$PCP_BINADM_DIR/pmlogger_check [-CNPpqsTV?] [-c control] [-l logfile]
$PCP_BINADM_DIR/pmlogger_daily [-EfKMNoprRV?] [-c control] [-k time] [-l logfile] [-m addresses] [-s size] [-t want] [-x time] [-X program] [-Y regex]
These shell scripts and associated control files may be used to create a customized regime of administration and management for Performance Co-Pilot (see PCPIntro(1)) archive log files.
pmlogger_check may be run at any time of the day and is intended to check that a desired set of pmlogger(1) processes are running. If not, it (re-)starts any missing logger processes.
pmlogger_daily is intended to be run once per day, preferably in the early morning, as soon after midnight as practicable. Its task is to aggregate, rotate and perform general housekeeping one or more sets of PCP archives.
To accommodate the evolution of PMDAs and changes in production logging environments, pmlogger_daily is integrated with pmlogrewrite(1) to allow optional and automatic rewriting of archives before merging. If there are global rewriting rules to be applied across all archives mentioned in the control file(s), then create the directory $PCP_SYSCONF_DIR/pmlogrewrite and place any pmlogrewrite(1) rewriting rules in this directory. For rewriting rules that are specific to only one family of archives, use the directory name from the control file(s) - i.e. the fourth field - and create a file, or a directory, or a symbolic link named pmlogrewrite within this directory and place the required rewriting rule(s) in the pmlogrewrite file or in files within the pmlogrewrite subdirectory. pmlogger_daily will choose rewriting rules from the archive directory if they exist, else rewriting rules from $PCP_SYSCONF_DIR/pmlogrewrite if that directory exists, else no rewriting is attempted.
As an alternate mechanism, if the file $PCP_LOG_DIR/pmlogger/.NeedRewrite exists when pmlogger_daily starts then this is treated the same as specifying -R on the command line and $PCP_LOG_DIR/pmlogger/.NeedRewrite will be removed once all the rewriting has been done.
- -c control, --control=control
Both pmlogger_check and pmlogger_daily are controlled by PCP logger control file(s) that specifies the pmlogger instances to be managed. The default control file is $PCP_PMLOGGERCONTROL_PATH, but an alternate may be specified using the -c option. If the directory $PCP_PMLOGGERCONTROL_PATH.d (or control.d from the -c option) exists, then the contents of any additional control files therein will be appended to the main control file (which must exist).
This option causes pmlogger_check to query the system service runlevel information for pmlogger, and use that to determine whether to start processes or not.
- -E, --expunge
This option causes pmlogger_daily to pass the -E flag to pmlogger_merge in order to expunge metrics with metadata inconsistencies and continue rather than fail. This is intended for automated daily log rotation where it is highly desirable for unattended daily archive merging, rewriting and compression to succeed. For further details, see pmlogger_merge(1) and description for the -x flag in pmlogextract(1).
- -f, --force
This option causes pmlogger_daily to forces action. Using this option in production is not recommended.
- -k time, --discard=time
After some period, old PCP archives are discarded. time is a time specification in the syntax of find-filter(1), so DD[:HH[:MM]]. The optional HH (hours) and MM (minutes) parts are 0 if not specified. By default the time is 14:0:0 or 14 days, but may be changed using this option.
Some special values are recognized for the time, namely 0 to keep no archives beyond the the ones being currently written by pmlogger(1), and forever or never to prevent any archives being discarded.
The time can also be set using the $PCP_CULLAFTER variable, set in either the environment or in a control file. If both $PCP_CULLAFTER and -k specify different values for time then the environment variable value is used and a warning is issued. I.e., if $PCP_CULLAFTER is set in the control file, it overrides -k given on the command line.
Note that the semantics of time are that it is measured from the time of last modification of each archive, and not from the original archive creation date. This has subtle implications for compression (see below) - the compression process results in the creation of new archive files which have new modification times. In this case, the time period (re)starts from the time of compression.
When this option is specified for pmlogger_daily then only the compression tasks are attempted, so no pmlogger rotation, no culling, no rewriting, etc. When -K is used and a period of 0 is in effect (from -x on the command line or $PCP_COMPRESSAFTER in the environment or via the control file) this is intended for environments where compression of archives is desired before the scheduled daily processing happens. To achieve this, once pmlogger_check has completed regular processing, it calls pmlogger_daily with just the -K option. Provided $PCP_COMPRESSAFTER is set to 0 along with any other required compression options to match the scheduled invocation of pmlogger_daily, then this will compress all volumes except the ones being currently written by pmlogger(1). If $PCP_COMPRESSAFTER is set to a value greater than zero, then manually running pmlogger_daily with the -x option may be used to compress volumes that are younger than the $PCP_COMPRESSAFTER time. This may be used to reclaim filesystem space by compressing volumes earlier than they would have otherwise been compressed. Note that since the default value of $PCP_COMPRESSAFTER is 0 days, the -x option has no effect unless the control file has been edited and $PCP_COMPRESSAFTER has been set to a value greater than 0.
