New LBC CMUs Configuration and Testing

  • the data probe does not respond through LBC software for housekeeping unless you are root; during shutdown when we finally powered up the electronics boxes again, I saw the same problem for testfilters
    power.py can not be launched using sudo , it complains about X
  • complaints from the ob.php and mainob.html in http log, but many, many others are gone now
  • I couldn't turn on telescope after importing an OB with telescope off - it didn't change it when I clicked and played - is this normal?
  • there are some warning messages that should be errors - do they just automatically lead to errors or not? for instance: system has been deactivated from filters and bad communication with serial device
    there's no way to recover from these except killing the software
  • do we need to restart winbind daily? (see IT 3292)
  • do we need the chmodHK cron job?
  • convert program needs LD_LIBRARY_PATH set - should we recompile ImageMagik so that it doesn't need that?
  • old SAMBA config used LBCNET as workgroup, new config is cmu.lbc.lbto.org - does it matter? it used to be guest only but now it's public ?
  • the windows software cannot open the remote log sometimes - something must be hung on the lbc control side, it needs lbckill to get out; can we make it a bigger problem on the Windows side? can we restart the RPC server on the CMU side?
    See Oct-2014 note in IT 2397

    This is a problem with running the test programs when the LBC software is up. I have modified the test programs to tell the user to restart LBC software afterward if it wasn't taken down.
  • get faulhaber driver from current CMU in Windows mode?
  • still core dumps sometimes (all of these were in the old system too)
    • in execute from PowerSwitch on startup
    • preset retry, with client handle NULL (camera.c:1126)
    • race condition between TCS uninit/init and reading DD entries - saw crash at tcs.c:389 when re-initializing one side of TCS
  • email SkyTech after summer vacation: details on BBB configuration, etc (draft started)
  • lbc in startup - after dgrp and applicom
  • clear the HK graphs - print empty graph not an old one - empty the files and re-run gnuplot
  • before on-sky tests, make sure the config files are the same between old/new CMU done
  • new lbc.zip for the windows PCs with changes to not MT the save in CCD controller and not launch the IA program just a change to the channel config file
  • check how DOFPIA accesses the CMU - make sure it still works however that is uses http://lbccontrol, not IP or longer name
  • test the fix for the PHP5 problem with XML files
  • I don't like the build for the libappio and the clearaccessio ; I think maybe these should be in a system dir - if not, then we should make the makefile build them from scratch - or copy them from SVN
  • add tklogger files before we turn off the old CMU; have to install tcl/tk wish
  • telescope user with aliases for start/stop/reboot
  • why does trajectory debug come out when you run testrotator ? it's hard-coded in testrotator
  • what is this in the /var/log/messages file?
    Apr 17 15:42:59 cmu kernel: molex driver - no more pci device !!! 
    Non problem - Molex says: After that the driver detect the card (card 1), the driver continue looking for other PCI card (because ApplicomIO can configure 8 cards) and because there is no second card, we receive the information message "molex driver - no more pci device !!!", informing that the driver doesn't find another card.
    In case of using two applicomIO card for example, the driver will detect and connect with the two cards then it will send message informing than there is no more pci device.

     Documentation updates:
  • add somewhere to populate the /images with the files necessary for windows PC(? or maybe not since we won't have them anymore)
  • now that I know more, need to document:
    • test.exe - things like you have to power up the CCD controller/shutter/temp sensor and make sure the instrument control is not running HK
    • crash recovery - won't do the powerup if you had only one side of HK enabled - it wants both
    • picture of the GUI and instrument control interactions - how the sw writes the uiupdate.php file as well as the lbc.status file and the GUI writes to the lbc.command pipe

16-Sep-2014

Looking back through logs, etc. to see what's been happening. 3 core's the night of 20140915

04:43, crash on TURNON (this was noted in the commissioning notes as an RPC error, but I don't see RPC complaints)

05:14, crash on preset (this is probably the one they really meant) with B CAMERA
RPC client is NULL during retries

