table of contents
| PMLOGGER_JANITOR(1) | General Commands Manual | PMLOGGER_JANITOR(1) |
NAME¶
pmlogger_janitor - management helper for Performance Co-Pilot archive files
SYNOPSIS¶
$PCP_BINADM_DIR/pmlogger_janitor [-NV?] [-c control] [-l logfile] [-X program] [-Y regex]
DESCRIPTION¶
Typically pmlogger_janitor is not used directly, rather it is called from pmlogger_check(1) to clean up any residual processes and state files associated with moribund instances of pmlogger(1).
If pmlogger_janitor is run directly, it needs to be run as the user $PCP_USER; the only exception is when the -N option is used.
OPTIONS¶
- -c control, --control=control
- Like pmlogger_check (1) and pmlogger_daily(1), pmlogger_janitor is controlled by PCP logger control file(s) that specify 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).
- -l file, --logfile=file
- In order to ensure that mail is not unintentionally sent when this script is run from cron(8) or systemd(1) diagnostics are always sent to log files. By default, this file is $PCP_LOG_DIR/pmlogger/pmlogger_janitor.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.
- -N, --showme
- This option enables a ``show me'' mode, where the actions are echoed, but not executed, in the style of ``make -n''. Using -N in conjunction with -V maximizes the diagnostic capabilities for debugging.
- -V, --verbose
- The -V option enables verbose tracing. By default pmlogger_janitor generates no output unless some error or warning condition is encountered. A second -V increases the verbosity. Using -N in conjunction with -V maximizes the diagnostic capabilities for debugging.
- -X program, --compressor=program
- This option specifies the program to use for compression. Refer to pmlogger_daily(1) for details.
- -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. Refer to pmlogger_daily(1) for details.
- -?, --help
- Display usage message and exit.
OPERATION¶
pmlogger_janitor operates in a number of passes looking for things to cleanup.
- Pass 1
- Scan the pmlogger port map files in the $PCP_TMP_DIR/pmlogger directory. Use the process id (PID) from the name of the port map file and ps(1) to check and if the process exists and it ``looks like'' an active pmlogger; if so append the PID and the directory where the archive is being written (extracted from the port map file) to a list of interest, otherwise the port map file is assumed to be associated with a pmlogger that has exited and the port map file is removed.
- Pass 2
- Use ps(1) to scan all processes and identify any pmlogger instances that were started by pmlogger_check(3) or pmlogger_daily(3), these will have a signature -m option in their arguments. Append the PIDs and directories where the archive is being written (extracted from the command line arguments) to the list of interest.
- Pass 3
- Parse all of the pmlogger.control(5) files from either the standard places, or control if the -c option is used. For each pmlogger instance, it should be in the list of interest, and if so remove it from that list.
If the list of interest is empty at the end of this Pass, we're done.
- Pass 4
- Each process remaining in the list of interest is a pmlogger that is running, but should not be; probably as a result of a change to the pmlogger.control(5) files. Send each process a SIGTERM.
- Pass 5
- For each process remaining in the list of interest after Pass 3 compress any archives in the directory pmlogger was writing into. This is the only step where the -X and -Y options might be used.
- Pass 6
- For each pmlogger sent a SIGTERM in Pass 4 check that it has exited and if not send it a SIGKILL.
FILES¶
- $PCP_LOG_DIR/pmlogger/pmlogger_janitor.log
- if the previous execution of pmlogger_janitor produced any output it is saved here. The normal case is no output in which case the file does not exist.
- $PCP_ARCHIVE_DIR/SaveLogs
- if this directory exists, then the log file from the -l argument for pmlogger_janitor will be saved in this directory with the name of the format <date>-pmlogger_janitor.log.<pid> This allows the log file to be inspected at a later time, even if several pmlogger_janitor executions have been launched in the interim. $PCP_ARCHIVE_DIR/SaveLogs needs to be owned by the user $PCP_USER.
- $PCP_TMP_DIR/pmlogger
- pmlogger maintains the files in this directory as the port map between the PID of the pmlogger instance, the IPC port that may be used to control each pmlogger instance (as used by pmlc(1)), the name of the host where pmcd(1) is providing the metrics for pmlogger, the remote connection string for the remote HTTP server that is receiving the logged data or the full path to the archive base name for a local logger and any annotation from the -m or -x options; all of this information is also available via the pmcd.pmlogger.* metrics, e.g. $ pminfo -f pmcd.pmlogger
PCP ENVIRONMENT¶
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).
SEE ALSO¶
PCPIntro(1), pmcd(1), pmlc(1), pmlogctl(1), pmlogger(1), pmlogger_check(1), pmlogger_daily(1), ps(1), systemd(1), pmlogger.control(5) and cron(8).
| PCP | Performance Co-Pilot |