Hi All,

I've installed midas (Rev:5294) on SLC6.3 (64bit), along with recent trunk versions of rootana and roody. 
All the packages compiled OK. The example code in $MIDASSYS/examples/experiment also runs OK 
provided that I don't launch roody. If I try to launch roody, then it immediately crashes the analyzer with 
the following trace:

#6 root_server_thread (arg=ox7f54fc001150) at src/mana.c:5154
#7 0x0000003219a1e13a in TThread::Function(void*) () from /usr/lib64/root/libThread.so.5.28
#8 0x0000003dd1207851 in start_thread () from /lib64/libpthread.so.0
#9 0x0000003dd0ee76dd in clone () from /lib64/libc.so.6

The line src/mana.c:5154 points to the following:

TObject *obj;
            if (strncmp(request + 10, "Any", 3) == 0)
               obj = folder->FindObjectAny(request + 14);
            else
               obj = folder->FindObject(request + 11);    // LINE 5154


Any suggestions on what may be going on here?  Thanks.


Cheng-Ju

> I've installed midas (Rev:5294) on SLC6.3 (64bit), along with recent trunk versions of rootana and roody. 
>
> #6 root_server_thread (arg=ox7f54fc001150) at src/mana.c:5154

You are connecting to mana, the old midas analyzer. The code for connecting to it is still present in roody,
but I cannot support the matching server code in mana.c - it is 2 revolutions behind the current state of
the ROOT object server (look in ROOTANA - the NetDirectory stuff and the latest is the XmlServer stuff).

I can offer 2 solutions - switch from mana.c to a ROOTANA based analyzer or graft the XmlServer code
into your analyzer (it is very simple - you need to create an XmlServer object and tell it which ROOT
containers you want to make visible to ROODY).

I guess you can also debug the old midas server code inside mana.c...

K.O.

Hi Konstantin,

Many thanks for your feedback.  I was able to keep the analyzer from exiting when launching roody by making some changes in the roody code. 
This at least allows me to keep moving forward. I will look into your suggestion of converting to ROOTANA based analyzer as well.

Regards,

Cheng-Ju


> > I've installed midas (Rev:5294) on SLC6.3 (64bit), along with recent trunk versions of rootana and roody. 
> >
> > #6 root_server_thread (arg=ox7f54fc001150) at src/mana.c:5154
> 
> You are connecting to mana, the old midas analyzer. The code for connecting to it is still present in roody,
> but I cannot support the matching server code in mana.c - it is 2 revolutions behind the current state of
> the ROOT object server (look in ROOTANA - the NetDirectory stuff and the latest is the XmlServer stuff).
> 
> I can offer 2 solutions - switch from mana.c to a ROOTANA based analyzer or graft the XmlServer code
> into your analyzer (it is very simple - you need to create an XmlServer object and tell it which ROOT
> containers you want to make visible to ROODY).
> 
> I guess you can also debug the old midas server code inside mana.c...
> 
> K.O.

Hi, when attempting to access a history window the following message is repeated
over and over in the MIDAS message log:

Thu Sep 6 11:37:16 2012 [mhttpd,ERROR] [history.c:886:hs_count_events,ERROR]
cannot find recent history file
Thu Sep 6 11:38:16 2012 [mhttpd,ERROR] [history.c:886:hs_count_events,ERROR]
cannot find recent history file
Thu Sep 6 11:38:16 2012 [mhttpd,ERROR] [history.c:886:hs_count_events,ERROR]
cannot find recent history file
Thu Sep 6 11:39:16 2012 [mhttpd,ERROR] [history.c:886:hs_count_events,ERROR]
cannot find recent history file
Thu Sep 6 11:39:16 2012 [mhttpd,ERROR] [history.c:886:hs_count_events,ERROR]
cannot find recent history file

It appears to be related to attempting to display a history graph that includes
some time periods that have no recorded history data. When I zoom in so that the
whole graph has data the error message goes away.

The graph displays fine either way, so this error message seems useless. Is
there a way to suppress it?

