Editing Scanco Administration (section)

==Administrative Commands==

===User accounts===

The <code>xtremect2</code> and <code>xt2recalib</code> passwords are the same and have all the same settings, except data disks DK0 points to DISK2 or to DISK6, respectively.

The <code>bonelab</code> and <code>bonelabknee</code> passwords are the same and have all the same settings, except they point to DISK2 and DISK6, respectively.

Any users who want to work at a VMS terminal should log in with <code>bonelab</code> or <code>bonelabknee</code>. This will ensure <code>uct_evaluation</code> is accessing the correct data.

Any user who wants to remotely log into the system will be using SSH authentication at users <code>xtremect2</code> or <code>xt2recalib</code>. Access by SSH requires Steve to set up your SSH public key. He has instructions on how to do this in his list of SOPs.

Any users who need to scan patients log into <code>xtremect2</code>, <code>microct2</code> or <code>xt2recalib</code> should use a password. This password is restricted to only a small handful of people.

====Set password on VMS system====

For OpenVMS access (which includes FTP, SSH as well) you can simply use:

 $ set password

Also see https://wiki.vmssoftware.com/SET_PASSWORD

We usually do the password change from system account for any users. It's easier as you can change any users password in the
same tool. Here how this goes (as system):

 $ cd sys$system
 $ mcr authorize
 UAF> mod  microct /pass=test1234 /nopwdlife /nopwdexp
 ... next user if wished ...
 UAF> exit

Then you may have to update the password for Samba access as well, if users use Samba. And that's also a reason we do it all as
system, as this can only be done there anyway.

 $ @samba$root:[bin]samba$define_commands
 $ pdbedit -x  microct
 $ pdbedit -a  microct

and enter new password.

===Query commands===
====Check licenses on VMS system====

This can be used to see what educational licenses (and other licenses) are currently on the system:

<pre>
$ SHOW LICENSE
</pre>

To apply for an educational license go to http://www.openvmsedu.com


[[Image:Redirect.jpg|15px]] [[#top|''Return to Contents'']] or [[Main_Page|''Return to Main Page'']]

====Check MAC address of VMS system====
You can check them using the command
  $ mcr lancp sh dev/char
For example, the result on percy.kin.ucalgary.ca gives

  Device Characteristics EIA0:
                  Value  Characteristic
                  -----  --------------
                   1500  Device buffer size
                 Normal  Controller mode
               External  Internal loopback mode
      00-02-A5-20-29-99  Hardware LAN address
                         Multicast address list
                CSMA/CD  Communication medium
      FF-FF-FF-FF-FF-FF  Current LAN address
                    128  Minimum receive buffers
                    256  Maximum receive buffers
                    Yes  Full duplex enable
                    Yes  Full duplex operational
            TwistedPair  Line media type
                    100  Line speed (mbps)
    Disabled/No Failset  Logical LAN state
                      0  Failover priority


[[Image:Redirect.jpg|15px]] [[#top|''Return to Contents'']] or [[Main_Page|''Return to Main Page'']]

===Update commands===

====Change hosts====
<pre>
$ mcr latcp show service *                      ! and if it shows you SCTHIN:
$ set host scthin                               ! or
$ set host scthin/lat
</pre>


[[Image:Redirect.jpg|15px]] [[#top|''Return to Contents'']] or [[Main_Page|''Return to Main Page'']]

====Update OpenVMS Licenses====
This needs to be done on all the Calgary workstations, XtremeCT, vivaCT40/uCT35, and uCT100.<br>
Request licenses from HP:
*School Identification (for U of C): 855803-9502
*License type: OpenVMS VAX/Alpha (uCT100: OpenVMW Integrity) Layered Products?  (Or try others)
*Member of DSPP or AlicanceONE: No
*Do you use a VAX or Alpha Emulator: No

On every workstation, you need to check which licenses are loaded and update those with a termination date.
<pre>
$ show license

Active licenses on node SCEVA3:

------- Product ID --------    ---- Rating ----- -- Version --
Product            Producer    Units PCL   Activ Version Release    Termination
C                  DEC             0  0     100    0.0  (none)       1-OCT-2012
CMS                DEC             0  0     100    0.0  (none)       1-OCT-2012
CXX-V              DEC             0  0     100    0.0  (none)       1-OCT-2012
FORTRAN            DEC             0  0     100    0.0  (none)       1-OCT-2012
LSE                DEC             0  0     100    0.0  (none)       1-OCT-2012
MMS                DEC             0  0     100    0.0  (none)       1-OCT-2012
VMSCLUSTER         DEC             0  0     100    0.0  (none)       1-OCT-2012
OPENVMS-I64-FOE    HP              4  1     0      0.0  (none)      (none)
</pre>
 

*HP will provide you with a script file ([[media:Openvms_license_update_hp.docx |'''Example script from HP''' ]]) containing a large number of licenses. 
*Modify the script to contain only the relevant licenses and load the licenses ([[Media:Openvms_license_update.docx |'''Example modified script''']]).
*Login as user: system
*Re-check the licenses on the workstation to ensure that the termination date has been updated.
*The date after expiry, if cannot log in, then there is a problem with the VMS Cluster license, which was installed by Scanco. Contact Mark Blanchard (mblanchard@scanco.ch) about this problem.


[[Image:Redirect.jpg|15px]] [[#top|''Return to Contents'']] or [[Main_Page|''Return to Main Page'']]

====Understanding how to set default parameters with OpenVMS====

Here is how the files all hang together...

LOGIN.COM is called automatically during login and calls
::MICROCT_SETUP.COM, which sets the scanner ID, data device, etc. This file calls
::::MICROCT_SETUP_COMMON.COM to create general settings. This file is located in DISK0:[MICROCT_COMMON] and calls
::::::MICROCT_FE.COM, if present, to setup the environment for Bert's FE program
::::::MICROCT_SETUP_SITE_SPECIFIC.COM to create local settings for all accounts (e.g. ICALC)
::::::MICROCT_SETUP_COMMON_*.COM, if present, typically to create application-specific settings for all accounts
::::::MICROCT_SETUP_GROUP_*.COM, if present, to create local settings for specific groups (e.g. group 3401)
::MICROCT_SETUP_USER.COM to create user-specific settings.

Note that the $UM designation is the MAIN directory, and the location of the binary files. To see a list: 

  $ dir DISK0:[MICROCT_COMMON.MAIN]

This is where you'll find IPL_SCANCO_FE.EXE, IPL_SCANCO_M.EXE, etc.

Lots of stuff here:

  Directory DISK0:[MICROCT_COMMON]

  3309.DIR;1          DATABASE.DIR;1      DEMO.DIR;1          EVAL.DIR;1         
  MAIN.DIR;1          TEMPLATES.DIR;1     TIPS.DIR;1          UIL.DIR;1          
  UTIL.DIR;1          WWW.DIR;1

UE and other symbols like UT, UM, UU, etc, are kind of search-lists of directory. As you see in the example below:
<pre>
$ dir ue:iplv6_upat_redo_e5e6_xt2*.com /size/date

Directory IDISK1:[XTREMECT2.][EVAL]

IPLV6_UPAT_REDO_E5E6_XT2_FIX_GOBJ.COM;2
                        15KB  19-DEC-2019 09:53:04.69
IPLV6_UPAT_REDO_E5E6_XT2_FIX_GOBJ.COM;1
                        15KB  26-JUL-2018 07:22:37.54
IPLV6_UPAT_REDO_E5E6_XT2_OLD.COM;1
                        15KB  19-DEC-2019 09:53:04.69

Total of 3 files, 45KB

Directory IDISK0:[MICROCT_COMMON.][EVAL]

IPLV6_UPAT_REDO_E5E6_XT2.COM;5
                        13KB   7-AUG-2017 14:31:15.91

Total of 1 file, 13KB
</pre>

UE shows files in the user specific eval directory and then the common files. "common" because these are common to all accounts on your system. 
If there is ever a file with the same file name in both locations, <b>it will always take the one on the specific directory</b>. Also if you save a file under ue:... it will be stored into your user specific directory.
Common directories are protected and files in there cannot be changed by a "normal" user account. Only calibration accounts can do so, after evaluated privileges with <code>$ spb</code>.