- -l file, --logfile=file
In order to ensure that mail is not unintentionally sent when these scripts are run from cron(8) diagnostics are always sent to log files. By default, this file is $PCP_LOG_DIR/pmlogger/pmlogger_check.log or $PCP_LOG_DIR/pmlogger/pmlogger_daily.log but this can be changed using the -l option. If this log file already exists when the script starts, it will be renamed with a .prev suffix (overwriting any log file saved earlier) before diagnostics are generated to the log file. The -l and -t options cannot be used together.
- -m addresses, --mail=addresses
Use of this option causes pmlogger_daily to construct a summary of the “notices” file entries which were generated in the last 24 hours, and e-mail that summary to the set of space-separated addresses. This daily summary is stored in the file $PCP_LOG_DIR/NOTICES.daily, which will be empty when no new “notices” entries were made in the previous 24 hour period.
This option may be used to disable archive merging (or renaming) and rewriting (-M implies -r). This is most useful in cases where the archives are being incrementally copied to a remote repository, e.g. using rsync(1). Merging, renaming and rewriting all risk an increase in the synchronization load, especially immediately after pmlogger_daily has run, so -M may be useful in these cases.
- -N, --showme
This option enables a “show me” mode, where the programs actions are echoed, but not executed, in the style of “make -n”. Using -N in conjunction with -V maximizes the diagnostic capabilities for debugging.
By default all possible archives will be merged. This option reinstates the old behaviour in which only yesterday's archives will be considered as merge candidates. In the special case where only a single input archive needs to be merged, pmlogmv(1) is used to rename the archive, otherwise pmlogger_merge(1) is used to merge all of the archives for a single host and a single day into a new PCP archive and the individual archives are removed.
- -p, --skip-primary
If this option is specified for pmlogger_check then any line from the control files for the primary pmlogger will be ignored. This option is intended for environments where some system daemon, like systemd(1), is responsible for controlling (starting, stopping, restarting, etc.) the primary pmlogger.
- -P, --only-primary
If this option is specified for pmlogger_check then only the primary logger entry in the control files will be processed. This is the logical opposite of the -p option described above and is intended for use by RC scripts that start only the primary logger, such as the pmlogger.service unit. The -p and -P options to pmlogger_check are mutually exclusive.
If this option is specified for pmlogger_daily then the status of the daily processing is polled and if the daily pmlogger(1) rotation, culling, rewriting, compressing, etc. has not been done in the last 24 hours then it is done now. The intent is to have pmlogger_daily called regularly with the -p option (at 30 mins past the hour, every hour in the default cron(8) set up) to ensure daily processing happens as soon as possible if it was missed at the regularly scheduled time (which is 00:10 by default), e.g. if the system was down or suspended at that time. With this option pmlogger_daily simply exits if the previous day's processing has already been done. Note that this option is not used on platforms supporting systemd(1) because the pmlogger_daily.timer service unit specifies a timer setting with Persistent=true. The -K and -p options to pmlogger_daily are mutually exclusive.
- -q, --quick
If this option is specified for pmlogger_check then the script will “quickstart” avoiding any optional processing like file compression.
- -r, --norewrite
This command line option acts as an override and prevents all archive rewriting with pmlogrewrite(1) independent of the presence of any rewriting rule files or directories.
- -R, --rewriteall
Sometimes PMDA changes require all archives to be rewritten, not just the ones involved in any current merging. This is required for example after a PCP upgrade where a new version of an existing PMDA has revised metadata. The -R command line forces this universal-style of rewriting. The -R option to pmlogger_daily is mutually exclusive with both the -r and -M options.
- -s size, --rotate=size
If the PCP “notices” file ($PCP_LOG_DIR/NOTICES) is larger than 20480 bytes, pmlogger_daily will rename the file with a “.old” suffix, and start a new “notices” file. The rotate threshold may be changed from 20480 to size bytes using the -s option.
- -s, --stop
Use of this option provides the reverse pmlogger_check functionality, allowing the set of pmlogger processes to be cleanly shutdown.
- -t period
To assist with debugging or diagnosing intermittent failures the -t option may be used. This will turn on very verbose tracing (-VV) and capture the trace output in a file named $PCP_LOG_DIR/pmlogger/daily.datestamp.trace, where datestamp is the time pmlogger_daily was run in the format YYYYMMDD.HH.MM. In addition, the period argument will ensure that trace files created with -t will be kept for period days and then discarded.
