2024-03-06 - Version 4.0 This release is a major release that fixes issues reported since the past six months and adds several new features. New features: * Add report of devices /sys/block/queue informations (scheduler, rotational, rq_affinity and add_random) to the System report. * Add report of PGDATA files to be able to detect symlinks and unwanted files in this directory. * Add reports for Wait Event when the pg_wait_sampling extension is use. First report show the distribution by Wait Event Type and second report show the distribution by Wait Events. * Add report for Global subtransactions counters if extension pg_subxact_counters is installed in the connection database. Subtransactions can lead to performance issue, report Counters to monitor the subtransactions (generation rate, overflow, state). * Add collect of /sys/kernel/debug/sched/migration_cost_ns with kernel >= 5.19 collect start-end metrics only twice (start and end) * Add a chapter in documentation about additional statistics collected from extensions pg_stat_statements, pg_wait_sampling and pg_subxact_counters. * Add option --no-pg_stats-dump to pgcluu_collectd to avoid collecting such statistics for performances reasons. Thanks for the patch to Frederic Yhuel. * Make SIGINT interrupts gracefully pgcluu_collectd like SIGHUP. Thanks to Frederic Yhuel for the patch. * Add ddump of the pg_stats view to pgcluu_collectd. The point is for the DBA to study the statistics, and nothing else. Thanks for the patch to Frederic Yhuel. It is useful to dump the pg_stats views. The dump can be imported in a table created like this: CREATE TABLE my_client_pgstats ( export_time TEXT, dbname TEXT, schemaname TEXT, tablename TEXT, attname TEXT, inherited BOOLEAN, null_frac REAL, avg_width INT, n_distinct REAL, most_common_vals TEXT[], most_common_freqs REAL[], histogram_bounds TEXT[], correlation REAL ); Note that the columns most_common_vals and histogram_bounds have the type TEXT[] instead of anyarray. Compatibility: There is not compatibility issues but a change of behavior with default to dump the pg_stats view. Following the number of table/indexes in the databases it could take more time collecting the statistic so in this case the interval of polling should be increase. There is an new option to disable the pg_stats dump, --no-pg_stats-dump, if you want to get back to the previous behavior. Here is the full list of fixes since previous release. - Fix graph labels for TCPv4 errors. Thanks to Christophe Courtois for the report. - Avoid reporting redundant index if they are not using the same type of index. Thanks to Christophe Courtois for the report. - Fix file not exists error message when looking at block devices information. - Fix broken report when /sys/kernel/debug/sched/migration_cost_ns is not readable by the user running pgcluu_colletd - Fix devices info (sar -d) for sar >= 12.4.0. Thanks to Christophe Courtois for the patch. - Fix legend of locks reports. - Fix report of locks per database. - Fix disable menu for buffercache use - Fix mount report that was not reporting NFS mount points. - Fix wait_all_childs() function. 2023-07-11 - Version 3.5 This release is a maintenance release that fixes issues reported since the past six months and adds support to PostgreSQL 16. - Fix pgcluu for PG16 changes on pg_stat_user_indexes. - Fix live/dead tuples statistics. Thanks to Frederic Yhuel for the patch. - Add e as expression for kind of extended statistics. - Add --disable-pidstat option to disable the collect of metrics from the pidstat command. This can be useful with old versions of pidstat that do not support the -U option. Thanks to Guillaume Lelarge for the report. - If on a secondary server, don't execute pg_current_wal_lsn(). Thanks to Guillaume Lelarge for the patch. Use relispartition only for PG version 10 and upper. Thanks to Guillaume Lelarge for the patch. 