08:50, crash on TURNON

What are these from?
2014/09/15 09:34:19.368279 W R ROTATOR           no delay, late by -4.9082 secs...
2014/09/15 09:35:23.048287 W B ROTATOR           no delay, late by -5.0105 secs...
2014/09/15 09:41:28.348306 W R ROTATOR           no delay, late by -4.9463 secs...
2014/09/15 09:42:38.288296 W R ROTATOR           no delay, late by -4.9650 secs...
2014/09/15 09:44:33.348283 W B ROTATOR           no delay, late by -4.9867 secs...
2014/09/15 09:49:48.208321 W B ROTATOR           no delay, late by -4.9290 secs...
2014/09/15 09:53:05.268290 W R ROTATOR           no delay, late by -4.9766 secs...
2014/09/15 09:56:19.768293 W B ROTATOR           no delay, late by -5.0473 secs...
2014/09/15 10:36:22.688302 W R ROTATOR           no delay, late by -5.0121 secs...
2014/09/15 11:28:46.068296 W R ROTATOR           no delay, late by -4.9713 secs...
2014/09/15 11:30:25.228343 W B ROTATOR           no delay, late by -4.9742 secs...
2014/09/15 11:47:52.328272 W B ROTATOR           no delay, late by -4.9333 secs...
2014/09/15 11:49:33.248324 W R ROTATOR           no delay, late by -4.9449 secs...
They correlate with long load times:
2014/09/15 09:34:19.368230 N R ROTATOR           loading tracking trajectory completed 3 points in 11.442 seconds
2014/09/15 09:35:23.048232 N B ROTATOR           loading tracking trajectory completed 3 points in 11.545 seconds
2014/09/15 09:41:28.348263 N R ROTATOR           loading tracking trajectory completed 3 points in 11.480 seconds
2014/09/15 09:42:38.288244 N R ROTATOR           loading tracking trajectory completed 3 points in 11.499 seconds
2014/09/15 09:44:33.348240 N B ROTATOR           loading tracking trajectory completed 3 points in 11.521 seconds
2014/09/15 09:49:48.208273 N B ROTATOR           loading tracking trajectory completed 3 points in 11.463 seconds
2014/09/15 09:53:05.268246 N R ROTATOR           loading tracking trajectory completed 3 points in 11.511 seconds
2014/09/15 09:56:19.768234 N B ROTATOR           loading tracking trajectory completed 3 points in 11.581 seconds
2014/09/15 10:36:22.688260 N R ROTATOR           loading tracking trajectory completed 2 points in 11.368 seconds
2014/09/15 11:28:46.068240 N R ROTATOR           loading tracking trajectory completed 6 points in 12.039 seconds
2014/09/15 11:30:25.228275 N B ROTATOR           loading tracking trajectory completed 6 points in 12.042 seconds
2014/09/15 11:47:52.328228 N B ROTATOR           loading tracking trajectory completed 7 points in 12.179 seconds
2014/09/15 11:49:33.248269 N R ROTATOR           loading tracking trajectory completed 7 points in 12.191 seconds

3-Sep-2014

Status after five nights on-sky (20140830,20140831,20140901,20140902,20140903)

  • sometimes the software crashes and restarts, but we do not get a core file - why?
  • found many php bugs - undefined vars, etc. the biggest was the uiupdate.tag file didn't have the "shortcut" fixed for the php keyword - this was causing the buttons to not update until the 5min timeout specified by the java script on the pages
  • for some reason, we core'ed several times in the rotator tracking function which JMH says we do not use - he instructed me to take it out so I made it false in the OB preparation instead of true
  • crashed several times in cleaning up the rotator trajectory; this may just be sloppiness, we weren't NULL'ing the pointer initially or after a free; also I was free'ing even if I didn't do the get
  • had to reboot the CMU once to get the GUI to do anything - it may have been the /var/log/httpd/error_log being pounded; just stopping and starting the httpd daemon didn't fix it
  • since red had shutter and data probe problems, it was down a lot, we couldn't even run HK on it - this caused us to be in a funky power state and some bugs were found
    • if you only have one side HK up and you kill/crash, it only brings up HK because it will not issue a POWERON if both sides aren't up - this is when you get the "inhibited channels must be explicitly disabled" message; it wants you to disable in the redchannel/bluechannel.conf file.
    • power.conf does not handle re-connecting to just one side of the telescope (I added it)

26-Aug-2014

Coming out of shutdown, we will use the new CMU (version 2.0 of linux software).

