Date: Sun, 10 Feb 2002 16:18:18 -0700
From: "s. a. snedden"
Subject: Re: 2.5m obslog 20020209 (MJD 52315)
Thanks, Jon for checking things out on a Sunday.
We did try more than 1 node, but we didn't know about the "special
address". All the nodes returned the same values without it.
Steph
Jon Brinkmann wrote:
>
> If the frequency of failures is increasing, perhaps we need to consider
> swapping the sdsshost ptvme card now instead of waiting for bright time.  I
> hate to do it, considering the risks involved.  The problem is that we have to
> power down sdsshost and the DA completely to do the swap.  Computers that never
> get powered down have a tendency to have hardware failures when they are
> powered down.
>
> I checked all the nodes this morning.  They're working fine.  I suspect that
> "ptvmeReset" did clear the problem as usual, but "ptvmeTest" was not testing
> what you thought.
>
> > ptvmeTest Results for node 0:
> >
> > im> ptvmeTest
> > Enter node ID of remote node (0-15): 0
>
> According to Ron R, You cannot test node 0, since it is the node in sdsshost.
> I think that's because there are no "loopback" connections on the ptvme bus,
> i.e., you must test from node 0 in sdsshost to another node.  Also, you need to
> enter the address.  Use:
>
>         "ptvmeTest -n 1 -a 0x18000000" to test the ptvme node in photo 1
>         "ptvmeTest -n 2 -a 0x18000000" to test the ptvme node in photo 2
>         "ptvmeTest -n 3 -a 0x18000000" to test the ptvme node in astro
>         "ptvmeTest -n 4 -a 0x18000000" to test the ptvme node in spectro
>
> I've appended an old E-mail on the subject.  Unfortunately, it's not in
> folklore, since it predates folklore.  Would one of the observers on tonight
> put it in folklore?
>
> Jon
>
> > To: 25m-observers@apo.nmsu.edu, cloomis@apo.nmsu.edu
> > Subject: DA PT/VME link diagnostic commands
> > Cc: au@pha.jhu.edu, jb@apo.nmsu.edu
> > Date: Mon,  3 Jan 2000 14:32:06 -0700 (MST)
> > From: jb@apo.nmsu.edu (Jon Brinkmann)
> >
> > While troubleshooting the imager and spectrograph DA systems this morning
> > with Ron Rechenmacker, I learned to use the diagnostic package "ptvme".
> > Here are my notes on it's use, which I will print out and put in the
> > "Observer's Bible" in the 2.5 meter control room:
> >
> > sdsshost % setup ptvme
> >
> > sdsshost % ptvmeReset                         (warm boot of the host PT/VME)
> >
> > sdsshost % ptvmeTest -?                                               (shows options)
> > sdsshost % ptvmeTest -n "address" -a "remote address"         (usual usage)
> >
> > ***   where the options for the flags are:
> >
> >       addresses on sdsshost:
> >                       0 = host (can't test)
> >                       1 = photo 1
> >                       2 = photo 2
> >                       3 = astro
> >                       4 = spectro
> >
> >       addresses on sdssmth:
> >                       0 = host (can't test)
> >                       1 = mt
> >
> >       remote address = 0x18000000             (Ron says this is "special")
> >
> > Here is typical command line and the data returned with no errors.
> > FYI:  error = 0 is the usual UNIX value returned for successful completion.
> >
> > sdsshost % ptvmeTest -n 1 -a 0x18000000
> > errno =  0: ioctl (fd=3, PT_IOCTL_GET_PRIORITY, &ptPriority) = 0
> >
> > errno =  0: ioctl (fd=3, PT_IOCTL_GET_VECTOR, &ptVector) = 0
> >
> > errno =  0: ioctl (fd=3, PT_IOCTL_GET_TIMEOUT, &ptTimeoutSave) = 0
> >
> > errno =  0: ioctl (fd=3, PT_IOCTL_GET_OP_MODE, &ptOpMode) = 0
> >
> > errno =  0: ioctl (fd=3, PT_IOCTL_GET_DEBUG_MODE, &ptDebugMode) = 0
> >
> > errno =  0: ioctl (fd=3, PT_IOCTL_GET_DEBUG, &ptDebugSave) = 0
> >
> > errno =  0: ioctl (fd=3, PT_IOCTL_GET_NODE_ID, 0) = 0x00
> > PID:               1716
> > IOPB size:           64 bytes
> > ping:                no
> > operation:         read
> > loopCnt:              1 (defaulted) 100 loops per dot
> > LWordCnt:    0x00000001 (defaulted)
> > controller:           0 (defaulted)
> > remNode:              1             --> /dev/ptvme0n1
> > remAddr:     0x18000000 (defaulted)
> > IxferMode:         0x00 (defaulted)
> > TxferMode:         0x00 (defaulted)
> >
> > errno =  0: open ("/dev/ptvme0n1", O_RDWR) = 3
> >             ptPriority  =       0x03
> >             ptVector    =       0x01
> >             ptTimeout   =        900 ticks
> >             ptOpMode    =          1 SINGLE_IO
> >             ptDebugMode =          0 NORMAL
> >             ptDebug     = 0x00000000
> > .                                                          1          3 ms 13:46
> >
> > errno =  0: lseek (fd=3, 0x00000000, SEEK_CUR) = 0x00000000
> >
> > 00000000: 01 95 A0 8C 04 05 06 07 08 09 0A 0B 0C 0D 0E 0F
> > 00000010: 10 11 12 13 14 15 16 17 18 19 1A 1B 1C 1D 1E 1F
> > 00000020: 20 21 22 23 24 25 26 27 28 29 2A 2B 2C 2D 2E 2F
> > 00000030: 30 31 32 33 34 35 36 37 38 39 3A 3B 3C 3D 3E 3F  ?>
> >
> > Jon
> > ---
> > Dr. Jon Brinkmann                             Apache Point Observatory
> > SDSS Instrument Scientist                     2001 Apache Point Road
> > Internet:     brinkmann@nmsu.edu              P.O. Box 59
> > URL:  http://www.apo.nmsu.edu/brinkmann/      Sunspot, NM 88349-0059
>