Thanks
Shaun

> > 
> > I guess you can also debug the old midas server code inside mana.c...
> > 

I ended up doing this. (After receiving some discussion by email).

Remembered that this is an old problem with the old midasServer network
protocol in mana.c - if mana.c is compiled 32-bit, it sends 32-bit pointers, if compiled 64-bit
it sends 64-bit pointers. On the receiving end (in roody), the ROOT TMessage object does not
provide any easy way to tell between them (i.e. object length is reported as 12 or 16 for the two cases).

To make things more interesting, the midasServer code in ROOTANA always sends 32-bit "pointers",
(which are not pointers but 32-bit integer cookies).

I use the ROOTANA midasServer to test ROODY (I have no working mana.c analyzers available),
and ROODY expects to receive 32-bit "pointers", so the two are consistent.

But if I compile my midasServer to send/receive 64-bit "pointers" (cookies), I reproduce this crash. What I can reproduce I can "fix".

If I change the code in ROODY to receive and return 64-bit "pointers" (cookies), both 32-bit and 64-bit midasServer seems to work okey.

This is committed as roody svn rev 248. (https://ladd00.triumf.ca/svn/roody/trunk)

It is the same fix as suggested by Cheng-Ju Stephen Lin [cjslin@lbl.gov].

I hope this helps (or breaks the ROODY midasServer connection for everybody. I hope not).

K.O.

Hi, I'm having some trouble getting ss_thread_kill() to work properly. It seems 
to kill the entire program instead 
of just the thread. Here is a test program to show the error:

_________________________________
#include <stdio.h>
#include <stdlib.h>
#include "midas.h"
#include "msystem.h"

INT f(void *param)
{
  for (int x = 0; x < 100; x++)
    sleep(1);
  return 0;
}

int main()
{
  printf("creating thread\n");
  midas_thread_t thr = ss_thread_create(f, NULL);
  sleep(2);
  printf("killing thread\n");
  ss_thread_kill(thr);
  printf("success\n");
  return 0;
}
_________________________________

Makefile:
_________________________________
FLAGS=-g -Wall -DLINUX -DOS_LINUX -I/home/deap/packages/midas/include 
LIBS=-L/home/deap/packages/midas/linux-m64/lib -lmidas -lpthread -lrt -lutil

main.exe: main.cpp 
	g++ $(FLAGS) -o $@ $^ $(LIBS)

_________________________________

Output when run:

_________________________________

[deap@deap04 multithread]$ ./main.exe 
creating thread
killing thread
Killed
[deap@deap04 multithread]$ 
_________________________________

The last "Killed" indicated the whole program got killed, when it should 
actually just kill the thread and then 
print "success".

I noticed the function in system.c uses pthread_kill(). Some google searches 
show me that it may be better to use 
pthread_cancel() (ie http://stackoverflow.com/questions/3438536/when-to-use-
pthread-cancel-and-not-pthread-kill ).


Shaun

The Linux thread functionality was introduced by Konstantin, so he might have a better idea about that.

What I usually do is a graceful thread shutdown just by a flag. Like

int stop_thread = 0;

INT f(void *param)
{
  for (int x = 0; x < 100; x++) {
    sleep(1);
    if (stop_thread) {
      // clean up things here...
      return 0;
    }
  }
  return 0;
}

int main()
{
 printf("creating thread\n");
 midas_thread_t thr = ss_thread_create(f, NULL);
 sleep(2);
 printf("killing thread\n");
 stop_thread = 1;
 sleep(2);
 printf("success\n");
 return 0;
}


This way I have a chance to clean up things in the thread, which otherwise I would not be able to.

> Hi, I'm having some trouble getting ss_thread_kill() to work properly. It seems 
> to kill the entire program instead of just the thread.

You cannot kill a thread. It's not a well defined operation. Most OSes do have the 
technical possibility to kill threads, but if you use them, you will not like the 
results. For a taste of small trouble, if a thread is holding a lock and you kill 
it, who's job is it to release the lock?

The best you can do is to ask the thread to gracefully shutdown itself. (I.e. by 
using global variable flags).