[[Image:Redirect.jpg|15px]] [[#top|''Return to Contents'']] or [[Main_Page|''Return to Main Page'']]

====Edit Symbol====

=====Edit SCANCO_SUBMIT_COMMAND=====

*Check the current setting:
<pre>
$ show sym *submit*
</pre>

In directory <code>DISK2:[MICROCT]MICROCT_SETUP_USER.COM</code> add the following line:
<pre>
$ SCANCO_SUBMIT_COMMAND :== SUBMIT/NOPRINT/QUE=SYS$SLOW /NOTIFY/LOG=SYS$SCRATCH:
</pre>
Then end session and log in again.  Definition of the symbol in the 
<code>MICRO_SETUP_USER.COM</code> supersedes the standard setting.


[[Image:Redirect.jpg|15px]] [[#top|''Return to Contents'']] or [[Main_Page|''Return to Main Page'']]

=====Edit AIX=====

[[Image:Bones2.jpg|20px]] <b>Warning:</b> Extreme caution!  Explicit permission for this change or similar changes <b>must</b> be obtained from Dr. Steven Boyd.

Log in as user calibration (Xtremect:  calibration;  VivaCT40:  calib5019;   uCT35:  calib2339), the password is the same as for the system user.
Set priviledges to bypass.
 $ set proc/priv=bypass
Or 
 $ spb
Change to the MICROCT_COMMANDS directory.
 $ cdc 
 $ sea *.com aix
 $ eve microct_setup_common.com
Edit the file which is called <code>MICROCT_SETUP_COMMON.COM</code>

''Another example'': To make the backspace/insert keys work properly, edit the <code>login.com</code> and insert the line
<pre>
$ set terminal/insert
</pre>
Anywhere suitable.


[[Image:Redirect.jpg|15px]] [[#top|''Return to Contents'']] or [[Main_Page|''Return to Main Page'']]

====Setup Time====
=====Change system date and time=====
The OpenVMS system should adjust for daylight savings, however, this is currently incorrect for the alpha systems.  The alpha systems will need to be manually moved forward in March and moved back in April. To do this, you will need to login as system
<pre>
$ set host 0
     user:  system
     password:
$ show time
$ set time=dd-mmm-yyyy:hh:mm:ss/cluster   !/cluster to change the time on all computers at once
$ set time=01-APR-2012:18:00/cluster      !Example:  Set the system to April 1, 2012, 6 pm.
</pre>


[[Image:Redirect.jpg|15px]] [[#top|''Return to Contents'']] or [[Main_Page|''Return to Main Page'']]

=====Set time zone=====
As user system:

 $ @sys$startup:utc$time_setup
Choose "CANADA" from the menu.
Choose "MOUNTAIN" from the menu.


[[Image:Redirect.jpg|15px]] [[#top|''Return to Contents'']] or [[Main_Page|''Return to Main Page'']]

=====Disable Distributed Time Synchronization Service (DTSS)=====
DTSS will automatically adjust the clocks for daylight savings (i.e., 2am to 1am on the 2nd Sunday in March, 2am to 3am on the 1st Sunday in November).  As user system:

 $ mcr ncl disable dtss
 $ mcr ncl delete dtss
 $ eve sys$startup:sylogicals.com

Add line to disable DTSS
  $ DEFINE/SYSTEM/EXEC NET$DISABLE_DTSS TRUE
Save and exit.
Now re-set the time zone.


[[Image:Redirect.jpg|15px]] [[#top|''Return to Contents'']] or [[Main_Page|''Return to Main Page'']]

=====Configure Network Time Protocol (NTP) server to synchronize clocks=====
*NTP must be configured for each node (computer).  Login as user system.
NTP can only synchronize the system clock if the error is less than 1000 seconds. 
 
