Hi Mathieu,

thanks for your contribution. Have you looked at the install.sh script I developed last week:

   https://daq00.triumf.ca/MidasWiki/index.php/Install_Script

which basically does the same, plus it modifies the environment and installs mhttpd as a service.

Actually I modeled the installation after the way Homebrew is installed in the first place (using curl).

I wonder if the two things can kind of be integrated. Would be great to get with brew always the newest midas version, and it would also 
check and modify the environment. 

If you tell me exactly what is wrong MidasConfig.cmake.in I'm happy to fix it.

Best,
Stefan

Thanks Stefan!
Actually, these two approaches are slightly different I guess:
- the installation script you are linking manages the
installation and the subsequent steps, but doesn't manage the dependencies: for instance on my machine, it didn't find root and so manalyzer
is built without root support.
Maybe this is just something to adapt?
Brew on the other hand manages root and so knows how to link these two
together.
- The nice thing I like about brew is that one can "ship bottles" aka compiled version of the code; it is great and fast for
deployment and avoid compilation issues.
- I like that your setup does deploy and launch all the necessary executables ! I know brew can do
this too via brew services (see an example here: https://github.com/Homebrew/homebrew-core/blob/HEAD/Formula/r/rabbitmq.rb#L83 ), maybe worth
investigating...?
- Brew relies on code tagging to better manage the bottles, so that it uses the tag to get a well-defined version of the
code and give a name to the version.
I had to implement my own tags e.g. midas-mod-2025-12-a to get a release.
I am not sure how to do in the
case of midas where the tags are not that frequent...

Thank you for the feedback, I will make the modifications (aka naming my formula
``midas-mod'') so that it doesn't collide with a future official midas one.

Concerning the MidasConfig.cmake issue, this is what I need
(note that the INTERFACE_INCLUDE_DIRECTORIES is pointing to 
/opt/homebrew/Cellar/midas/midas-mod-2025-12-a/)

set_target_properties(midas::midas PROPERTIES
  INTERFACE_COMPILE_DEFINITIONS "HAVE_CURL;HAVE_MYSQL;HAVE_SQLITE;HAVE_FTPLIB"
 INTERFACE_COMPILE_OPTIONS "-I/opt/homebrew/Cellar/mariadb/12.1.2/include/mysql;-I/opt/homebrew/Cellar/mariadb/12.1.2/include/mysql/mysql"
 INTERFACE_INCLUDE_DIRECTORIES "/opt/homebrew/Cellar/midas/midas-mod-2025-12-a/;${_IMPORT_PREFIX}/include"
  INTERFACE_LINK_LIBRARIES "/opt/
homebrew/opt/zlib/lib/libz.dylib;-lcurl;-L/opt/homebrew/Cellar/mariadb/12.1.2/lib/ -lmariadb;/opt/homebrew/opt/sqlite/lib/libsqlite3.dylib"
)

whereas by default INTERFACE_INCLUDE_DIRECTORIES points to the source code location (in the case of brew, something like /private/<some-
hash> ).
Brew deletes the source code at the end of the installation, whereas midas seems to rely on the fact that the source code is still
present...
Does it help?
A way to fix is to search for this ``/private'' path and replace it, but this isn't ideal I guess...
This is what I
did in the midas formula:
--------
# Fix broken CMake export paths if they exist
    cmake_files = Dir["#{lib}/**/*manalyzer*.cmake"]
   cmake_files.each do |file|
      if File.read(file).match?(%r{/private/tmp/midas-[^/"]+})
        inreplace file, %r{/private/tmp/midas-
[^/"]+},
prefix.to_s
      end
      inreplace file, %r{/tmp/midas-[^/"]+}, prefix.to_s if File.read(file).match?(%r{/tmp/midas-[^/"]+})
   end
    cmake_files = Dir["#{lib}/**/*midas*.cmake"]
    cmake_files.each do |file|
      if File.read(file).match?(%r{/private/tmp/midas-
[^/"]+})
        inreplace file, %r{/private/tmp/midas-[^/"]+},
prefix.to_s
      end
      inreplace file, %r{/tmp/midas-[^/"]+},
prefix.to_s if File.read(file).match?(%r{/tmp/midas-[^/"]+})
    end
-----

I guess this code could be changed into some bash commands and
added to your script?