- -T, --terse
This option to pmlogger_check produces less verbose output than the default. This is most suitable for a pmlogger “farm” where many instances of pmlogger are expected to be running.
- -V, --verbose
The output from the cron execution of the scripts may be extended using the -V option to the scripts which will enable verbose tracing of their activity. By default the scripts generate no output unless some error or warning condition is encountered. Using -N in conjunction with -V maximizes the diagnostic capabilities for debugging.
- -x time, --compress-after=time
Archive data files can optionally be compressed after some period to conserve disk space. This is particularly useful for large numbers of pmlogger processes under the control of pmlogger_check.
time is a time specification in the syntax of find-filter(1), so DD[:HH[:MM]]. The optional HH (hours) and MM (minutes) parts are 0 if not specified.
Some special values are recognized for the time, namely 0 to apply compression as soon as possible, and forever or never to prevent any compression being done.
If transparent_decompress is enabled when libpcp was built (can be checked with the pmconfig(1) -L option), then the default behaviour is compression “as soon as possible”. Otherwise the default behaviour is to not compress files (which matches the historical default behaviour in earlier PCP releases).
The time can also be set using the $PCP_COMPRESSAFTER variable, set in either the environment or in a control file. If both $PCP_COMPRESSAFTER and -x specify different values for time then the environment variable value is used and a warning is issued. For important other detailed notes concerning volume compression, see the -K and -k options (above).
- -X program, --compressor=program
This option specifies the program to use for compression - by default this is xz(1). The environment variable $PCP_COMPRESS may be used as an alternative mechanism to define program. If both $PCP_COMPRESS and -X specify different compression programs then the environment variable value is used and a warning is issued.
- -Y regex, --regex=regex
This option allows a regular expression to be specified causing files in the set of files matched for compression to be omitted - this allows only the data file to be compressed, and also prevents the program from attempting to compress it more than once. The default regex is ".(index|Z|gz|bz2|zip|xz|lzma|lzo|lz4)$" - such files are filtered using the -v option to egrep(1). The environment variable $PCP_COMPRESSREGEX may be used as an alternative mechanism to define regex. If both $PCP_COMPRESSREGEX and -Y specify different values for regex then the environment variable value is used and a warning is issued.
- -?, --help
Display usage message and exit.
Warning: The $PCP_PMLOGGERCONTROL_PATH file and files within the $PCP_PMLOGGERCONTROL_PATH.d directory must not be writable by any user other than root.
The control file(s) should be customized according to the following rules that define for the current version (1.1) of the control file format.
- Lines beginning with a “#” are comments. A special case is lines beginning “#!#”; these are control lines for a pmlogger(1) that has been stopped using pmlogctl(1).
Lines beginning with a “$” are assumed to be assignments to environment variables in the style of sh(1), and all text following the “$” will be eval'ed by the script reading the control file, and the corresponding variable exported into the environment. This is particularly useful to set and export variables into the environment of the administrative scripts, e.g.
There must be a version line in the initial control file of the form:
There should be one line in the control file(s) for each pmlogger instance of the form:
host y|n y|n directory args
- Fields within a line of the control file(s) are usually separated by one or more spaces or tabs (although refer to the description of the directory field for some important exceptions).
- The first field is the name of the host that is the source of the performance metrics for this pmlogger instance.
- The second field indicates if this is a primary pmlogger instance (y) or not (n). Since the primary logger must run on the local host, and there may be at most one primary logger for a particular host, this field can be y for at most one pmlogger instance, in which case the host name must be the name of the local host.
- The third field indicates if this pmlogger instance needs to be started under the control of pmsocks(1) to connect to a pmcd through a firewall (y or n).
- The fourth field is a directory name. All files associated with this pmlogger instance will be created in this directory, and this will be the current directory for the execution of any programs required in the maintenance of those archives. A useful convention is that primary logger archives for the local host with hostname myhost are maintained in the directory $PCP_ARCHIVE_DIR/myhost (this is where the default pmlogger start-up script in $PCP_RC_DIR/pcp will create the archives), while archives for the remote host mumble are maintained in $PCP_ARCHIVE_DIR/mumble.
The directory field may contain embedded shell syntax that will be evaluated by sh(1) to produce the real directory name to be used. The allowed constructs are:
- Any text (including white space) enclosed with $( and ).
- Any text (including white space) enclosed with ` and ` (back quotes).
- Any text (including white space) enclosed with " and " (double quotes).
- Any word containing a $ (assumed to introduce an environment variable name).