What is the procedure to go back to 1.7 - the version from Feb-2014?

  1. change IP addresses on old-cmu and cmu
  2. move profibus connection in CRB from new cmu to older one
  3. samba configuration?
  4. windows PCs are using cmu.lbc.lbto.org in the go.bat on the machine; the go.bat sitting on /images is using CMU - which maybe ok since it will get that version if we go back to old CMU; but we actually had to get it kicked or something - it needed
    go.bat copies update.bat

Stephen got it set up:
 [root@rm580f-1 ~]# host cmu.lbc.lbto.org
cmu.lbc.lbto.org has address 192.168.59.101
[root@rm580f-1 ~]# host lbccontrol.mountain.lbto.org
lbccontrol.mountain.lbto.org is an alias for cmu.lbc.lbto.org.
cmu.lbc.lbto.org has address 192.168.59.101
[root@rm580f-1 ~]# host cmu-old.lbc.lbto.org
cmu-old.lbc.lbto.org has address 192.168.59.10 

He said: I changed the hostnames on the machines themselves as well, but I usually like to reboot them to make sure that it isn't stuck somewhere (like crond and ovms). I figure the power outage will take of that, but just be aware that you may still see interesting messages.
I changed the netbios name of the old cmu from "CMU" to "cmu-old". I think it was probably the NetBIOS name that was causing the weirdness with your Windows mounts yesterday. The file to change there is /etc/samba/smb.conf in case you need to fully revert.

To change the IP address:
more /etc/sysconfig/network
NETWORKING=yes
HOSTNAME=cmu.lbc.lbto.org 
Run the hostname command to reset
Run /etc/init.d/network restart


We had some complaints on the Windows side early on:
... has remembered the connection 
Had to do a net use delete to get rid of it

Also - had to change the precedence in lmhosts for DNS

18-Aug-2014

First time anything was powered up since the roof leaking early in summer shutdown.
JamesR took a look at the electronics for water damage while working heater relay wiring issues. Robert and John put the filters into both sides and I was going to test the balance, etc.

dataprobe problems when running as non-root.

Robert and JohnM got the filters installed on red side by 1:45. Of course, the software was problematic when I first tried connecting to red, so they moved to the blue side to install. I was not able to cleanly connect and run the red filter wheels until almost 3pm. On the lift, we ran through parking and 10 random filters.

Blue was much more problematic. While trying to connect, we had to power-cycle the dataprobes a couple of times (this takes a lift to the electronics box) and once I decided to run as root only, we had fewer dataprobe problems and then got profibus complaints. With some research into drawings and documents we went back onto the lift to check connections between the LBC hub and electronics boxes. No smoking guns - IT 5207.

  • I need to track down why the dataprobe access needs to be root -- although it's not clear why some things work and others don't. Why does power.py have no problems toggling power and seeing status as lbccontrol ? The same is true of =testpower3 .
  • SourceExtractor results are different between the current version and the old version on the Windows PCs - need to make sure I understand if it's better or worse. Windows side is using 2.2.2 (Sep-2001); current LBT version is 2.8.6 (Apr-2013)
  • the profibus problem is hardware - I am confident it's not related to the new board in the new CMU or software. It works on red side and it worked on both sides before shutdown.

25-July-2014

Tested the PHP wrapper for use of DOMDocument - I think the updated OB generation will work.

Moved the applicom driver build to /lbt/lbc_runtime/applicomio

15-July-2014 Testing

On the mountain with INAF guys. Set up the IBM test PC for them in the lab. It's still running XP from our testing last summer. We don't have the XP ethernet driver (well, I don't anyway), so it does not have network on it.

Ran testprofibus on the old CMU and it does not work as non-root user either. I remembered the installation of the applicomio module also creates a /dev/ac on the filesystem. I looked at that and changed the permissions to be 666. That fixed running the testprofibus as non-root user!! Changed the startup script /etc/init.d/applicomio to do the chmod .

