Alarm System: Difference between revisions

From MidasWiki
Jump to navigation Jump to search
No edit summary
No edit summary
Line 26: Line 26:
* Selection of alarm message destination (to system message log or to elog)
* Selection of alarm message destination (to system message log or to elog)
* email or SMS alerts can be sent
* email or SMS alerts can be sent
* Program control on run transition
* Alarm triggered when a Program is not running




Line 39: Line 39:
As of alarm.c rev 4683, "System message interval" is set to 0 ensures that every alarm is recorded into the MIDAS log file. (In previous revisions, this setting may still miss some alarms).
As of alarm.c rev 4683, "System message interval" is set to 0 ensures that every alarm is recorded into the MIDAS log file. (In previous revisions, this setting may still miss some alarms).


= Types of alarm =
There are four [[/Programs ODB tree#Alarm Type|types of alarm]]:


==  ''program not running'' alarms ==


These alarms are controlled through the [[/Programs ODB tree]] rather than the [[/Alarms ODB tree]] except of course for the [[/Alarms ODB tree#Alarms class|alarm class]].  All the ODB
keys mentioned in this section are in the <span style="color: purple; font-style:italic;">/Programs/<client-name></span> subtree unless otherwise specified.


These alarms are enabled by setting the ODB key [[/Programs ODB tree#Alarm class|Alarm class]] to a valid Alarm class specified in the [[/Alarms ODB tree]]. Each time al_check() runs, every client listed in the <span style="color: purple; font-style:italic;">/Programs</span> ODB tree is tested using "cm_exist()" and if the client is not running, the time of first failure is recorded in the ODB key [[/Programs ODB tree#First failed|First failed]].
<div id="Alarm class"></div>
= Alarms structure =
The [[/Alarms ODB tree]] structure is split into 2 sections:
*"Alarms" which define the condition to be tested. The user can create as many [[/Alarms ODB tree#Alarms subtree|Alarms]] as desired, but each must be one of the four defined [[#Alarm Types|Alarm types]] .
*"Classes" which define the action to be taken when the alarm occurs. Two Classes (Alarm and Warning) are defined by default. The user can add more [[/Alarms ODB tree#Classes subtree|Classes]] as desired.
<br><br>


If the client has not been running for longer than the time set in ODB key [[/Programs ODB tree#Check interval|Check interval]], an alarm is triggered (if enabled by [[/Programs ODB tree#Alarm class|Alarm class]] and the program is restarted (if enabled by [[/Programs ODB tree#Auto restart|Auto restart]]).
In order to make the system flexible, each alarm class may perform different actions when an alarm is given. For example, it may
* write a system message (see [[/Alarms ODB tree#Write System Message|Write System Message]])
* write to the elog (see [[/Alarms ODB tree#Write Elog Message|Write Elog Message]])
* stop the run (see [[/Alarms ODB tree#Stop run|Stop run]])
* spawn a detached script listed in the ODB variable [[/Alarms ODB tree#Execute command|Execute command]]. This feature is used when an Alarm triggers Email or SMS alerts (see [[#Alarm triggering Email or SMS alerts|example]]).
<br>


The "not running" condition is tested every 10 seconds (each time al_check() is called), but the frequency of "program not running" alarms can be reduced by increasing the value of  
= Alarm Types =
[[/Programs ODB tree#Check interval|Check interval]]
 
The four available Alarm Types are shown in Table 1. They are defined in [http://ladd00.triumf.ca/~daqweb/doc/midas-old/html/midas_8h.html&nbsp midas.h]. The alarm type is entered into the [[/Alarms ODB tree#Type|Type]] key.
 
