Date: Fri, 11 Dec 2009 12:22:48 -0500
Subject: Re: 2009/12/10 - 2.5m obslog 20091206 (MJD 55176)
From: Craig Loomis
On Fri, Dec 11, 2009 at 8:03 AM,   wrote:
> APACHE POINT OBSERVATORY SDSS 2.5M OBSERVING LOG
>
>                 December  10, 2009
>
>                       (MJD 55176)
>
> Our network is very slow. Shsshost2 replies with large delays.
  I believe this is due to one of the DNS servers being down
(192.41.211.10). Some hosts are configure to round-robin quickly, some
not. If you touch one of the slower ones, it takes a while to start
any TCP connection.
> 03:55Z   Something strange to note.  In ./list, the cart and plate IDs
> were 0,0 for all exposures until 104619 (Cart 13, second exposure).  After
> that, all plate and cart IDs were -1,-1.
  The platedb was down at the start of the observations due to a
problem on the sdss-db machine; I started platedb and the guider a bit
later. The _guider_ is assumed to be the definitive source of the
cartridge/plate info, and without a 'guider loadCartridge' the plate
and cartridge ids will not be recorded in the standard header cards.
That said, there are SPnCARTID cards which record what the specmechs
see, and we could also directly record what the mcp sees.
> cart 16 plate 3663 is mounted.
>
> 08:52Z We try to run hartmann, but STUI thinks that petals are closed,
> once MCP shows they are opened. After "mcp info" the problem disappeared.
>
> Piston "c" at sp2 got stuck and did not move after 3000 steps movement was
> requested.
> Hartmanns (104636, 104638), then calibrations (104640,41)
>
> GSOGTF
>
> We got stuck at this place because the recommended scale change was 0.014%
> but "sop shaleChange delta=0.014 spew a python code error:
> sdss3@hub25m[1000]$ cat ttt
> 09:38:13 APO.dmbiz 9 sop f text="command failed: TypeError(\"unsupported
> operand type(s) for *: 'float' and 'NoneType'\",) at
  Interesting. The tcc had been restarted but never queried for the
scaleFac. I need to add 'show scale' to the connection commands.
  There is no significant problem with sop, though it should fail a
bit more informatively.
> 11:10Z 104645 does not want to be read. It is repeating to report
> "PREREADING".
> We did boss exposure abort, boss exposure readout.
> After this 104645 was successfully read-out (spectra look ok).
>
 This came after a problem communication with one of the DAQs. Ticket
#475 filed.
 - craig