All functions supported in trigger expressions are listed here.
FUNCTION  

Description  Parameters  Comments  
abschange  
The amount of absolute difference between the previous and latest value.  Supported value types: float, int, str, text, log For example: (previous value;latest value=abschange) 1;5=4 3;1=2 0;2.5=2.5 For strings returns: 0  values are equal 1  values differ 

avg (sec#num,<time_shift>)  
Average value of an item within the defined evaluation period.  sec or #num  maximum evaluation period^{1} in seconds or in latest collected values (preceded by a hash mark) time_shift (optional)  evaluation point is moved the number of seconds back in time  Supported value types: float, int Examples: ⇒ avg(#5) → average value for the five latest values ⇒ avg(1h) → average value for an hour ⇒ avg(1h,1d) → average value for an hour one day ago. The time_shift parameter is supported since Zabbix 1.8.2. It is useful when there is a need to compare the current average value with the average value time_shift seconds back. 

band (<sec#num>,mask,<time_shift>)  
Value of “bitwise AND” of an item value and mask.  sec (ignored, equals #1) or #num (optional)  the Nth most recent value mask (mandatory)  64bit unsigned integer (0  18446744073709551615) time_shift (optional)  evaluation point is moved the number of seconds back in time  Supported value types: int Take note that #num works differently here than with many other functions (see last()).Although the comparison is done in a bitwise manner, all the values must be supplied and are returned in decimal. For example, checking for the 3rd bit is done by comparing to 4, not 100. Examples: ⇒ band(,12)=8 or band(,12)=4 → 3rd or 4th bit set, but not both at the same time ⇒ band(,20)=16 → 3rd bit not set and 5th bit set. This function is supported since Zabbix 2.2.0. 

change  
The amount of difference between the previous and latest value.  Supported value types: float, int, str, text, log For example: (previous value;latest value=change) 1;5=+4 3;1=2 0;2.5=2.5 See also: abschange for comparison For strings returns: 0  values are equal 1  values differ 

count (sec#num,<pattern>,<operator>,<time_shift>)  
Number of values within the defined evaluation period.  sec or #num  maximum evaluation period^{1} in seconds or in latest collected values (preceded by a hash mark) pattern (optional)  required pattern operator (optional) Supported operators :eq  equal ne  not equal gt  greater ge  greater or equal lt  less le  less or equal like  matches if contains pattern (casesensitive) band  bitwise AND regexp  case sensitive match of regular expression given in pattern iregexp  case insensitive match of regular expression given in pattern Note that: eq (default), ne, gt, ge, lt, le, band, regexp, iregexp are supported for integer items eq (default), ne, gt, ge, lt, le, regexp, iregexp are supported for float items like (default), eq, ne, regexp, iregexp are supported for string, text and log items time_shift (optional)  evaluation point is moved the number of seconds back in time  Supported value types: float, integer, string, text, log Float items match with the precision of 0.000001. With band as third parameter, the second pattern parameter can be specified as two numbers, separated by '/': number_to_compare_with/mask. count() calculates “bitwise AND” from the value and the mask and compares the result to number_to_compare_with. If the result of “bitwise AND” is equal to number_to_compare_with, the value is counted.If number_to_compare_with and mask are equal, only the mask need be specified (without '/'). With regexp or iregexp as third parameter the second pattern parameter can be an ordinary or global (starting with '@') regular expression. In case of global regular expressions case sensitivity is inherited from global regular expression settings. For the purpose of regexp matching, float values will always be represented with 4 decimal digits after '.'. Also note that for large numbers difference in decimal (stored in database) and binary (used by Zabbix server) representation may affect the 4th decimal digit. Examples: ⇒ count(10m) → number of values for last 10 minutes ⇒ count(10m,"error",eq) → number of values for last 10 minutes that equal 'error' ⇒ count(10m,12) → number of values for last 10 minutes that equal '12' ⇒ count(10m,12,gt) → number of values for last 10 minutes that are over '12' ⇒ count(#10,12,gt) → number of values within last 10 values that are over '12' ⇒ count(10m,12,gt,1d) → number of values for preceding 10 minutes up to 24 hours ago that were over '12' ⇒ count(10m,6/7,band) → number of values for last 10 minutes having '110' (in binary) in the 3 least significant bits. ⇒ count(10m,,,1d) → number of values for preceding 10 minutes up to 24 hours ago The #num parameter is supported since Zabbix 1.6.1.The time_shift parameter and string operators are supported since Zabbix 1.8.2.The band operator is supported since Zabbix 2.2.0. The regexp, iregexp operators are supported since Zabbix 3.2.0. 