 
{|  style="text-align: left; width: 100%; background-color: white;" border="3" cellpadding="2" cellspacing="2"
|+ Table 1 : Defined Alarm Types
|-
| colspan="2" rowspan="1" style="vertical-align: top; background-color: lavender; font-weight: bold;" | Alarm Type
| colspan="1" rowspan="1" style="vertical-align: top; background-color: lavender; font-weight: bold;" | INT value
| colspan="1" rowspan="1" style="vertical-align: top; background-color: lavender; font-weight: bold;" | Explanation
 
|- 
| colspan="1" rowspan="1"  style="vertical-align: top; background-color: white; font-weight: bold;"  |Internal alarms
| colspan="1" rowspan="1"  style="vertical-align: top; background-color: white; font-weight: normal;" | AT_INTERNAL
|1
|Trigger on internal (program) alarm setting through the use of the al_...() functions.
 
|- 
| style="vertical-align: top; background-color: white; font-weight: bold;" |Program alarms
|AT_PROGRAM
|2
|Triggered on condition of the state of the defined task (i.e. program not running)
 
|- 
| style="vertical-align: top; background-color: white; font-weight: bold;" |Evaluated alarms
|AT_EVALUATED
|3
|Triggered by ODB value on given arithmetical condition.
 
|- 
| style="vertical-align: top; background-color: white; font-weight: bold;" |Periodic alarms
|AT_PERIODIC
|4
|Triggered by timeout condition defined in the alarm setting.
 
|}
 
 
==  Program Alarm ==
 
''program'' (or rather ''program not running'') alarms, when enabled, warn the user when a program is not running.
 
''Program'' alarms are enabled by setting the ODB key [[/Programs ODB tree#Alarm class|/Programs/<client-name>/Alarm class]] to a valid Alarm class specified in the [[/Alarms ODB tree]]. The first time the alarm is triggered, an <span style="color: purple; font-style:italic;">/Alarms/Alarms/<client-name></span> subtree will be created automatically. The ''program'' alarm will not be visible in the [[Alarms Page]] until the alarm has triggered, and the subtree created.
 
Each time al_check() runs, every client listed in the <span style="color: purple; font-style:italic;">/Programs</span> ODB tree is tested using "cm_exist()" and if the client is not running, the time of first failure is recorded in the ODB key [[/Programs ODB tree#First failed|/Programs/<client-name>/First failed]].
 
If the client has not been running for longer than the time set in ODB key [[/Programs ODB tree#Check interval|/Programs/<client-name>/Check interval]], an alarm is triggered (if enabled by [[/Programs ODB tree#Alarm class|Alarm class]]) and the program is restarted (if enabled by [[/Programs ODB tree#Auto restart|/Programs/<client-name>/Auto restart]] and a valid [[/Programs ODB tree#Start command|Start command]] is supplied).
 
The "not running" condition is tested every 10 seconds (each time al_check() is called), but the frequency of ''program not running'' alarms can be reduced by increasing the value of the ODB key
[[/Programs ODB tree#Check interval|/Programs/<client-name>/Check interval]]
(default value 60 seconds). This can be useful if  [[/Alarms ODB tree#System message interval|System message interval]] in the specified alarm class subtree is set to zero.
(default value 60 seconds). This can be useful if  [[/Alarms ODB tree#System message interval|System message interval]] in the specified alarm class subtree is set to zero.


==  ''evaluated alarms'' ==
are programmed through the [[/Alarms ODB tree]]. See [[/Alarms ODB tree#Evaluated Alarm conditions|evaluated alarms]].


==  ''periodic alarms'' ==
== Periodic Alarm ==
are programmed through the [[/Alarms ODB tree]]. See[[/Alarms ODB tree#Periodic Alarm|periodic alarms]].
The periodic alarm is activated periodically according to the time in [[#Check interval|Check interval]]. An example of a periodic alarm is "Demo Periodic" in the  [[/Alarms ODB tree#Example|example]].
 
== Evaluated Alarm  ==
 
Evaluated alarms require an ''alarm condition'' which is entered into the ODB key [[/Alarms ODB tree#Condition|Condition]] in the  [[/Alarms ODB tree#<alarm_name> subtree|<alarm_name> subtree]].
The condition may be simply a '''comparison''' between any ODB variable and a threshold parameter, e.g.
 