Synchronize the system clock to UofC's NTP server (ensures time is correct to within 1000s):

 $ mcr sys$system:tcpip$ntpdate ntp.cpsc.ucalgary.ca

Enable NTP:

 $ cd sys$startup
 $ @tcpip$config
   3 - Server components
   12 - NTP
   1 – Enable service on this node
   E – Exit menu
   E – Exit configuration procedure
   N – don’t start service

Create the NTP configuration file:

 $ sys$sysroot:[tcpip$ntp]
 $ copy/log tcpip$ntp.template tcpip$ntp.conf
 $ eve tcpip$ntp.conf

Comment out existing peer and server entries (# for comments).
Add entries for local servers.
        server ntp.cpsc.ucalgary.ca
        server ntp1.cpsc.ucalgary.ca
Save and exit.
 

Start NTP:

 $ @sys$startup:tcpip$ntp_startup

Check the log:

 $ type/cont/int=1 sys$sysroot:[tcpip$ntp]tcpip$ntp_run.log

Sample log file:

 $ Set NoOn
 $ VERIFY = F$VERIFY(F$TRNLNM("SYLOGIN_VERIFY"))
   11 Apr 09:22:34 ntp[547708872]: ntpd version = 4.1.0
   11 Apr 09:22:34 ntp[547708872]: precision = 976 usec
   11 Apr 09:22:34 ntp[547708872]: drift file 'SYS$SPECIFIC:[TCPIP$NTP]TCPIP$NTP.DRIFT' does not exist: no such file or directory
   11 Apr 09:22:34 ntp[547708872]: will attempt to create drift file 
 


[[Image:Redirect.jpg|15px]] [[#top|''Return to Contents'']] or [[Main_Page|''Return to Main Page'']]

====Setup Email====
=====Setup email server=====
The SMTP protocol is not enabled by default. If you use <code>tcpip$config</code> you can switch on the mail server under Client Components. As system:

 $ set host 0
     user:  system
     password:
 $ @SYS$MANAGER:TCPIP$CONFIG
Or
 $ @tcpip$config
 -> 2 (client components)
 -> 6 (SMTP)
 -> 2 (Enable & Start services)

*Select option 2 (client components) from the menu.
*Select option 6 (SMTP) from the next menu to enable and start it.

Since your smtp server is not registered anywhere, it can be that emails sent from this server will be filtered out as spam on the way to your mailbox. In that case you would need a so called MX record for your mail server. Contact UofC IT to resolve this issue.


[[Image:Redirect.jpg|15px]] [[#top|''Return to Contents'']] or [[Main_Page|''Return to Main Page'']]

=====Send an email=====
Example:  Send an email without an attachment.
 $ mail sys$scratch:email_message.txt/subject="email subject" "hbuie@ucalgary.ca,mkan@ucalgary.ca"

Example:  Send an email with an attachment.
 $ MIME :== $MIME.EXE
 $ MIME MIME> NEW /NOEDIT sys$scratch:file.txt
 MIME> ADD /TEXT thisattached.txt
 MIME> ADD /BINARY thatattached.zip
 MIME> ADD /MESSAGE this_is_written.txt
 MIME> SAVE
 MIME> EXIT
 $ MAIL /SUBJECT="MIME-encoded Message" - sys$scratch:file.txt "hbuie@ucalgary.ca,mkan@ucalgary.ca"
 $
*MIME> ADD /TEXT *.txt does not work


[[Image:Redirect.jpg|15px]] [[#top|''Return to Contents'']] or [[Main_Page|''Return to Main Page'']]

===Action commands===
====Send a message to terminals====
<pre>
$ reply /all "Merry Christmas to Bone Imaging Lab!"

  Reply received on SCEVA2 from user XTREMECT2 at _SCEVA2$FTA8473:   09:21:58 
  Merry Christmas to Bone Imaging Lab!