P.S. I did not implement the ss_thread stuff, I do not know what ss_thread_kill() 
does, but I recommend that you do not use it.

P.P.S. Programming using threads is complicated, I recommend that you read at least 
some literature on the topic before using threads. At the least you must understand 
the common pitfalls and mistakes. At the least, you must know about deadlocks, 
livelocks, race conditions and semaphore priority inversions.

K.O.

> > Hi, I'm having some trouble getting ss_thread_kill() to work properly. It seems 
> > to kill the entire program instead of just the thread.
> 
> You cannot kill a thread. It's not a well defined operation. Most OSes do have the 
> technical possibility to kill threads, but if you use them, you will not like the 
> results. For a taste of small trouble, if a thread is holding a lock and you kill 
> it, who's job is it to release the lock?
> 
> The best you can do is to ask the thread to gracefully shutdown itself. (I.e. by 
> using global variable flags).
> 
> P.S. I did not implement the ss_thread stuff, I do not know what ss_thread_kill() 
> does, but I recommend that you do not use it.
> 
> P.P.S. Programming using threads is complicated, I recommend that you read at least 
> some literature on the topic before using threads. At the least you must understand 
> the common pitfalls and mistakes. At the least, you must know about deadlocks, 
> livelocks, race conditions and semaphore priority inversions.
> 
> K.O.

Yes, but unfortunately what I was attempting to do was use a library function that I
can't alter. It sometimes gets stuck and I wanted a way to kill it. Anyway I ended up
not doing this at all in c++; I was able to do what I needed in python.

Shaun

When building MIDAS on Fedora 17 64-bit, the default zlib 1.2.5 shared library
is linked to.  When recording data, the "/Logger/Channels/*/Statistics/Bytes
written" value does not get set correctly beyond the first few seconds of the
run.  Occasionally, it appears to not get set at all, and mlogger aborts the run.

Installing zlib 1.2.3 in static form to /usr/local/lib (the default location),
and changing the NEED_ZLIB section of the MIDAS Makefile to the following seems
to function as a workaround:

ifdef NEED_ZLIB
CFLAGS   += -DHAVE-ZLIB
LIBS     += /usr/local/lib/libz.a
endif

Several Fedora 17 libraries expect zlib 1.2.5 specifically, so it seems safest
to not replace the default zlib shared library.

Some extra details are that the VME CPU is an XVB602, and the most recent GE-IP
drivers are being used for VME communication.  Fedora 17 was chosen to avoid a
bug with the VGA output in Fedora 13-16.

If is not so easy to get out of zlib how many bytes have been written actually. I used an undocumented function, 
which breaks down on 64-bit systems.

I now rewrote the code in mlogger.cxx to use lseek() to "measure" actually the output file and set the values 
correctly. I tried on a few systems but am not 100% sure if it works everywhere. Can you please double check?

The fix is in SVN revision 5347.

/Stefan

Dear Midas Experts

I am Wenliang Li, a graduate student from University of Regina. Our group have
encountered some difficulty on outputting ADC and TDC data into ROOT tree with
the MIDAS SVN Revision: 5347.

Our Linux Distribution: Scientific Linux release 6.0 (Carbon)
ROOT Version:           ROOT 5.28
gcc version:            g++ (GCC) 4.4.4 20100726 (Red Hat 4.4.4-13)
kernel version:         2.6.32-279.19.1.el6.i686


I am using the given example $MIDASSYS/examples/experiment to generate some
data, and the issue is that the analyzer refuses to turn on the  ADC0 and TDC0
back switches. 

If the ADC and TDC banks are switched off, the analyzer will successfully output
the histograms but not the ROOT tree, and the Trigger and Scaler root trees are
completely empty.

With the same example experiment: $MIDASSYS/examples/experiment, this issue does
not occur on MIDAS SVN Revision: 4309.