- All other fields are interpreted as arguments to be passed to pmlogger(1). Most typically this would be the -c option.
The following sample control lines specify a primary logger on the local host (bozo), and non-primary loggers to collect and log performance metrics from the hosts wobbly and boing.
$version=1.1 bozo y n $PCP_ARCHIVE_DIR/bozo -c config.default wobbly n n "/store/wobbly/$(date +%Y)" -c ./wobbly.config boing n n $PCP_ARCHIVE_DIR/boing -c ./pmlogger.config
Typical crontab(5) entries for periodic execution of pmlogger_daily and pmlogger_check are given in $PCP_SYSCONF_DIR/pmlogger/crontab (unless installed by default in /etc/cron.d already) and shown below.
# daily processing of archive logs 14 0 * * * $PCP_BINADM_DIR/pmlogger_daily # every 30 minutes, check pmlogger instances are running 25,55 * * * * $PCP_BINADM_DIR/pmlogger_check
When using systemd(1) on Linux, no crontab entries are needed as the timer mechanism provided by systemd is used instead.
the PCP logger control file. For a new installation this file contains no pmlogger(1) control lines (the real control files are all in the $PCP_PMLOGGERCONTROL_PATH.d directory), but this file is still processed to support any legacy configurations therein from earlier PCP releases.
Warning: this file must not be writable by any user other than root.
optional directory containing additional PCP logger control files, typically one per host
Warning: the files herein must not be writable by any user other than root.
sample crontab for automated script execution by $PCP_USER (or root). Exists only if the platform does not support the /etc/cron.d mechanism.
default pmlogger configuration file location for the local primary logger, typically generated automatically by pmlogconf(1).
default location for archives of performance information collected from the host hostname
transient lock file to guarantee mutual exclusion during pmlogger administration for the host hostname - if present, can be safely removed if neither pmlogger_daily nor pmlogger_check are running
PCP archive folio created by mkaf(1) for the most recently launched archive containing performance metrics from the host hostname
PCP “notices” file used by pmie(1) and friends
if the previous execution of pmlogger_check produced any output it is saved here. The normal case is no output in which case the file does not exist.
if the previous execution of pmlogger_daily produced any output it is saved here. The normal case is no output in which case the file does not exist.
if this directory exists, then the log file from the -l argument of a newly launched pmlogger(1) for hostname will be linked into this directory with the name archive.log where archive is the basename of the associated pmlogger(1) PCP archive files. This allows the log file to be inspected at a later time, even if several pmlogger(1) instances for hostname have been launched in the interim. Because the cron-driven PCP archive management scripts run under the uid of the user “pcp”, $PCP_ARCHIVE_DIR/hostname/SaveLogs typically needs to be owned by the user “pcp”.
if this file exists, then this is treated as equivalent to using -R on the command line and the file will be removed once all rewriting has been done.
Environment variables with the prefix PCP_ are used to parameterize the file and directory names used by PCP. On each installation, the file /etc/pcp.conf contains the local values for these variables. The $PCP_CONF variable may be used to specify an alternative configuration file, as described in pcp.conf(5).
The default behaviour, when pmlogger(1) configuration comes from pmlogconf(1), is to regenerate the configuration file and check for changes whenever pmlogger(1) is started from pmlogger_check. If the PMDA configuration is stable, this is not necessary, and setting $PMLOGGER_CHECK_SKIP_LOGCONF to yes disables the regeneration and checking.
Earlier versions of pmlogger_daily used find(1) to locate files for compressing or culling and the -k and -x options took only integer values to mean “days”. The semantics of this was quite loose given that find(1) offers different precision and semantics across platforms.
The current implementation of pmlogger_daily uses find-filter(1) which provides high precision intervals and semantics that are relative to the time of execution and are consistent across platforms.
egrep(1), find-filter(1), PCPIntro(1), pmconfig(1), pmlc(1), pmlogconf(1), pmlogctl(1), pmlogger(1), pmlogger_daily_report(1), pmlogger_merge(1), pmlogextract(1), pmlogmv(1), pmlogrewrite(1), pmsocks(1), systemd(1), xz(1) and cron(8).
find-filter(1), LOGARCHIVE(5), pcp-atop(1), pcp-atopsar(1), PCPIntro(1), pmDiscoverSetup(3), pmdumplog(1), pmfind_check(1), pmlc(1), pmlogctl(1), pmlogextract(1), pmlogger(1), pmlogger_daily_report(1), pmlogger_merge(1), pmlogger_rewrite(1), pmloglabel(1), pmsearch(1), pmsnap(1).
The man page pmlogger_daily(1) is an alias of pmlogger_check(1).