Tested one of the Windows PCs using the cmu.lbc.lbto.org name in the bat file instead of the IP address and it was OK. So, now that the DNS/search orders/IP addresses are all sorted this is happy now.

9-July-2014 Testing

Housekeeping
I configured the two portservers according to: http://lbccontrol/devices/DiGiPortServerTS16/configuration.shtml
but minicom still wasn't communicating with the shutter.
finally, going back and forth with the old CMU, got the settings correct to make it communicate via minicom, but HK wouldn't work as lbccontrol - wouldn't read the portserver power status....

Had to install gnuplot and change some settings in the tag file to make the plot work.

 [lbccontrol@cmu-new current]$ which gnuplot
/usr/bin/which: no gnuplot in (.:/usr/lib/qt-3.3/bin:/usr/local/bin:/bin:/usr/bin:/usr/local/sbin:/usr/sbin:/sbin:/home/lbccontrol/bin)
[lbccontrol@cmu-new current]$ sudo su
[sudo] password for lbccontrol: 
[root@cmu-new 2.0]# yum install gnuplot
Loaded plugins: fastestmirror, refresh-packagekit, security
Loading mirror speeds from cached hostfile
 * base: repos.lax.quadranet.com
 * extras: mirrors.easynews.com
 * updates: repos.dfw.quadranet.com
base                                                                                                                                             | 3.7 kB     00:00     
extras                                                                                                                                           | 3.4 kB     00:00     
updates                                                                                                                                          | 3.4 kB     00:00     
updates/primary_db                                                                                                                               | 3.4 MB     00:00     
Setting up Install Process
Resolving Dependencies
--> Running transaction check
---> Package gnuplot.i686 0:4.2.6-2.el6 will be installed
--> Processing Dependency: gnuplot-common = 4.2.6-2.el6 for package: gnuplot-4.2.6-2.el6.i686
--> Running transaction check
---> Package gnuplot-common.i686 0:4.2.6-2.el6 will be installed
--> Finished Dependency Resolution

Dependencies Resolved

========================================================================================================================================================================
 Package                                       Arch                                Version                                    Repository                           Size
========================================================================================================================================================================
Installing:
 gnuplot                                       i686                                4.2.6-2.el6                                base                                430 k
Installing for dependencies:
 gnuplot-common                                i686                                4.2.6-2.el6                                base                                552 k

Transaction Summary
========================================================================================================================================================================
Install       2 Package(s)

Total download size: 981 k
Installed size: 2.4 M
Is this ok [y/N]: y
Downloading Packages:
(1/2): gnuplot-4.2.6-2.el6.i686.rpm                                                                                                              | 430 kB     00:00     
(2/2): gnuplot-common-4.2.6-2.el6.i686.rpm                                                                                                       | 552 kB     00:00     
------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Total                                                                                                                                   844 kB/s | 981 kB     00:01     
Running rpm_check_debug
Running Transaction Test
Transaction Test Succeeded
Running Transaction
  Installing : gnuplot-common-4.2.6-2.el6.i686                                                                                                                      1/2 
  Installing : gnuplot-4.2.6-2.el6.i686                                                                                                                             2/2 
  Verifying  : gnuplot-common-4.2.6-2.el6.i686                                                                                                                      1/2 
  Verifying  : gnuplot-4.2.6-2.el6.i686                                                                                                                             2/2 

Installed:
  gnuplot.i686 0:4.2.6-2.el6                                                                                                                                            

Dependency Installed:
  gnuplot-common.i686 0:4.2.6-2.el6                                                                                                                                     

Complete!

It's lbc.c in the housekeeping thread that actually creates the png files. Tried from the command line, it didn't like the fonts. commented out the set linestyle lines and now it does blue, but not red, complains:
[root@cmu-new dynamic]# gnuplot hkplot-test.plt 
Could not find/open font when opening font "arial", using internal non-scalable font

set logscale y1 10
              ^
"hkplot-test.plt", line 33: unknown axis

note: when changing the tag file (hkplot.tag) in UI area, have to restart lbc software to pick it up.