date  
Current date in YYYYMMDD format.  Supported value types: any Example of returned value: 20150731 

dayofmonth  
Day of month in range of 1 to 31.  Supported value types: any This function is supported since Zabbix 1.8.5. 

dayofweek  
Day of week in range of 1 to 7 (Mon  1, Sun  7).  Supported value types: any  
delta (sec#num,<time_shift>)  
Difference between the maximum and minimum values within the defined evaluation period ('max()' minus 'min()').  sec or #num  maximum evaluation period^{1} in seconds or in latest collected values specified (preceded by a hash mark) time_shift (optional)  evaluation point is moved the number of seconds back in time  Supported value types: float, int The time_shift parameter is supported since Zabbix 1.8.2. 

diff  
Checking if last and previous values differ.  Supported value types: float, int, str, text, log Returns: 1  last and previous values differ 0  otherwise 

forecast (sec#num,<time_shift>,time,<fit>,<mode>)  
Future value, max, min, delta or avg of the item.  sec or #num  maximum evaluation period^{1} in seconds or in latest collected values specified (preceded by a hash mark) time_shift (optional)  evaluation point is moved the number of seconds back in time time  forecasting horizon in seconds fit (optional)  function used to fit historical data Supported fits :linear  linear function polynomialN  polynomial of degree N (1 <= N <= 6) exponential  exponential function logarithmic  logarithmic function power  power function Note that: linear is default, polynomial1 is equivalent to linear mode (optional)  demanded output Supported modes :value  value (default) max  maximum min  minimum delta  maxmin avg  average Note that: value estimates item value at the moment now + time max, min, delta and avg investigate item value estimate on the interval between now and now + time  Supported value types: float, int If value to return is larger than 1.7976931348623157E+308 or less than 1.7976931348623157E+308, return value is cropped to 1.7976931348623157E+308 or 1.7976931348623157E+308 correspondingly. Becomes not supported only if misused in expression (wrong item type, invalid parameters), otherwise returns 1 in case of errors. Examples: ⇒ forecast(#10,,1h) → forecast of item value after one hour based on last 10 values ⇒ forecast(1h,,30m) → forecast of item value after 30 minutes based on last hour data ⇒ forecast(1h,1d,12h) → forecast of item after 12 hours based on one hour one day ago ⇒ forecast(1h,,10m,exponential) → forecast of item value after 10 minutes based on last hour data and exponential function ⇒ forecast(1h,,2h,polynomial3,max) → forecast of maximum value item can reach in next two hours based on last hour data and cubic (third degree) polynomial ⇒ forecast(#2,,20m) → estimate the value of an item which was 20 minutes ago based on last two values (this can be more precise than using last() or prev(), especially if item is updated rarely, say, once an hour) This function is supported since Zabbix 3.0.0. Negative time values are supported since Zabbix 3.0.6 and 3.2.2.See also additional information on predictive trigger functions. 

fuzzytime (sec)  
Checking how much the passive agent time differs from the Zabbix server/proxy time.  sec  seconds  Supported value types: float, int Returns: 1  difference between the passive item value (as timestamp) and Zabbix server/proxy timestamp is less than or equal to T seconds 0  otherwise Usually used with the 'system.localtime' item to check that local time is in sync with the local time of Zabbix server. Note that 'system.localtime' must be configured as a passive check. Can be used also with vfs.file.time[/path/file,modify] key to check that file didn't get updates for long time. Example: ⇒ fuzzytime(60)=0 → detect a problem if time difference is over 60 seconds This function is not recommended for use in complex trigger expressions (with multiple items involved), because it may cause unexpected results (time difference will be measured with the most recent metric), e.g. in Host:system.localtime.fuzzytime(60)}=0 or {Host:trap.last()}<>0 

iregexp (<pattern>,<sec#num>)  
This function is a non casesensitive analogue of regexp().  see regexp()  Supported value types: str, log, text  
last (<sec#num>,<time_shift>)  
The most recent value.  sec (ignored, equals #1) or #num (optional)  the Nth most recent value time_shift (optional)  evaluation point is moved the number of seconds back in time  Supported value types: float, int, str, text, log Take note that #num works differently here than with many other functions.For example: last() is always equal to last(#1) last(#3)  third most recent value (not three latest values) Zabbix does not guarantee exact order of values if more than two values exist within one second in history. The #num parameter is supported since Zabbix 1.6.2.The time_shift parameter is supported since Zabbix 1.8.2. 

