'\"! tbl | mmdoc
'\"macro stdmacro
.\"
.\" Copyright (c) 2012-2018,2021 Red Hat.
.\" Copyright (c) 2000-2004 Silicon Graphics, Inc.  All Rights Reserved.
.\"
.\" This program is free software; you can redistribute it and/or modify it
.\" under the terms of the GNU General Public License as published by the
.\" Free Software Foundation; either version 2 of the License, or (at your
.\" option) any later version.
.\"
.\" This program is distributed in the hope that it will be useful, but
.\" WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
.\" or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
.\" for more details.
.\"
.\"
.TH PMFETCH 3 "PCP" "Performance Co-Pilot"
.SH NAME
\f3pmFetch\f1,
\f3pmFetchHighRes\f1,
\f3pmHighResFetch\f1 \- get performance metric values
.SH "C SYNOPSIS"
.ft 3
#include <pcp/pmapi.h>
.sp
.nf
int pmFetch(int \fInumpmid\fP, pmID *\fIpmidlist\fP, pmResult **\fIresult\fP);
.br
int pmFetchHighRes(int \fInumpmid\fP, pmID *\fIpmidlist\fP, pmHighResResult **\fIresult\fP);
.fi
.sp
cc ... \-lpcp
.ft 1
.SH DESCRIPTION
.de CR
.ie t \f(CR\\$1\fR\\$2
.el \fI\\$1\fR\\$2
..
.\" some useful acronyms ... always cite the full text at the first use
.\" and use uppercase acronym thereafter
.\" Performance Metrics Application Programming Interface (PMAPI)
.\" Performance Metrics Name Space (PMNS)
.\" Performance Metrics Collector Daemon (PMCD)
.\" Performance Metric Identifier (PMID)
Given a list of Performance Metric Identifiers (PMID)s,
e.g. as constructed by
.BR pmLookupName (3),
via
.I pmidlist
and
.IR numpmid ,
fetch the values for these performance metrics.
.PP
A call to either
.B pmFetch
or
.B pmFetchHighRes
is executed in the context of a source of metrics,
instance profile and collection time,
previously established by calls to
the appropriate context and profile functions, namely some of
.BR pmNewContext (3),
.BR pmDupContext (3),
.BR pmUseContext (3),
.BR pmAddProfile (3),
.BR pmDelProfile (3)
and
.BR pmSetMode (3).
.PP
The principal result from
.B pmFetch
is returned in the argument
.I result
as a tree, using the following component data structures;
.PP
.ft CR
.nf
.in +0.5i
typedef struct {
      unsigned int vtype : 8;        /* value type (same as pmDesc.type) */
      unsigned int vlen : 24;        /* bytes for vtype/vlen + vbuf */
      char         vbuf[1];          /* one or more values */
} pmValueBlock;

typedef struct {
      int      inst;                 /* instance identifier */
      union {
            pmValueBlock *pval;      /* pointer to value-block */
            int          lval;       /* integer value insitu */
      } value;
} pmValue;

typedef struct {
      pmID      pmid;                /* metric identifier */
      int       numval;              /* number of values or error code */
      int       valfmt;              /* value style, insitu or ptr */
      pmValue   vlist[1];            /* set of instances/values */
} pmValueSet;

/* Result returned by pmFetch() */
typedef struct {
      struct timeval timestamp;      /* time stamped by collector */
      int            numpmid;        /* number of PMIDs */
      pmValueSet     *vset[1];       /* set of value sets */
} pmResult;
.in
.fi
.ft 1
.PP
The principal result from
.B pmFetchHighRes
is exactly the same in terms of value sets, however a higher resolution
timestamp (nanosecond instead of microsecond precision) is available in
the
.CR pmHighResResult
structure.
.PP
.ft CR
.nf
.in +0.5i
/* Result returned by pmFetchHighRes() */
typedef struct {
      struct timespec timestamp;      /* time stamped by collector */
      int             numpmid;        /* number of PMIDs */
      pmValueSet      *vset[1];       /* set of value sets */
} pmHighResResult;
.in
.fi
.ft 1
.PP
To accommodate metrics with multiple value instances, the
.CR numval
field indicates how many values are returned for each requested PMID.
The field
.CR valfmt
in the
.CR pmValueSet
structure indicates if the values for this metric are stored
.I insitu
in the
.CR lval
field, i.e. a 32-bit integer quantity (either int, unsigned int,
long or unsigned long) or if the values are held in associated
.CR pmValueBlock
structures.