Windows PC updates
SAMBA is not connecting to the correct machine.

Even when profibus unhappy, filters come out in "enabled", ...ok not "ready"

Complaints from windows pc about not finding BTech/flat.zip; none of these were populated on the /images dir

Turned on the power to profibus and portservers manually tried the testprofibus again (after clearaccessio) and it worked!
18:33
Try from lbc again?

18:51 top-level software is strugglingto turn on the portservers - why?
2014/07/09 18:51:29.262204 N B HKEEPING          device "192.168.59.114" outlet 2 switching checking
2014/07/09 18:51:39.262508 N R HKEEPING          device "192.168.59.124" outlet 2 switching checking failed, rc=-1
2014/07/09 18:51:39.262686 N B HKEEPING          device "192.168.59.114" outlet 2 switching checking failed, rc=-1
2014/07/09 18:51:49.263219 N R HKEEPING          device "192.168.59.124" outlet 2 switching checking failed, rc=-1
2014/07/09 18:51:49.263262 N B HKEEPING          device "192.168.59.114" outlet 2 switching checking failed, rc=-1
But, they worked when I tried the turn on again - now they're up. (HK is up)

Will profibus work this time? and the windows PCs come up?

I think it must have core'd -- new "session started" at 18:54

ulimit -a shows 0 core file size...

What are these?
lbc.log:2014/07/09 16:45:15.751007 S B HKEEPING          powerdump: no device in use
lbc.log:2014/07/09 17:06:36.916787 S B HKEEPING          powerdump: no device in use
lbc.log:2014/07/09 17:30:52.955267 S B HKEEPING          powerdump: no device in use
lbc.log:2014/07/09 18:06:38.094421 S B HKEEPING          powerdump: no device in use
lbc.log:2014/07/09 18:22:56.357959 S B HKEEPING          powerdump: no device in use
lbc.log:2014/07/09 18:29:18.064448 S B HKEEPING          powerdump: no device in use
lbc.log:2014/07/09 18:47:07.016384 S B HKEEPING          powerdump: no device in use
lbc.log:2014/07/09 18:54:31.609965 S B HKEEPING          powerdump: no device in use
if PowerDump is called when there are no devices in the list....

YEA...everything came up at 19:18....

Brought down to test filters standalone. Why won't it run as lbccontrol?? ugh - hung with just this in the filters.log:
 2014/07/09 19:30:41.904025 < Log Service: session started >  

We have a blash offset for Wheel2 on blue?? could that be causing our problems in production?
no -- the problem in production is wheel 1

Ran through 10 random filters for both red and blue. See filters.log file in CMURehost on ksummers.

When profibus is powered up, all three leds on the splitter are yellow, the card in the back of the new CMU is green (led on the card).
when not powered up, the "Ch 2" (RED) and "Ch 3" (BLUE) leds on the splitter are red and the led on the card is yellow.
ProfibusSplitterLEDs-NoPowerOnTel.JPG

Trying testrotator. Failed with "applicom driver not accessible"; this is with lbccontrol
try root (20:08): ... worked -- why?
try red again, as lbccontrol 20:14 -- same thing. This is a permissions problem with the applicom module!!

why does the tracking debug come out when you run testrotator?

20:32 -- restart everything in overall software

Could not import from the fast OB page. Page goes blank and nothing happens in the high-level software. I was able to browse to an OB I created yesterday (in Repository/20140708) and import that one and run it.
http://forums.phpfreaks.com/topic/241116-is-there-any-equivalent-function-of-dump-mem-in-php5/
If not authorized, do I not set the FACSUM variables in IIF? I didn't see the seq updates, etc.

The windows boot drive has been put in the cabinet in CRB on a shelf - it is clearly labelled, and has the USB cable attached.
The windows boot drive is in the LBC cabinet 1, in a box - labelled and has the USB cable.


8-July-2014 Testing

Complaining about applicom driver.