logeventid (<pattern>)  
Checking if event ID of the last log entry matches a regular expression.  pattern (optional)  regular expression describing the required pattern, Perl Compatible Regular Expression (PCRE) style.  Supported value types: log Returns: 0  does not match 1  matches This function is supported since Zabbix 1.8.5. 

logseverity  
Log severity of the last log entry.  Supported value types: log Returns: 0  default severity N  severity (integer, useful for Windows event logs: 1  Information, 2  Warning, 4  Error, 7  Failure Audit, 8  Success Audit, 9  Critical, 10  Verbose). Zabbix takes log severity from Information field of Windows event log. 

logsource (<pattern>)  
Checking if log source of the last log entry matches a regular expression.  pattern (optional)  regular expression describing the required pattern, Perl Compatible Regular Expression (PCRE) style.  Supported value types: log Returns: 0  does not match 1  matches Normally used for Windows event logs. For example, logsource("VMware Server"). 

max (sec#num,<time_shift>)  
Highest value of an item within the defined evaluation period.  sec or #num  maximum evaluation period^{1} in seconds or in latest collected values (preceded by a hash mark) time_shift (optional)  evaluation point is moved the number of seconds back in time  Supported value types: float, int The time_shift parameter is supported since Zabbix 1.8.2. 

min (sec#num,<time_shift>)  
Lowest value of an item within the defined evaluation period.  sec or #num  maximum evaluation period^{1} in seconds or in latest collected values (preceded by a hash mark) time_shift (optional)  evaluation point is moved the number of seconds back in time  Supported value types: float, int The time_shift parameter is supported since Zabbix 1.8.2. 

nodata (sec,<mode>)  
Checking for no data received.  sec  evaluation period in seconds. The period should not be less than 30 seconds because the history syncer process calculates this function only every 30 seconds. nodata(0) is disallowed. mode  if set to strict, this function will be insensitive to proxy availability (see comments for details).  Supported value types: any Returns: 1  if no data received during the defined period of time 0  otherwise Since Zabbix 5.0, the 'nodata' triggers monitored by proxy are, by default, sensitive to proxy availability  if proxy becomes unavailable, the 'nodata' triggers will not fire immediately after a restored connection, but will skip the data for the delayed period. Note that for passive proxies suppression is activated if connection is restored more than 15 seconds and no less than 2 & ProxyUpdateFrequency seconds later. For active proxies suppression is activated if connection is restored more than 15 seconds later. To turn off sensitiveness to proxy availability, use the second parameter, e.g.: nodata(5m,strict); in this case the function will work the same as before 5.0.0 and fire as soon as the evaluation period (five minutes) without data has past. Note that this function will display an error if, within the period of the 1st parameter:  there's no data and Zabbix server was restarted  there's no data and maintenance was completed  there's no data and the item was added or reenabled Errors are displayed in the Info column in trigger configuration. This function may not work properly if there are time differences between Zabbix server, proxy and agent. See also: Time synchronization requirement. 

now  
Number of seconds since the Epoch (00:00:00 UTC, January 1, 1970).  Supported value types: any  
percentile (sec#num,<time_shift>,percentage)  
Pth percentile of a period, where P (percentage) is specified by the third parameter.  sec or #num  maximum evaluation period^{1} in seconds or in latest collected values (preceded by a hash mark) time_shift (optional)  evaluation point is moved the number of seconds back in time percentage  a floatingpoint number between 0 and 100 (inclusive) with up to 4 digits after the decimal point  Supported value types: float, int This function is supported since Zabbix 3.0.0. 

prev  
Previous value.  Supported value types: float, int, str, text, log Returns the same as last(#2). 

regexp (<pattern>,<sec#num>)  
Checking if the latest (most recent) value matches regular expression.  pattern (optional)  regular expression, Perl Compatible Regular Expression (PCRE) style. sec or #num (optional)  maximum evaluation period^{1} in seconds or in latest collected values (preceded by a hash mark). In this case, more than one value may be processed.  Supported value types: str, text, log Returns: 1  found 0  otherwise If more than one value is processed, '1' is returned if there is at least one matching value. This function is casesensitive. 

str (<pattern>,<sec#num>)  
Finding a string in the latest (most recent) value.  pattern (optional)  required string sec or #num (optional)  maximum evaluation period^{1} in seconds or in latest collected values (preceded by a hash mark). In this case, more than one value may be processed.  Supported value types: str, text, log Returns: 1  found 0  otherwise If more than one value is processed, '1' is returned if there is at least one matching value. This function is casesensitive. Tip: You may use the 'count' function with the like operator to count string values that match a pattern. 