The output error messages in the analyzer window are shown if the ADC and TDC
banks are switched to 1:

*************************
Connect to experiment ...OK
Root server listening on port 9090...
Loading previous online histos from /home/billlee/experiment/test_exp/last.root
Running analyzer online. Stop with "!"
Error in <TTree::Branch>: The pointer specified for ADC0 is not of a class known
to ROOT and (null) is not a known class
ROOT TTree rebooked
Error in <TTree::Branch>: The pointer specified for ADC0 is not of a class known
to ROOT and (null) is not a known class
Error in <TTree::Branch>: The pointer specified for TDC0 is not of a class known
to ROOT and (null) is not a known class
ROOT TTree rebooked
***********************
***************************



If I analyze the data with TDC and ADC bank switched set to be 1 :
$ analyzer -i runXXXXX.mid -o runXXXXX.root

I get the following error messages:


************************************************************************
************************************************************************


Root server listening on port 9090...
Running analyzer offline. Stop with "!"
Error in <TTree::Branch>: The pointer specified for ADC0 is not of a class known
to ROOT and (null) is not a known class
Error in <TTree::Branch>: The pointer specified for TDC0 is not of a class known
to ROOT and (null) is not a known class
Set run number 1 in ODB
Load ODB from run 1...OK

 *** Break *** segmentation violation



===========================================================
There was a crash.
This is the entire stack trace of all threads:
===========================================================

Thread 2 (Thread 0x7f46c6853700 (LWP 10808)):
#0  0x0000003b63a0e84d in accept () from /lib64/libpthread.so.0
#1  0x0000003b64e370f4 in TUnixSystem::AcceptConnection(int) () from
/usr/lib64/root/libCore.so.5.28
#2  0x0000003b6647849c in TServerSocket::Accept(unsigned char) () from
/usr/lib64/root/libNet.so.5.28
#3  0x000000000040c50e in root_socket_server (arg=<value optimized out>) at
src/mana.c:5275
#4  0x00007f46c8dc513a in TThread::Function(void*) () from
/usr/lib64/root/libThread.so.5.28
#5  0x0000003b63a07851 in start_thread () from /lib64/libpthread.so.0
#6  0x0000003b62ee811d in clone () from /lib64/libc.so.6

Thread 1 (Thread 0x7f46c8b94720 (LWP 10800)):
#0  0x0000003b62eabfdd in waitpid () from /lib64/libc.so.6
#1  0x0000003b62e3e899 in do_system () from /lib64/libc.so.6
#2  0x0000003b62e3ebd0 in system () from /lib64/libc.so.6
#3  0x0000003b64e3da31 in TUnixSystem::StackTrace() () from
/usr/lib64/root/libCore.so.5.28
#4  0x0000003b64e3d3f3 in TUnixSystem::DispatchSignals(ESignals) () from
/usr/lib64/root/libCore.so.5.28
#5  <signal handler called>
#6  0x000000000041245f in TIter (file=<value optimized out>,
pevent=0x7f46c5281010, par=0x665180) at /usr/include/root/TCollection.h:148
#7  write_event_ttree (file=<value optimized out>, pevent=0x7f46c5281010,
par=0x665180) at src/mana.c:2872
#8  0x0000000000412a4c in process_event (par=0x665180, pevent=0x7f46c5281010) at
src/mana.c:3195
#9  0x0000000000412e42 in analyze_run (run_number=1,
input_file_name=0x7fff4d738340 "run00001.mid", output_file_name=<value optimized
out>) at src/mana.c:4178
#10 0x0000000000413372 in loop_runs_offline () at src/mana.c:4366
#11 0x0000000000413ba5 in main (argc=<value optimized out>, argv=<value
optimized out>) at src/mana.c:5579
===========================================================


The lines below might hint at the cause of the crash.
If they do not help you then please submit a bug report at
http://root.cern.ch/bugs. Please post the ENTIRE stack trace
from above as an attachment in addition to anything else
that might help us fixing this issue.
===========================================================
#6  0x000000000041245f in TIter (file=<value optimized out>,
pevent=0x7f46c5281010, par=0x665180) at /usr/include/root/TCollection.h:148
#7  write_event_ttree (file=<value optimized out>, pevent=0x7f46c5281010,
par=0x665180) at src/mana.c:2872
#8  0x0000000000412a4c in process_event (par=0x665180, pevent=0x7f46c5281010) at
src/mana.c:3195
#9  0x0000000000412e42 in analyze_run (run_number=1,
input_file_name=0x7fff4d738340 "run00001.mid", output_file_name=<value optimized
out>) at src/mana.c:4178
#10 0x0000000000413372 in loop_runs_offline () at src/mana.c:4366
#11 0x0000000000413ba5 in main (argc=<value optimized out>, argv=<value
optimized out>) at src/mana.c:5579
===========================================================