Thank you very much again!
Mathieu


> Hi Mathieu,
> 
> thanks for your contribution. Have you looked at the
install.sh script I developed last week:
> 
>    https://daq00.triumf.ca/MidasWiki/index.php/Install_Script
> 
> which basically does the
same, plus it modifies the environment and installs mhttpd as a service.
> 
> Actually I modeled the installation after the way Homebrew is
installed in the first place (using curl).
> 
> I wonder if the two things can kind of be integrated. Would be great to get with brew always
the newest midas version, and it would also 
> check and modify the environment. 
> 
> If you tell me exactly what is wrong
MidasConfig.cmake.in I'm happy to fix it.
> 
> Best,
> Stefan

> Actually, these two approaches are slightly different I guess:
> - the installation script you are linking manages the
> installation and the subsequent steps, but doesn't manage the dependencies: for instance on my machine, it didn't find root and so manalyzer
> is built without root support.
> Maybe this is just something to adapt?

Yes indeed. From your perspective, you probably always want ROOT with MIDAS. But at PSI here we have several installation where we do not
need ROOT. These are mainly beamline control PCs which just connect to EPICS or pump station controls replacing Labview installations. All
graphics there is handled with the new mplot graphs which is better in some case.

I therefore added a check into install.sh which tells you explicitly if ROOT is found and included or not. Then it's up to the user to choose to
install ROOT or not.

> Brew on the other hand manages root and so knows how to link these two
> together.

If you really need it, yes.

> - The nice thing I like about brew is that one can "ship bottles" aka compiled version of the code; it is great and fast for
> deployment and avoid compilation issues.
> - I like that your setup does deploy and launch all the necessary executables ! I know brew can do
> this too via brew services (see an example here: https://github.com/Homebrew/homebrew-core/blob/HEAD/Formula/r/rabbitmq.rb#L83 ), maybe worth
> investigating...?

Indeed this is an advantage of brew, and I wholeheartedly support it therefore. If you decide to support this for the midas
community, I would like you to document it at 

  https://daq00.triumf.ca/MidasWiki/index.php/Installation

Please talk to Ben <bsmith@triumf.ca> who manages the documentation and can give you write access there. The downside is that you will
then become the supporter for the brew and all user requests will be forwarded to you as long as you are willing to maintain the package ;-)

> - Brew relies on code tagging to better manage the bottles, so that it uses the tag to get a well-defined version of the
> code and give a name to the version.
> I had to implement my own tags e.g. midas-mod-2025-12-a to get a release.
> I am not sure how to do in the
> case of midas where the tags are not that frequent...

Yes we always struggle with the tagging (what is a "release", when should we release, ...). Maybe it's the simplest if we tag once per month
blindly with midas-2026-02a or so. In the past KO took care of the tagging, he should reply here with his thoughts.


> Thank you for the feedback, I will make the modifications (aka naming my formula
> ``midas-mod'') so that it doesn't collide with a future official midas one.

Nope. The idea is that YOU do the future official midas realize from now on ;-)


> Concerning the MidasConfig.cmake issue, this is what I need ...

Let's take this offline not to spam others.

Best,
Stefan

Due my oversight, the code for extracting ODB dumps from MIDAS data files from rootana/old_analyzer/event_dump.cxx was missing in 
manalyzer.cxx.

This is now corrected, the new manalyzer command line flag is "--save-odb", to use it:

daq00:manalyzer$ ./manalyzer_test.exe --save-odb ~/git/midas/testexpt/run00002.mid.lz4
...
Saving begin of run ODB dump for run 2 from "/home/olchansk/git/midas/testexpt/run00002.mid.lz4" to "run2bor.json"
...
Saving end of run ODB dump for run 2 from "/home/olchansk/git/midas/testexpt/run00002.mid.lz4" to "run2eor.json"
...

manalyzer commit f4cbcb7426083edc9f74298965c90a3a91f461ab

K.O.

Dear all, I implemented an optional timeout for the wait ODBvalue command. The way it works is similar to the standard wait command:

WAIT ODBvalue, /Equipment/HV/Variables/Measured[3], <, 100, timeout, 60

where the "timeout" keyword start a countdown in seconds. If the ODB condition is not met after 60 seconds the sequencer moves on to the next line.

To use this feature you must recompile the msequencer, delete /Sequencer/State and start the freshly compiled msequencer. This will add two ODBs to the /Sequencer/State: "Timeout value" (the countdown) and "Timeout limit" (the limit given in the wait command).

I suggest that we add something similar to the pysequencer using the same ODBs.