The
.CR pmValueBlock
structure is always used for floating point values (float or double)
and also accommodates arbitrary sized binary data such as
`string-valued' metrics and metrics with aggregated or complex data types.
The maximum length of a
.CR pmValueBlock
buffer is
.B PM_VAL_VLEN_MAX
bytes.
If the
.CR pmValueBlock
format is used, the
.CR vtype
field indicates the data type of the value.
This field has the same interpretation as the
.CR type
field in the
.B pmDesc
structure,
see
.BR pmLookupDesc (3).
.PP
Note that the insitu value may be a signed or unsigned 32 bit integer,
signed or unsigned 32 bit long value (on 32 bit platforms),
In the special cases described below, it may also be a 32 bit floating
point value.
If the application needs to know the type of an insitu value,
which is almost always the case, it is necessary to
fetch the descriptor for the metric
and interpret the
.B type
field, as described in detail in
.BR pmLookupDesc (3).
When the
.CR pmResult
is received from a PCP1.x
.BR pmcd ,
insitu values may also be 32 bit floating point values
(of type
.BR PM_TYPE_FLOAT ).
In all cases, it is good practice to use
.BR pmLookupDesc (3)
to fetch the descriptor for the metric and interpret the
.B type
field therein.
Note also that the
.BR PMAPI (3)
will automatically translate from the PCP2.0 format
to the PCP1.x format when a PCP1.x client requests 32 bit floating point values
from a PCP2.0
.BR pmcd ,
but the reverse translation does not occur (because the PCP2.0
.B pmcd
cannot automatically distinguish between arbitrary 32 bit floating point values
and 32 bit integers).
.PP
If one value (i.e. associated with a particular instance)
for a requested metric is `unavailable' (at the requested time),
then there is no associated
.CR pmValue
structure in the
.IR result .
If there are no available values for a metric,
then
.CR numval
will be zero and the associated
.CR pmValue[]
instance will be empty (\c
.CR valfmt
is undefined in these circumstances,
however
.CR pmid
will be correctly set to the PMID of the metric with no values).
.PP
As an extension of this protocol,
if the Performance Metrics Collection System (PMCS)
is able to provide a reason why no values are available
for a particular metric,
this is encoded as a standard error code in the corresponding
.CR numval .
Since the error codes are all negative,
values for a requested metric are `unavailable' if
.CR numval
is less than, or equal to, zero.
A performance metric's value may be `unavailable'
for a number of reasons; the following list is illustrative but
not exhaustive:
of the software for the associated Performance Metric Domain
.IP "+"
Collection is not currently activated
in the software for the associated Performance Metric Domain
.IP "+"
The associated PMID is not known
.IP "+"
The current system configuration does not include
the associated hardware component and/or the associated software module,
e.g. a disk is not installed, or off-line, or Oracle is not installed
.IP "+"
The metric is one for which an instance profile is required,
and none was provided (there are a small number of metrics in this category,
typically ones with very large, and/or very
dynamic instance domains, and/or expensive metric instantiation methods).
.IP "+"
If the current context involves fetching metrics from an archive,
values may be unavailable in the region around a
.I <mark>
record (see
.BR pmlogextract (1))
that indicate a temporal discontinuity in the time-series of
metric values.
.PP
In general, we may not be able to differentiate between the various cases,
and if differentiation is not possible,
.CR numval
will simply be zero.
.PP
The argument definition and the result specifications have been constructed
to ensure that for each PMID in the requested
.I pmidlist
there is exactly one
.CR pmValueSet
in the
.IR result ,
and further the PMIDs appear in exactly the same sequence in both
.I pmidlist
and
.IR result .
This makes the number
and order of entries in
.I result
completely deterministic,
and greatly simplifies the application programming logic
after the call to
.B pmFetchHighRes
or
.BR pmFetch .
.PP
The
.I result
structure returned by
.B pmFetch
is dynamically allocated using
a combination of
.BR malloc (3)
calls
and specialized allocation strategies,
and should be released when no longer required by calling
.BR pmFreeResult (3)
\- under no circumstances should
.BR free (3)
be called directly to release this space.
.PP
As common error conditions are encoded
in the
.I result
data structure, we'd expect only cataclysmic events
to cause an error value to be returned.
One example would be if the metrics source context was a remote host,
and that host or the PMCS on that host became unreachable.
Otherwise the value returned by the
.B pmFetch
function will be non-negative.