When i run it without profibus attached on old CMU:
[root@CMU lbccontrol]# ./testprofibus

------------------------------
 Profibus Service test program
-------------------------------

Opening Log Service ...
ProfiBus Blue Encoder1 Initialization ...
ProfiBus Blue Encoder2 Initialization ...
ProfiBus Blue Encoder3 Initialization ...
ProfiBus Red Encoder1 Initialization ...
ProfiBus Red Encoder2 Initialization ...
ProfiBus Red Encoder3 Initialization ...
Reading 3 red and 3 blue encoders 30 times ...
1404861384 --&gt; ERROR-B-enc1      ERROR-B-enc2      ERROR-B-enc3      ERROR-R-enc1      ERROR-R-enc2      ERROR-R-enc3     

This is not the error I'm getting now - it's complaining
2014/07/08 22:16:57.629153 N B PROFIBUS ENCODER1 Profibus board 1 device 10: board initialization error 93 (applicom driver not accessible) 

Also, TCS is complaining. Why?

Jul  8 22:52:52 cmu-new LBT: The configuration file /lbt/tcs/current/tcs/etc/tcs.conf was added.
Jul  8 22:54:57 cmu-new LBT: Using IIF IP address 127.0.0.1
Jul  8 22:54:57 cmu-new LBT: connect failure to RPC server on 127.0.0.1 at port 50021 for function IIF:Authorize:execute (errno 111)
Jul  8 22:54:57 cmu-new LBT: IIF::init() failed: IIF Subsystem cannnot be found
Jul  8 22:54:57 cmu-new LBT: IIF::init() failed: IIF Subsystem cannnot be foundUsing IIF IP address 127.0.0.1

Old CMU has:

Jul  7 10:24:25 CMU LBT: Using IIF IP address 192.168.58.17
Jul  8 02:46:25 CMU LBT: Using IIF IP address 192.168.58.17
Jul  8 18:25:12 CMU LBT: Using IIF IP address 192.168.58.17
Jul  8 23:07:38 CMU LBT: Using IIF IP address 192.168.58.17

Stephen changed the DNS search order(?) and that made it correct.

error messages wrong in HK:
2014/07/09 00:03:14.998150 W -                   BLU Channel HouseKeeping data file not found [successfully completed:134953133]
2014/07/09 00:03:14.998251 W -                   RED Channel HouseKeeping data file not found [successfully completed:134953133]

housekeeping is not connecting; and I fixed the above complaint.

installed minicom to take a look:
[lbccontrol@cmu-new conf]$ sudo yum install minicom
[sudo] password for lbccontrol: 
[lbccontrol@cmu-new conf]$ sudo yum install minicom
[sudo] password for lbccontrol: 
Loaded plugins: fastestmirror, refresh-packagekit, security
Loading mirror speeds from cached hostfile
 * base: centos.mirrors.hoobly.com
 * extras: centos.mirror.lstn.net
 * updates: centos.host-engine.com
base                                                                                                                                   | 3.7 kB     00:00     
extras                                                                                                                                 | 3.4 kB     00:00     
updates                                                                                                                                | 3.4 kB     00:00     
updates/primary_db                                                                                                                     | 3.4 MB     00:03     
Setting up Install Process
Resolving Dependencies
--&gt; Running transaction check
---&gt; Package minicom.i686 0:2.3-6.1.el6 will be installed
--&gt; Processing Dependency: lrzsz for package: minicom-2.3-6.1.el6.i686
--&gt; Running transaction check
---&gt; Package lrzsz.i686 0:0.12.20-27.1.el6 will be installed
--&gt; Finished Dependency Resolution

Dependencies Resolved

==============================================================================================================================================================
 Package                              Arch                              Version                                       Repository                         Size
==============================================================================================================================================================
Installing:
 minicom                              i686                              2.3-6.1.el6                                   base                              242 k
Installing for dependencies:
 lrzsz                                i686                              0.12.20-27.1.el6                              base                               66 k

Transaction Summary
==============================================================================================================================================================
Install       2 Package(s)