> Dear all, I implemented an optional timeout for the wait ODBvalue command. The way it works is similar to the standard wait command:
> 
> WAIT ODBvalue, /Equipment/HV/Variables/Measured[3], <, 100, timeout, 60
> 
> where the "timeout" keyword start a countdown in seconds. If the ODB condition is not met after 60 seconds the sequencer moves on to the next line.
> 
> To use this feature you must recompile the msequencer, delete /Sequencer/State and start the freshly compiled msequencer. This will add two ODBs to the /Sequencer/State: "Timeout value" (the countdown) and "Timeout limit" (the limit given in the wait command).
> 
> I suggest that we add something similar to the pysequencer using the same ODBs.

How can the MSL code figure out if the wait succeeded or timed out?

Stefan

> 
> How can the MSL code figure out if the wait succeeded or timed out?
> 
> Stefan

You get a message, something like:
17:52:12.293 2026/05/29 [Sequencer,INFO] WAIT ODBValue timeout after 10.0 seconds: /Equipment/Test/Variables/V < 1 not satisfied

Do we need something else?

Zaher

> > 
> > How can the MSL code figure out if the wait succeeded or timed out?
> > 
> > Stefan
> 
> You get a message, something like:
> 17:52:12.293 2026/05/29 [Sequencer,INFO] WAIT ODBValue timeout after 10.0 seconds: /Equipment/Test/Variables/V < 1 not satisfied
> 
> Do we need something else?
> 
> Zaher

I mean how can the following code determine the timeout?

> > > 
> > > How can the MSL code figure out if the wait succeeded or timed out?
> > > 
> > > Stefan
> > 
> > You get a message, something like:
> > 17:52:12.293 2026/05/29 [Sequencer,INFO] WAIT ODBValue timeout after 10.0 seconds: /Equipment/Test/Variables/V < 1 not satisfied
> > 
> > Do we need something else?
> > 
> > Zaher
> 
> I mean how can the following code determine the timeout?

My intention with this was dealing with something like setting a cryostat temperature or any non-critical parameter. If it is not reached within a given timeout we give up and move on with the plan rather than sitting and wasting a whole night of beam. If your ODBvalue is "mission critical" then the wait command should not be used with a timeout. If you do use the timeout option then you will have to check in the following lines what is the state of your ODBvalue (very easy). To me this is the simplest and most useful way for our use case.

> > > > 
> > > > How can the MSL code figure out if the wait succeeded or timed out?
> > > > 
> > > > Stefan
> > > 
> > > You get a message, something like:
> > > 17:52:12.293 2026/05/29 [Sequencer,INFO] WAIT ODBValue timeout after 10.0 seconds: /Equipment/Test/Variables/V < 1 not satisfied
> > > 
> > > Do we need something else?
> > > 
> > > Zaher
> > 
> > I mean how can the following code determine the timeout?
> 
> My intention with this was dealing with something like setting a cryostat temperature or any non-critical parameter. If it is not reached within a given timeout we give up and move on with the plan rather than sitting and wasting a whole night of beam. If your ODBvalue is "mission critical" then the wait command should not be used with a timeout. If you do use the timeout option then you will have to check in the following lines what is the state of your ODBvalue (very easy). To me this is the simplest and most useful way for our use case.

I was more thinking like a return value 0/1 if the wait function. If you change the condition, you only have to change it in one location. More like normal C functions work.

Stefan 

Hello, 

The UCN group at TRIUMF has used the history system extensively with a MySQL backend. They see long wait times (up to 20 mins) to add new logged variables to MIDAS equipment. I did some testing and it seems to be a MySQL issue that is fixed in recent versions (UCN uses an older version of MySQL). 

Likely no update is needed to MIDAS source. 


Derek 

Hi,
 has anyone tried runnning midas frontend on a Linux running 
on a Motorola MVME167 motorola embedded CPU?
I have seen people running Linux on a MV167 
(http://www.sleepie.demon.co.uk/linuxvme/)
so in principle this can be done.

The reason I am asking is that we have a lot of them in house 
and we would like to avoid paying for VxWorks
(I have succesfully run Midas on a mvme167/VxWorks node)

Or maybe one has come up with a much better solution 
[short of dumping mv167 into a sewer :)]

Piotr

I would like to know whether all parameters used by the frontend code have to be in the "Experiment/
Run Parameters" section.  This section can become big and difficult to maintain, because it is one single 
big section of experim.h (EXP_PARAM_DEFINED).  I have parameters the various frontends read at the 
beginning of each run, which set the hardware settings of various devices.  I would like to place these in 
a section all their own, organized by device.  Is this doable? 