[midas.c:1973:,ERROR] cm_disconnect_experiment not called at end of program

**********************************************************************************************
**********************************************************************************************







I wonder if there is any program syntax change between MIDAS Version 4309 and
5347, and are there any simple working setup example which can output root tree
with the newest version of MIDAS?
 
In the end, I would like to thank the continuous effort from Triumf and PSI on
developing MIDAS, it is a pleasure to work with.

Many thanks
Bill 

Dear Bill,

the Midas analyzer "mana.c" is currently not maintained. At PSI we use the ROME framework (which might be too complicated for a 
small experiment) and at TRIUMF the ROOTANA framework is used:

http://ladd00.triumf.ca/~olchansk/rootana/

You might be better off switching to that one.

Best regards,
Stefan

The odb.c function merge_records() has a fixed size array of 10000 bytes to handle the data and it 
aborts with an assert() if passed data bigger than that. It is called from db_create_record() which 
already allocates a data buffer of correct size for it's operations.
K.O.

odbedit uses a fixed size buffer for ODB data. If an array in ODB is bugger than this size, 
db_get_data() correctly returns DB_TRUNCATED and there is no memory overwrite, but the following 
code for printing the data does not know about this truncation and proceeds printing memory 
values contained after the end of the fixed size data buffer - in the current case, this memory 
somehow had the contents of the shell history - this very confused the MIDAS users as they though 
that the funny printout was actually in ODB. This is in the print_key() function. If db_get_data() 
returns DB_TRUNCATED, it should allocate a buffer of bigger size. This (and the previous) bug found 
by the TIGRESS experiment. K.O.

I find that new mhttpd has strange behaviour for ODB strings.

- I create a new STRING variable in ODB through mhttpd.  It defaults to size 32.

- I then edit the STRING variable through mhttpd, writing a new string larger
than 32 characters.

- Initially everything looks fine; it seems as if the new string value has been
accepted.

- But if you reload the page, or navigate back to the page, you realize that
mhttpd has silently truncated the string back to 32 characters.

You can reproduce this problem on a test page here:

http://midptf01.triumf.ca:8081/AnnMessage

Older versions of mhttpd (I'm testing one from 2 years ago) don't have this
'feature'.  For older mhttpd the string variable would get resized when a larger
string was inputted.  That definitely seems like the right behavior to me.

I am using fresh copy of midas from bitbucket as of this morning. (How do I get
a particular tag/hash of the version of midas that I am using?)

I can confirm part of the problem - the new inline-edit function - after you finish editing - shows you what you 
have typed, not what's actually in ODB - at the very end it should do an ODBGet() to load the actual ODB 
contents and show *that* to the user.

The truncation to 32 characters - most likely it is a failure to resize the ODB string - is probably in mhttpd and 
I can take a quick look into it.

There is a 3rd problem - the mhttpd ODB editor "create" function does not ask for the string length to create.

Actually, in ODB, "create" and "set string size" are two separate functions - db_create_key(TID_STRING) creates 
a string of length zero, then db_set_data() creates an empty string of desired length.

In the new AJAX interface these two functions are separate (ODBCreate just calls db_create_key()).

In the present ODBSet() function the two are mixed together - and the ODB inline edit function uses ODBSet().

K.O.