  /Runinfo/Run number > 100
or it may be an '''evaluated condition'''. One can write conditions like
 
  /Equipment/HV/Variables/Input[*] < 100
or
 
  /Equipment/HV/Variables/Input[2-3] < 100
to check all values from an array or a certain range. If one array element fulfills the alarm condition, the alarm is triggerrd. In addition, bit-wise alarm conditions are possible, e.g.
 
  /Equipment/Environment/Variables/Input[0] & 8
The alarm is triggered if bit #2 is set in Input[0].
 
The value of an evaluated alarm is computed using al_evaluate_condition() in [http://ladd00.triumf.ca/~daqweb/doc/midas-old/html/alarm_8c_source.html alarm.c].
 
== Internal Alarm ==
These are triggered in a program using a call to
[http://ladd00.triumf.ca/~daqweb/doc/midas/doc/html/group__alfunctioncode.html#gac024cd8160dc8b9418f05a63678f6c68&nbsp al_trigger_alarm()]. See also description of al_trigger_alarm() sequence  [[#Implementation of the MIDAS Alarm System|above]].
 
 
There is nothing surprising in these alarms. Each alarm is checked with a time period set by ODB key [[ /Alarms ODB tree#Check interval|Check interval]] in the [[/Alarms ODB tree]].
 
 
 
 
 
= Alarm triggering Email or SMS alerts =
 
It is possible to have the MIDAS alarm system send email or SMS alerts to cell phones when alarms are triggered. This can be configured by defining an ODB alarm on a critical ODB parameter, e.g.
 
/Alarms/Alarms/Liquid Level
Active                  y
Triggered                0 (0x0)
Type                    3 (0x3)
Check interval          60 (0x3C)
Checked last    1227690148 (0x492D10A4)
Time triggered first    (empty)
Time triggered last    (empty)
Condition              /Equipment/Environment/Variables/Input[0] < 10
Alarm Class            Level Alarm
Alarm Message          Liquid Level is only %s
 
In this example, the alarm triggers an alarm of class "Level Alarm". This alarm class is defined as follows:
 
/Alarms/Classes/Level Alarm
Write system message    y
Write Elog message      n
System message interval 600 (0x258)
System message last    0 (0x0)
Execute command        /home/midas/level_alarm '%s'
Execute interval        1800 (0x708)
Execute last            0 (0x0)
Stop run                n
Display BGColor        red
Display FGColor        black
 
The key here is to call a script "level_alarm", which can send emails. Use something like:
 
#/bin/csh
echo $1 | mail -s \"Level Alarm\" your.name@domain.edu
odbedit -c 'msg 2 level_alarm \"Alarm was sent to your.name@domain.edu\"'
 
The second command just generates a MIDAS system message for confirmation. Most cell phones (depends on the provider) have an email address. If you send an email there, it will be translated into a SMS message.
 
The script file above can of course be more complicated. A perl script could be used that parses an address list, so other interested parties can register by adding his/her email address to that list. The script may also collects some other slow control variables (like pressure, temperature) and combine them into the SMS message.
 
For very sensitive systems, having an alarm via SMS may not be sufficient, since the alarm system could be down (e.g. computer crash, network failure). In this case 'negative alarms' can be used. For example, every 30 minutes the system may send an SMS with the current parameter values. If the expected message is not received, it may indicate that something in the MIDAS system is wrong.
 


==  ''internal'' alarms ==
are triggered in a program using a call to
[http://ladd00.triumf.ca/~daqweb/doc/midas/doc/html/group__alfunctioncode.html#gac024cd8160dc8b9418f05a63678f6c68&nbsp al_trigger_alarm()]. The alarm is configured through the [[/Alarms ODB tree]].


There is nothing surprising in these alarms. Except "program not running" alarms (see above), each alarm is checked with a time period set by ODB key [[ /Alarms ODB tree#Check interval|Check interval]] in the [[/Alarms ODB tree]]. The value of an evaluated alarm is computed using al_evaluate_condition().


= Setting up the Alarm system =
See the [[/Alarms ODB tree]] for details and
[[/Alarms ODB tree#/Alarms tree structure|Alarms tree structure]].  See also [[/Alarms ODB tree#Examples|Examples]].