  4 terminals have been notified on SCEVA2.
$
  1 terminal has been notified on SCSRV2.
$
  1 terminal has been notified on SC3309.
$
  3 terminals have been notified on SCEVA1.
$
</pre>


[[Image:Redirect.jpg|15px]] [[#top|''Return to Contents'']] or [[Main_Page|''Return to Main Page'']]

====<code>mcr</code> vs <code>run</code> command====

The command <code>mcr</code> is the same as <code>run</code>, but only with mcr you can hand over 'in-line' parameters to an .exe file:

<pre>$ run ut:ctheader.exe c000124.isq</pre>
does not work, but you  have to say
<pre>$ mcr ut:ctheader.exe c000124.isq</pre>

But funnily if you define a symbol in <code>microct_setup_user.com</code> (or <code>login.com</code>),

<pre>$ ctheader :== $ut:ctheader.exe</pre>

with just neither run nor mcr, but the 'funny' $ sign, then you can also hand over the ISQ name in-line.

Small detail with big effect...


[[Image:Redirect.jpg|15px]] [[#top|''Return to Contents'']] or [[Main_Page|''Return to Main Page'']]

====Using fda, pipe and search====

The <code>fda</code> (file do all) command in conjunction with <code>sys$pipe</code> and <code>search</code> can be quite helpful. Generally, use <code>fda</code> if you want to have a wildcard to capture many files in one command line, otherwise simply use <code>sys$pipe</code> and <code>search</code>.

It's easiest to simply see examples. In this case, read the ISQ headers and use <code>...</code> to include all subdirectories:

 $ fda ctheader dk0:[microct.data.0001234...]*.isq

Here we include all AIM files, no matter whether it's a C1234567.AIM or C1234567_SEG.AIM or whatever else:

 $ fda "aix -l" dk0:[microct.data.0001234.00005678]*.aim

Note the use of double-quotes around the target of the <code>fda</code> command when there are more than one strings in the command.

In this example we ensure the filename is exactly 8 characters using "%" instead of "*":

 $ fda aix dk0:[microct.data...]C%%%%%%%.AIM

[[Image:Bones2.jpg|20px]] <b>Warning! </b> It's a good practise to always test your wildcard usage with <code>dir</code> before applying a large command:

 $ dir dk0:[xxx...]*.aim

Combine <code>fda</code> with <code>sys$pipe</code> and <code>search</code>:

 $ pipe fda "aix -l" dk0:[microct...]*.aim | search sys$pipe "Slope"

And you can redirect the output into a file:

 $ fda/output=sys$scratch:all_headers.log aix dk0:[microct...]*.aim
 $ pipe fda type *.log | search sys$pipe "BV/TV" /output=BVTV.txt
 $ pipe fda/output=sys$scratch:zscan.txt "aix -l" C%%%%%%%.AIM | search sys$pipe "Original Creation-Date","Index Patient","Index Measurement","Patient Name","Site","Time"

Again, if the command consists of multiple words with spaces between (e.g. mcr ut:xxx.exe), then enclose the command in double-quotes.

 $ fda "mcr ut:uct_ext_header_display" dk0:[microct...]*.isq

Grab more than one line using <code>pipe</code> and <code>search</code> command:

 $ pipe aix *.aim | search sys$pipe "dim", "pos", "name"

Some common examples I use, without <code>fda</code>, are:

 pipe aix -l D0013580.AIM | search sys$pipe "Original file","Index Patient","Index Measurement","Patient Name","Site","Time"
 pipe ctheader D0013580.ISQ | search sys$pipe "Patient Name","Patient Index","Measurement Index","Creation Date","Site"

[[Image:Redirect.jpg|15px]] [[#top|''Return to Contents'']] or [[Main_Page|''Return to Main Page'']]

====Putting multiple commands on a single line====

The key is to use the <code>PIPE</code> command. Basically separate all commands with ;

<pre>
$ pipe  cd DISK2:[XTREMECT2.DATA.00003089.00016318]  ;  dir
</pre>

It has also some nice features of passing input and output along to the next command (see help), eg if you want to see the integration time of a certain scan :