strlen (<sec#num>,<time_shift>)  
Length of the latest (most recent) value in characters (not bytes).  sec (ignored, equals #1) or #num (optional)  the Nth most recent value time_shift (optional)  evaluation point is moved the number of seconds back in time  Supported value types: str, text, log Take note that #num works differently here than with many other functions.Examples: ⇒ strlen()(is equal to strlen(#1)) → length of the latest value ⇒ strlen(#3) → length of the third most recent value ⇒ strlen(,1d) → length of the most recent value one day ago. This function is supported since Zabbix 1.8.4. 

sum (sec#num,<time_shift>)  
Sum of collected values within the defined evaluation period.  sec or #num  maximum evaluation period^{1} in seconds or in latest collected values (preceded by a hash mark) time_shift (optional)  evaluation point is moved the number of seconds back in time  Supported value types: float, int The time_shift parameter is supported since Zabbix 1.8.2. 

time  
Current time in HHMMSS format.  Supported value types: any Example of returned value: 123055 

timeleft (sec#num,<time_shift>,threshold,<fit>)  
Time in seconds needed for an item to reach a specified threshold.  sec or #num  maximum evaluation period^{1} in seconds or in latest collected values (preceded by a hash mark) time_shift (optional)  evaluation point is moved the number of seconds back in time threshold  value to reach fit (optional)  see forecast()  Supported value types: float, int If value to return is larger than 1.7976931348623157E+308, return value is cropped to 1.7976931348623157E+308. Returns 1.7976931348623157E+308 if threshold cannot be reached. Becomes not supported only if misused in expression (wrong item type, invalid parameters), otherwise returns 1 in case of errors. Examples: ⇒ timeleft(#10,,0) → time until item value reaches zero based on last 10 values ⇒ timeleft(1h,,100) → time until item value reaches 100 based on last hour data ⇒ timeleft(1h,1d,0) → time until item value reaches 0 based on one hour one day ago ⇒ timeleft(1h,,200,polynomial2) → time until item reaches 200 based on last hour data and assumption that item behaves like quadratic (second degree) polynomial This function is supported since Zabbix 3.0.0. Unit symbols in threshold parameter are supported since Zabbix 3.0.6 and 3.2.2.See also additional information on predictive trigger functions. 

trendavg (period,period_shift)  
Average of trend values within the defined time period.  period  the time period (minimum '1h'), defined as <N><time unit> whereN  number of time unitstime unit  h (hour), d (day), w (week), M (month) or y (year).period_shift  the time period offset: starts with now  specifying the current time, and is followed by any combination of time operations:/<time unit>  defines the beginning and end of the time unit, for example, midnight to midnight for a day, MondaySunday for a week, first daylast day of the month for a month,+N<time unit> or N<time unit>  adds or subtracts N time units (see examples)  Examples: ⇒ trendavg(1h,now/h) → average for the current hour (e.g. 12:0013:00) ⇒ trendavg(1h,now/h1h) → average for the previous hour (11:0012:00) ⇒ trendavg(1h,now/h2h) → average for two hours ago (10:0011:00) ⇒ trendavg(1M,now/M1y) → average for the same month a year ago Note that time operations of the period_shift parameter are calculated from left to right without priorities. For example, now/d+1h/w will be parsed like ((now/d)+1h)/w .This function is supported since 5.2.0. 

trendcount (period,period_shift)  
Number of successfully retrieved trend values within the defined time period.  period  the time period (minimum '1h'), defined as <N><time unit> whereN  number of time unitstime unit  h (hour), d (day), w (week), M (month) or y (year).period_shift  the time period offset: starts with now  specifying the current time, and is followed by any combination of time operations:/<time unit>  defines the beginning and end of the time unit, for example, midnight to midnight for a day, MondaySunday for a week, first daylast day of the month for a month,+N<time unit> or N<time unit>  adds or subtracts N time units (see examples)  Examples: ⇒ trendcount(1h,now/h) → count for the current hour (e.g. 12:0013:00) ⇒ trendcount(1h,now/h1h) → count for the previous hour (11:0012:00) ⇒ trendcount(1h,now/h2h) → count for two hours ago (10:0011:00) ⇒ trendcount(1M,now/M1y) → count for the same month a year ago Note that time operations of the period_shift parameter are calculated from left to right without priorities. For example, now/d+1h/w will be parsed like ((now/d)+1h)/w .This function is supported since 5.2.0. 