[[Category:Alarms]]
[[Category:Alarms]]

Revision as of 17:32, 13 August 2014


Links

Introduction

MIDAS provides an alarm system, which by default is turned off. When the alarm system is activated and an alarm condition is detected, alarm messages are sent by the system which appear as an alarm banner on the mhttpd status page, and as a message on any windows running odbedit clients. The alarm system is flexible and can be extensively customized for each experiment using the mhttpd Alarms Page or odbedit.

The alarm system is built-in and part of the main experiment scheduler. This means no separate task is necessary to benefit from the alarm system. Its setup and activation is done through the /Alarms ODB tree. The alarm system includes several other features such as sequencing and control of the experiment. The alarm capabilities are:

  • Alarm setting on any ODB variable against a threshold parameter.
  • Alarm triggered by evaluated condition
  • Selection of Alarm check frequency
  • Selection of Alarm trigger frequency
  • Customization alarm scheme; under this scheme multiple choices of alarm type can be selected
  • Selection of alarm message destination (to system message log or to elog)
  • email or SMS alerts can be sent
  • Alarm triggered when a Program is not running


Implementation of the MIDAS Alarm System

The alarm system source code is alarm.c. Alarms are checked inside alarm.c::al_check(). This function is called by cm_yield() every 10 seconds and by rpc_server_thread(), also every 10 seconds. For remote MIDAS clients, their al_check() issues an RPC_AL_CHECK RPC call into the MIDAS server utility mserver, where rpc_server_dispatch() calls the local al_check(). As result, all alarm checks run inside a process directly attached to the local MIDAS shared memory (inside a local client or inside an mserver process for a remote client). Each and every MIDAS client runs the alarm checks. To prevent race conditions between different MIDAS clients, access to al_check() is serialized using the ALARM semaphore. Inside al_check(), alarms are triggered using al_trigger_alarm(), which in turn calls al_trigger_class(). Inside al_trigger_class(), the alarm is recorded into an elog or into midas.log using cm_msg(MTALK).

Special note should be made of the ODB setting system message interval, which has a surprising effect - after an alarm is recorded into system messages (using cm_msg(MTALK)), no record is made of any subsequent alarms until the time interval set by this variable elapses. With default value of 60 seconds, after one alarm, no more alarms are recorded for 60 seconds. Also, because all the alarms are checked at the same time, only the first triggered alarm will be recorded.

As of alarm.c rev 4683, "System message interval" is set to 0 ensures that every alarm is recorded into the MIDAS log file. (In previous revisions, this setting may still miss some alarms).



Alarms structure

The /Alarms ODB tree structure is split into 2 sections:

  • "Alarms" which define the condition to be tested. The user can create as many Alarms as desired, but each must be one of the four defined Alarm types .
  • "Classes" which define the action to be taken when the alarm occurs. Two Classes (Alarm and Warning) are defined by default. The user can add more Classes as desired.



In order to make the system flexible, each alarm class may perform different actions when an alarm is given. For example, it may


Alarm Types

The four available Alarm Types are shown in Table 1. They are defined in midas.h. The alarm type is entered into the Type key.


Table 1 : Defined Alarm Types
Alarm Type INT value Explanation
Internal alarms AT_INTERNAL 1 Trigger on internal (program) alarm setting through the use of the al_...() functions.
Program alarms AT_PROGRAM 2 Triggered on condition of the state of the defined task (i.e. program not running)
Evaluated alarms AT_EVALUATED 3 Triggered by ODB value on given arithmetical condition.
Periodic alarms AT_PERIODIC 4 Triggered by timeout condition defined in the alarm setting.


Program Alarm

program (or rather program not running) alarms, when enabled, warn the user when a program is not running.

Program alarms are enabled by setting the ODB key /Programs/<client-name>/Alarm class to a valid Alarm class specified in the /Alarms ODB tree. The first time the alarm is triggered, an /Alarms/Alarms/<client-name> subtree will be created automatically. The program alarm will not be visible in the Alarms Page until the alarm has triggered, and the subtree created.