<pre>
$ pipe ctheader C0014998.ISQ | search sys$pipe "Sampletime"
         Sampletime : 900000 [µs]
</pre>

You can even submit a multi-line command to a remote computer using SSH:

<pre>
$ ssh -p22 -c 3des-cbc -o KexAlgorithms=+diffie-hellman-group1-sha1 -o HostKeyAlgorithms=+ssh-dss microct@12.618.196.561 'commands.txt'
</pre>
where 'commands.txt' can be replaced by 'pipe  cd DISK2:[XTREMECT2.DATA.00003089.00016318]  ;  dir'

[[Image:Redirect.jpg|15px]] [[#top|''Return to Contents'']] or [[Main_Page|''Return to Main Page'']]

====Make commands available to all users with symbols or paths====

DCL$PATH is one way to do that, WE don't use it...
<pre>
$ def dcl$path disk1:[microct.main],disk1:[microct.util]
</pre>
then, if you type
<code>
$ uct_t
</code>
it will first try to

<code>
$ @uct_t.com
</code>
and then
<code>
$ r uct_t.exe
</code>

however, we usually define it by giving full path:

<pre>
$ progx :== $disk1:[microct.main]progx.exe
$ com_proc_x :== @disk1:[microct.main]procedure.com
</pre>

then, you can just type the shortcut ('symbol' in VMS language)

We use the 'cd' to set the default path, you may also use this in your command procedure:

<code>proc.com</code>:

<pre>
$ cd 'p1'
$ dir *.*/size

$ define COM disk1:[microct.commands]
$ p :== @COM:PROC.COM
</pre>

these both go into <code>login.com</code>, but see the Wiki entry above to confirm exactly where...

then, you can just type:

<pre>
$ p disk4:[microct.data.0000007.000009]*.ISQ
</pre>

[[Image:Redirect.jpg|15px]] [[#top|''Return to Contents'']] or [[Main_Page|''Return to Main Page'']]

====Stopping disconnected sessions====

As <code>system</code> user you can check if there are any disconnected sessions, with <code>$ show user/full</code>:

<pre>
$ sh user/full
      OpenVMS User Processes at  5-JUN-2021 03:14:06.45
    Total number of users = 2,  number of processes = 7

 Username Process Name     PID     Terminal
 MICROCT  DECW$MWM       00000EA4  MBA6597: (disconnected)
 MICROCT  DECW$SESSION   00000EA2
 MICROCT  VUE$MICROCT_3  00000EA6  MBA6601: (disconnected)
 MICROCT  VUE$MICROCT_4  00000EA7  (subprocess of 00000EA2)
 MICROCT  VUE$MICROCT_5  0000289E  (subprocess of 00000EA2)
 MICROCT  _FTA24:        00012284  FTA24:
 SYSTEM   _FTA30:        000129DC  FTA30:
</pre>

and kill any as follows:
<pre>
$ stop/id=00000EA4
$ stop/id=00000EA6
</pre>

There may also be FTP connections that aren't closed properly. You can check for open FTP connection with this:

<pre>
$ sh sys/proc=tcpip$ftp*
OpenVMS V8.3-1H1  on node SC4209    5-JUN-2021 03:12:53.14   Uptime  73 13:39:10
  Pid    Process Name    State  Pri      I/O       CPU       Page flts  Pages
00000426 TCPIP$FTP_1     LEF      9   786428   0 00:00:12.51      1089    751  N
000129E0 TCPIP$FTPC00046 HIB      4      106   0 00:00:00.03       431    353  N
</pre>

and again kill any with:

<pre>
$ stop/id=000129E0
</pre>

But <b>DO NOT</b> kill the <code>TCPIP$FTP_1</code>  as this is the actual FTP server process.

Remember, in a cluster there will almost always be other open sessions which have their DECW$SM.LOG files locked. You don't need to delete them all!

[[Image:Redirect.jpg|15px]] [[#top|''Return to Contents'']] or [[Main_Page|''Return to Main Page'']]