trenddelta (period,period_shift)  
Difference between the maximum and minimum in trend values within the defined time period.  period  the time period (minimum '1h'), defined as <N><time unit> whereN  number of time unitstime unit  h (hour), d (day), w (week), M (month) or y (year).period_shift  the time period offset: starts with now  specifying the current time, and is followed by any combination of time operations:/<time unit>  defines the beginning and end of the time unit, for example, midnight to midnight for a day, MondaySunday for a week, first daylast day of the month for a month,+N<time unit> or N<time unit>  adds or subtracts N time units (see examples)  Examples: ⇒ trenddelta(1h,now/h) → delta for the current hour (e.g. 12:0013:00) ⇒ trenddelta(1h,now/h1h) → delta for the previous hour (11:0012:00) ⇒ trenddelta(1h,now/h2h) → delta for two hours ago (10:0011:00) ⇒ trenddelta(1M,now/M1y) → delta for the same month a year ago Note that time operations of the period_shift parameter are calculated from left to right without priorities. For example, now/d+1h/w will be parsed like ((now/d)+1h)/w .This function is supported since 5.2.0. 

trendmax (period,period_shift)  
The maximum in trend values within the defined time period.  period  the time period (minimum '1h'), defined as <N><time unit> whereN  number of time unitstime unit  h (hour), d (day), w (week), M (month) or y (year).period_shift  the time period offset: starts with now  specifying the current time, and is followed by any combination of time operations:/<time unit>  defines the beginning and end of the time unit, for example, midnight to midnight for a day, MondaySunday for a week, first daylast day of the month for a month,+N<time unit> or N<time unit>  adds or subtracts N time units (see examples)  Examples: ⇒ trendmax(1h,now/h) → maximum for the current hour (e.g. 12:0013:00) ⇒ trendmax(1h,now/h1h) → maximum for the previous hour (11:0012:00) ⇒ trendmax(1h,now/h2h) → maximum for two hours ago (10:0011:00) ⇒ trendmax(1M,now/M1y) → maximum for the same month a year ago Note that time operations of the period_shift parameter are calculated from left to right without priorities. For example, now/d+1h/w will be parsed like ((now/d)+1h)/w .This function is supported since 5.2.0. 

trendmin (period,period_shift)  
The minimum in trend values within the defined time period.  period  the time period (minimum '1h'), defined as <N><time unit> whereN  number of time unitstime unit  h (hour), d (day), w (week), M (month) or y (year).period_shift  the time period offset: starts with now  specifying the current time, and is followed by any combination of time operations:/<time unit>  defines the beginning and end of the time unit, for example, midnight to midnight for a day, MondaySunday for a week, first daylast day of the month for a month,+N<time unit> or N<time unit>  adds or subtracts N time units (see examples)  Examples: ⇒ trendmin(1h,now/h) → minimum for the current hour (e.g. 12:0013:00) ⇒ trendmin(1h,now/h1h) → minimum for the previous hour (11:0012:00) ⇒ trendmin(1h,now/h2h) → minimum for two hours ago (10:0011:00) ⇒ trendmin(1M,now/M1y) → minimum for the same month a year ago Note that time operations of the period_shift parameter are calculated from left to right without priorities. For example, now/d+1h/w will be parsed like ((now/d)+1h)/w .This function is supported since 5.2.0. 

trendsum (period,period_shift)  
Sum of trend values within the defined time period.  period  the time period (minimum '1h'), defined as <N><time unit> whereN  number of time unitstime unit  h (hour), d (day), w (week), M (month) or y (year).period_shift  the time period offset: starts with now  specifying the current time, and is followed by any combination of time operations:/<time unit>  defines the beginning and end of the time unit, for example, midnight to midnight for a day, MondaySunday for a week, first daylast day of the month for a month,+N<time unit> or N<time unit>  adds or subtracts N time units (see examples)  Examples: ⇒ trendsum(1h,now/h) → sum for the current hour (e.g. 12:0013:00) ⇒ trendsum(1h,now/h1h) → sum for the previous hour (11:0012:00) ⇒ trendsum(1h,now/h2h) → sum for two hours ago (10:0011:00) ⇒ trendsum(1M,now/M1y) → sum for the same month a year ago Note that time operations of the period_shift parameter are calculated from left to right without priorities. For example, now/d+1h/w will be parsed like ((now/d)+1h)/w .This function is supported since 5.2.0. 
^{1} The function is evaluated starting with the first received value (unless the time_shift
parameter is used).
Since Zabbix 3.2, nodata(), date(), dayofmonth(), dayofweek(), now() and time() functions are calculated for unsupported items, too. Other functions require that the referenced item is in a supported state.