Each time al_check() runs, every client listed in the /Programs ODB tree is tested using "cm_exist()" and if the client is not running, the time of first failure is recorded in the ODB key /Programs/<client-name>/First failed.

If the client has not been running for longer than the time set in ODB key /Programs/<client-name>/Check interval, an alarm is triggered (if enabled by Alarm class) and the program is restarted (if enabled by /Programs/<client-name>/Auto restart and a valid Start command is supplied).

The "not running" condition is tested every 10 seconds (each time al_check() is called), but the frequency of program not running alarms can be reduced by increasing the value of the ODB key /Programs/<client-name>/Check interval (default value 60 seconds). This can be useful if System message interval in the specified alarm class subtree is set to zero.


Periodic Alarm

The periodic alarm is activated periodically according to the time in Check interval. An example of a periodic alarm is "Demo Periodic" in the example.

Evaluated Alarm

Evaluated alarms require an alarm condition which is entered into the ODB key Condition in the <alarm_name> subtree. The condition may be simply a comparison between any ODB variable and a threshold parameter, e.g.

/Runinfo/Run number > 100

or it may be an evaluated condition. One can write conditions like

 /Equipment/HV/Variables/Input[*] < 100

or

 /Equipment/HV/Variables/Input[2-3] < 100

to check all values from an array or a certain range. If one array element fulfills the alarm condition, the alarm is triggerrd. In addition, bit-wise alarm conditions are possible, e.g.

 /Equipment/Environment/Variables/Input[0] & 8

The alarm is triggered if bit #2 is set in Input[0].

The value of an evaluated alarm is computed using al_evaluate_condition() in alarm.c.

Internal Alarm

These are triggered in a program using a call to al_trigger_alarm(). See also description of al_trigger_alarm() sequence above.


There is nothing surprising in these alarms. Each alarm is checked with a time period set by ODB key Check interval in the /Alarms ODB tree.



Alarm triggering Email or SMS alerts

It is possible to have the MIDAS alarm system send email or SMS alerts to cell phones when alarms are triggered. This can be configured by defining an ODB alarm on a critical ODB parameter, e.g.

/Alarms/Alarms/Liquid Level
Active                   y
Triggered                0 (0x0)
Type                     3 (0x3)
Check interval          60 (0x3C)
Checked last    1227690148 (0x492D10A4)
Time triggered first    (empty)
Time triggered last     (empty)
Condition               /Equipment/Environment/Variables/Input[0] < 10
Alarm Class             Level Alarm
Alarm Message           Liquid Level is only %s

In this example, the alarm triggers an alarm of class "Level Alarm". This alarm class is defined as follows:

/Alarms/Classes/Level Alarm
Write system message    y
Write Elog message      n
System message interval 600 (0x258)
System message last     0 (0x0)
Execute command         /home/midas/level_alarm '%s'
Execute interval        1800 (0x708)
Execute last            0 (0x0)
Stop run                n
Display BGColor         red
Display FGColor         black

The key here is to call a script "level_alarm", which can send emails. Use something like:

#/bin/csh
echo $1 | mail -s \"Level Alarm\" your.name@domain.edu
odbedit -c 'msg 2 level_alarm \"Alarm was sent to your.name@domain.edu\"'

The second command just generates a MIDAS system message for confirmation. Most cell phones (depends on the provider) have an email address. If you send an email there, it will be translated into a SMS message.

The script file above can of course be more complicated. A perl script could be used that parses an address list, so other interested parties can register by adding his/her email address to that list. The script may also collects some other slow control variables (like pressure, temperature) and combine them into the SMS message.

For very sensitive systems, having an alarm via SMS may not be sufficient, since the alarm system could be down (e.g. computer crash, network failure). In this case 'negative alarms' can be used. For example, every 30 minutes the system may send an SMS with the current parameter values. If the expected message is not received, it may indicate that something in the MIDAS system is wrong.