Total download size: 308 k
Installed size: 943 k
Is this ok [y/N]: y
Downloading Packages:
(1/2): lrzsz-0.12.20-27.1.el6.i686.rpm                                                                                                 |  66 kB     00:00     
(2/2): minicom-2.3-6.1.el6.i686.rpm                                                                                                    | 242 kB     00:00     
--------------------------------------------------------------------------------------------------------------------------------------------------------------
Total                                                                                                                         844 kB/s | 308 kB     00:00     
Running rpm_check_debug
Running Transaction Test
Transaction Test Succeeded
Running Transaction
  Installing : lrzsz-0.12.20-27.1.el6.i686                                                                                                                1/2 
  Installing : minicom-2.3-6.1.el6.i686                                                                                                                   2/2 
  Verifying  : lrzsz-0.12.20-27.1.el6.i686                                                                                                                1/2 
  Verifying  : minicom-2.3-6.1.el6.i686                                                                                                                   2/2 

Installed:
  minicom.i686 0:2.3-6.1.el6                                                                                                                                  

Dependency Installed:
  lrzsz.i686 0:0.12.20-27.1.el6                                                                                                                               

Complete!

How does minicom get to the serial devices? They're on portservers, but not directly on the CMU...

Outstanding as of Tucson testing, before on the mountain

  • dgrp daemon not starting on reboot because it is not configured for the port servers yet
  • Add to portserver configuration.shtml the command /usr/bin/dgrp/config/dgrp_cfg_node init R 192.168.59.122 16
  • issues with the lbc.command file - when I create it like it is on the existing CMU, the gui hangs... (pipe with just write access)
  • what is this in the /var/log/messages file?
    Apr 17 15:42:59 cmu kernel: molex driver - no more pci device !!! 
  • add notes to installation procedures for how to boot in windows
  • tftp is not installed
  • when we go to the mountain, set up the NTP sync
  • add apache config info to web page somewhere
  • get faulhaber driver from current CMU in Windows mode?

  • no IDL license, so lbciaRun hangs when it tries to run
  • image magik convert program needs LD_LIBRARY_PATH to run - but it's the same exe as from Tucson systems - why?
  • do we need to restart winbind daily? (see IT 3292)

Details (April-2014)

Two Dell servers set up the week of 15-April-2014.

using local /lbt/lbc_runtime for:
  • TCS
  • cfitsio
  • sextractor
  • IDL
  • image magik
php issues "short notation"

changed paths so we run from /lbccontrol/current instead of /home/lbccontrol
lbccontrol is its own disk

try to do everything as the lbccontrol user not root

cron jobs on current CMU

-rwxrwxrwx    1 root     root           28 Aug  5  2003 00-logwatch -> ../log.d/scripts/logwatch.pl
-rwxr-xr-x    1 root     root          135 Jan 26  2003 00webalizer
-rwxr-xr-x    1 root     root          276 Jan 24  2003 0anacron
-rwxr-xr-x    1 root     root          308 Sep 12  2013 checkDisk
-rwxr-xr-x    1 root     root           86 Oct 10  2013 chmodHK
-rwxr-xr-x    1 root     root           51 Jan 24  2003 logrotate
-rwxr-xr-x    1 root     root          418 Feb 10  2003 makewhatis.cron
-rwxr-xr-x    1 root     root          100 Oct 15  2011 reset_winbindd
-rwxr-xr-x    1 root     root          104 Feb 27  2003 rpm
-rwxr-xr-x    1 root     root          132 Feb 19  2003 slocate.cron
-rwxr-xr-x    1 root     root          100 Feb 19  2003 tetex.cron
-rwxr-xr-x    1 root     root          193 Feb 10  2003 tmpwatch

I Attachment Action Size Date Who Comment
LBCDailyCronJobs.txttxt LBCDailyCronJobs.txt manage 25 K 29 May 2014 - 23:06 UnknownUser Dump of all the cron.daily files for safekeeping
ProfibusSplitterLEDs-NoPowerOnTel.JPGJPG ProfibusSplitterLEDs-NoPowerOnTel.JPG manage 210 K 21 Jul 2014 - 20:34 UnknownUser ProfiBus splitter LEDs with no power on the telescope
Topic revision: r30 - 18 May 2016, KelleeSummers
This site is powered by FoswikiCopyright © by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding Foswiki? Send feedback