> I find that new mhttpd has strange behaviour for ODB strings.
> 
> - I create a new STRING variable in ODB through mhttpd.  It defaults to size 32.
> 
> - I then edit the STRING variable through mhttpd, writing a new string larger
> than 32 characters.
> 
> - Initially everything looks fine; it seems as if the new string value has been
> accepted.
> 
> - But if you reload the page, or navigate back to the page, you realize that
> mhttpd has silently truncated the string back to 32 characters.
> 
> You can reproduce this problem on a test page here:
> 
> http://midptf01.triumf.ca:8081/AnnMessage
> 
> Older versions of mhttpd (I'm testing one from 2 years ago) don't have this
> 'feature'.  For older mhttpd the string variable would get resized when a larger
> string was inputted.  That definitely seems like the right behavior to me.
> 
> I am using fresh copy of midas from bitbucket as of this morning. (How do I get
> a particular tag/hash of the version of midas that I am using?)

I confirm the second part of the problem.

Inline edit uses ODBSet(), which uses the "jset" AJAX call to mhttpd which does not extend string variables.

This is the jset code. The best I can tell it truncates string variables to the existing size in ODB:

db_find_key(hDB, 0, str, &hkey)
db_get_key(hDB, hkey, &key);
memset(data, 0, sizeof(data));
size = sizeof(data);
db_sscanf(getparam("value"), data, &size, 0, key.type);
db_set_data_index(hDB, hkey, data, key.item_size, index, key.type);

These original jset/jget functions are a little bit too complicated and there is no documentation (what exists is done by me trying to read the existing code).

We now have a jcopy/ODBMCopy() as a sane replacement for jget, but nothing comparable for jset, yet.

I think this quirk of inline edit cannot be fixed in javascript - the mhttpd code for "jset" has to change.

K.O.


> 
> I can confirm part of the problem - the new inline-edit function - after you finish editing - shows you what you 
> have typed, not what's actually in ODB - at the very end it should do an ODBGet() to load the actual ODB 
> contents and show *that* to the user.
> 
> The truncation to 32 characters - most likely it is a failure to resize the ODB string - is probably in mhttpd and 
> I can take a quick look into it.
> 
> There is a 3rd problem - the mhttpd ODB editor "create" function does not ask for the string length to create.
> 
> Actually, in ODB, "create" and "set string size" are two separate functions - db_create_key(TID_STRING) creates 
> a string of length zero, then db_set_data() creates an empty string of desired length.
> 
> In the new AJAX interface these two functions are separate (ODBCreate just calls db_create_key()).
> 
> In the present ODBSet() function the two are mixed together - and the ODB inline edit function uses ODBSet().
> 
> K.O.
> 
> 
> 
> > I find that new mhttpd has strange behaviour for ODB strings.
> > 
> > - I create a new STRING variable in ODB through mhttpd.  It defaults to size 32.
> > 
> > - I then edit the STRING variable through mhttpd, writing a new string larger
> > than 32 characters.
> > 
> > - Initially everything looks fine; it seems as if the new string value has been
> > accepted.
> > 
> > - But if you reload the page, or navigate back to the page, you realize that
> > mhttpd has silently truncated the string back to 32 characters.
> > 
> > You can reproduce this problem on a test page here:
> > 
> > http://midptf01.triumf.ca:8081/AnnMessage
> > 
> > Older versions of mhttpd (I'm testing one from 2 years ago) don't have this
> > 'feature'.  For older mhttpd the string variable would get resized when a larger
> > string was inputted.  That definitely seems like the right behavior to me.
> > 
> > I am using fresh copy of midas from bitbucket as of this morning. (How do I get
> > a particular tag/hash of the version of midas that I am using?)

Actually this was no bug, but a missing feature. Strings were never meant to be extended via the web interface. 
Now I added that feature to the current version. Please check it.

/Stefan

> This is the jset code. The best I can tell it truncates string variables to the existing size in ODB:
> 
> db_find_key(hDB, 0, str, &hkey)
> db_get_key(hDB, hkey, &key);
> memset(data, 0, sizeof(data));
> size = sizeof(data);
> db_sscanf(getparam("value"), data, &size, 0, key.type);
> db_set_data_index(hDB, hkey, data, key.item_size, index, key.type);


Correct. So I added some code which extends strings if necessary (NOT string arrays, they are more complicated to handle).