Hi Jan,

I usually keep under /Experiment/Run Parameters only those settings which are kind of "global" and thus of
interest to frontend *and* analyzer, like a run mode (data/calibration/cosmic/...). Settings more specific to a
frontend I keep under /Equipment/<name>/Settings where <name> is the equipment name the specific frontend
produces. In your case each frontend will then get its own tree (related to each fragment). Please note that
both discussed trees can contain a whole tree with subdirectories, which lets you organize your data better.

Best regards, Stefan.

   We've had mhttpd aborting regularly since upgrading from midas-1.9.3.  This
happens during elog queries, and is due to an elog file that was incorrectly
modified by hand.  The modification to the file occurred 6 months ago.
   el_retrieve(midas.c:15683) now has several assert statements, one of which
aborts the program on reading the bad entry.

   Why is assert used, instead of an error return from the function (if
necessary), and maybe an error message in the log file?  Assert statements are
often removed, using NDEBUG, for normal use.

Chris

   The problem elog entry had one character removed, so end-of-file came before
the end of the message.  This could probably occur without the file being
altered, if the disk containing the elog fills.

>    We've had mhttpd aborting regularly since upgrading from midas-1.9.3.  This
> happens during elog queries, and is due to an elog file that was incorrectly
> modified by hand.

(sorry for delayed reply, for reasons unknown, I did not get an email notice when this was posted)

Yes, I agree, error handling in midas elog code is insufficient (note missing error checks for
read() and lseek() system calls). Anything but "perfect" elog files would cause funny errors and
malfunctions.

>  The modification to the file occurred 6 months ago.
>    el_retrieve(midas.c:15683) now has several assert statements, one of which
> aborts the program on reading the bad entry.

I added those to fix problems with "broken last NN days" and with infinite looping in the elog code
that we observed in TWIST.

You are welcome to replace the assert() statements with proper error handling. I used to have some code
that could report the filename of the bad elog file. Can we also report the exact file location for broken
files.

Please send me the diff, I will commit it to midas cvs.

>    Why is assert used, instead of an error return from the function (if
> necessary), and maybe an error message in the log file?  Assert statements are
> often removed, using NDEBUG, for normal use.

I use assert() in several ways:

0) I want a core dump each time X happens. (This is the only reasonable action when facing memory/stack
corruption. The problems in the elog code were stack corruption).
1) "I am too lazy to write proper error handling code" so I just crash and burn. This includes the
case where "proper error handling" would be "too invasive".
2) the error is too bad (or too deep) and there is no reasonable way to recover. Print an error message
and dump core (for later analysis). I sometimes use "cm_msg(); abort()". (assert is "printf("error"); abort()")

Please refer to literature for philosophic discussions on uses of assert() (Argh! Stefan will have my
head again!), but I will mention that "abort() early, abort() often" I find very effective. BTW, this technique
is heavily used in the Linux kernel (oops(), bug(), panic()) with some good effect, too.

>    The problem elog entry had one character removed, so end-of-file came before
> the end of the message.  This could probably occur without the file being
> altered, if the disk containing the elog fills.

Yes, I think you are right. In TWIST, we have seen disk-full conditions break both elog and history.

K.O.

What is the api call to determine the index of the frontend when specifying the
-i parameter during execution of the frontend? 

> What is the api call to determine the index of the frontend when specifying the
> -i parameter during execution of the frontend? 

INT get_frontend_index();

- Stefan

When i compile the experiment example of 1.9.5 the problem happened:

adccalib.c: In function `INT adc_calib_init()':
adccalib.c:114: `H1_BOOK' undeclared (first use this function)
adccalib.c:114: (Each undeclared identifier is reported only once for each
   function it appears in.)
make: *** [adccalib.o] Error 1

my ROOT is 4.01 and Zlib is 1.2.2

> When i compile the experiment example of 1.9.5 the problem happened:
> 
> adccalib.c: In function `INT adc_calib_init()':
> adccalib.c:114: `H1_BOOK' undeclared (first use this function)
> adccalib.c:114: (Each undeclared identifier is reported only once for each
>    function it appears in.)
> make: *** [adccalib.o] Error 1
> 
> my ROOT is 4.01 and Zlib is 1.2.2

We're in the process of fixing in the proper manner this problem, in the mean time
please add to the analyzer makefile the definition: -DUSE_ROOT at the line:
...
ROOTCFLAGS += -DHAVE_ROOT -DUSE_ROOT