-26 First light Methanol multibeam receiver

• 2006-01-01 Implementation by JR of fix to PDFB1 configs to correct PTU problem diagnosed by RTE. Effect was sawtooth drift pattern of ~200ns in 5 min of TOAs.

• 2005-11 Input levels to PDFB1 reduced to avoid saturation problems.

• 2005-06-20 Commissioning of PDFB1 (Prototype DFB, 256 MHz bw)

• 2004-11-12 New feed polarisation parameters introduced, FITS Version 1.24.

• 2004-09 Wide Band Correlator first bin - last bin problem fixed

• 2004-09-07 MB Receiver reinstalled

• 2004-05-21 Discovery of configuration dependent time delays in the WBC (RNM email 19/06/2004)

• 2003-10 to 2003-12 More WBC commissioning tests. Most problems solved

• 2003-09-11 Removal of MB receiver, installation and commissioning of 1050cm receiver, installation of H-OH receiver with upgraded preamps.

• 2003-02 Commissioning of Wide Band Correlator system. Limited configs and still instrumental problems.