.PP
Similarly, the
.I result
structure returned by
.B pmFetchHighRes
operates under the same principles, and should be released via
.BR pmFreeHighResResult (3).
.PP
If the current context involves fetching metrics from a
Performance Metrics Collector Daemon (PMCD), then the return value
may be used to encode out-of-band changes in the state of the
PMCD and the associated
Performance Metrics Daemon Agents (PMDAs), as a bit-wise ``or'' of the
following values:
.sp 0.5v
.IP \fBPMCD_RESTART_AGENT\fR 20n
An attempt has been made to restart at least one failed PMDA.
.IP \fBPMCD_ADD_AGENT\fR
At least one PMDA has been started.
.IP \fBPMCD_DROP_AGENT\fR
PMCD has noticed the termination of at least one PMDA.
.IP \fBPMCD_AGENT_CHANGE\fR
A convenience macro for any of the three PMDA changes.
.IP \fBPMCD_LABEL_CHANGE\fR
PMCD has been informed of changes to global (context) labels,
or new metrics have appeared which have associated labels.
.IP \fBPMCD_NAMES_CHANGE\fR
PMCD has been informed that the namespace has been modified,
such that new metrics have appeared or existing metrics have
been removed.
.IP \fBPMCD_HOSTNAME_CHANGE\fR
The hostname on the host where PMCD is running has changed.
This may be the result of changes from temporary to permanent
hostname after a system reboot or
some subsequent explicit change to the system's hostname.
.PP
The default is to return zero to indicate no change in state,
however the
.CR pmResult
returned by
.B pmFetch
(or
.CR pmHighResResult
returned by
.BR pmFetchHighRes )
has the same interpretation independent of the return value being
zero or greater than zero.
.PP
.B pmHighResFetch
is a previous name for
.B pmFetchHighRes
that has been maintained for backwards compatibility.
.SH SEE ALSO
.BR pmcd (1),
.BR pmAddProfile (3),
.BR PMAPI (3),
.BR pmDelProfile (3),
.BR pmDupContext (3),
.BR pmExtractValue (3),
.BR pmFetchArchive (3),
.BR pmFreeHighResResult (3),
.BR pmFreeResult (3),
.BR pmGetInDom (3),
.BR pmLookupDesc (3),
.BR pmLookupLabels (3),
.BR pmLookupName (3),
.BR pmNewContext (3),
.BR pmSetMode (3),
.BR pmUseContext (3)
and
.BR pmWhichContext (3).
.PP
Note that
.B pmFetch
and
.B pmFetchHighRes
are the most primitive methods of fetching metric values from the PMCS.
See the
.BR pmFetchGroup (3)
API for a higher level method that insulates
the user from the intricacies of looking up metric names and metadata,
setting up instance profiles,
.CR pmResult
traversal, conversions, and scaling.
.SH DIAGNOSTICS
As mentioned above,
.B pmFetch
and
.B pmFetchHighRes
return error codes
.I insitu
in the argument
.IR result .
If no result is returned,
e.g. due to IPC failure using the current PMAPI context, or
end of file on an archive, then these routines
will return a negative error code which may be examined using
.BR pmErrStr (3).
.IP \f3PM_ERR_EOL\f1
When fetching records from an archive,
.B pmFetch
returns this error code to indicate the end of the archive has been
passed (or the start of the archive has been passed, if the direction
of traversal is backwards in time).
If the ``mode'' for the current PMAPI context (see
.BR pmSetMode (3))
is
.B PM_MODE_INTERP
then the time origin is advanced, even when this error code is
returned.
In this way applications that position the time outside the range
defined by the records in the archive, and then commence to
.B pmFetch
will eventually see valid results once the time origin moves inside
the temporal span of the archive.
.SH ENVIRONMENT
Many of the performance metrics exported from PCP agents have the
semantics of
.I counter
meaning they are expected to be monotonically increasing.
Under some circumstances, one value of these metrics may be smaller
than the previously fetched value.
This can happen when a counter of finite precision overflows, or
when the PCP agent has been reset or restarted, or when the
PCP agent is exporting values from some
underlying instrumentation that is subject to some asynchronous
discontinuity.
.sp 0.5v
The environment variable
.B PCP_COUNTER_WRAP
may be set to indicate that all such cases of a decreasing ``counter''
should be treated
as a counter overflow, and hence the values are assumed to have
wrapped once in the interval between consecutive samples.
This ``wrapping'' behavior was the default in earlier PCP versions, but
by default has been disabled in PCP version 1.3 and later.