2022-12-26 - Version 3.4 This release is a maintenance release that fixes issues reported since the past six months. - Fix unwanted report of missing index from partition. Thanks to Christophe Courtois for the report. - Fix cases where detection of redundant indexes was depending on the creation order.Thanks to Guillaume Lelarge for the report. Move Cancelled queries report from Replication sub menu to upper level. Thanks to Guillaume Lelarge for the report. - Allow -k option to make pgcluu_collectd exit gracefully. Previous behavior was to exit immediately, it now calls the HUP signal (-1). This change is done to not interrupt the running execution of the metrics collect, the process will die only if all SQL queries are executed. For an immediate interrupt use the TERM (-15) signal. - Remove all ORDER BY clause in the metrics collect queries to improve the performances. Ordering is done when reports are generated. - Fix collect of the pg_stat_user_indexes, pg_stat_user_tables and corresponding io stats. - Fix --rotate-daily spill directory. Thanks to Chistophe Courtois for the patch. - Fix potential use of uninitialized value when sar and pidstat files are not present. Thanks to geostart123 for the report. - [CGI] verify that pidstat statistics exists before trying to load them. Thanks to geostar123 foir the report. - [CGI] Fix wrong error message when attempt to read binary files and remove useless code. Thanks to geostar123 for the report. - Remove newline from indexes definition to avoid data import error when building reports. 2022-06-02 - Version 3.3 This release is a maintenance release that fixes issues reported since previous release. It also adds new reports and features. - Add TCP utilization and TCP errors reports, from sar -nTCP,ETCP. - Add a comment in the suggested index if it could be redundant with existing ones. - Add support to run pgcluu_collect as user with the pg_monitor role. Here is the full list of changes since previous release. - Add support to run pgcluu_collect as user with the pg_monitor role. Thanks to Nicolas Gollet for the report. - Fix redundant index query to work with PG < 10, regexp_match() must be replaced by regexp_matches() with some additional rewrite. - Replace call to pg_ls_dir by pg_ls_waldir to be able to use pg_monitor privilege starting from PG10. Thanks to Nicolas Gollet for the report. - Fix impossibility to use --lock-timeout. Thanks to Christophe Courtois for the report. - Remove lock_timeout from internally set of PGOPTIONS when querying pgbouncer, the old value is restored after. Thanks to Christophe Courtois for the report. - Fix empty pgbouncer statistics file when running in interactive mode. Thanks to Christophe Courtois for the report. - Add a comment in the CREATE INDEX when the suggested index may be redundant with existing ones. Thanks to Christophe Courtois for the patch. - Fix false report of redundant index when the WHERE clause is different. Thanks to Guillaume Lelarge for the report. - Fix in pgcluu for OUTPUT_DIR always be quoted, especially if their value may contain a white space or separator character. Thanks to Nicolas Gollet for the patch. - Fix network errors disabled. Thanks to Christophe Courtois for the patch. - Add TCP utilization and TCP errors reports, from sar -nTCP,ETCP. Thanks to Christophe Courtois for the patch. - Fix pgbouncer command. Thanks to Nicolas Gollet for the patch. - Modification of the query on pg_stat_statement so that under PostgreSQL 13 and higher total_time = total_plan_time + total_execution_time. Thanks to Nicolas Gollet for the patch. - Fix STATS_REPORT_RETENTION directory in config file. Don't prepend DESTDIR to paths in config file. Thanks to Christoph Berg for the patch. 2021-09-30 - Version 3.2 This release is a maintenance release that fixes issues reported since previous release. It also adds new reports and features. * Add STATS_REPORT_OUTDIR configuration directive to pgcluu.conf to set the directory where the pgCluu reports will be saved. Default to /var/lib/pgcluu/report/. * Add command line option -t | --lock-timeout with default to 3 seconds to self terminate a SQL query that could be lock by the activity of an other session like a drop table for example. * Add collect and report of unused trigger functions. * Add system statistics dedicated to PostgreSQL process through stats reported by pidstat. * Add report of CPU scaling_governor to check if it set to powersave, setting it to performances can improve performances up to 20% following the PostgreSQL load. For more information on this kernel parameter see https://wiki.archlinux.org/index.php/CPU_frequency_scaling#Scaling_governors Backward compatibility: - Change default retention time to 30 days instead of no storage limit to avoid possible out of memory on uncontrolled installation. Thanks to Jean Pierre Huart for the report. Here is the full list of changes since previous release. - Remove service time report as per sysstat v12.1.2 it has been removed. - Add STATS_REPORT_OUTDIR configuration directive to pgcluu.conf to set the directory where the pgCluu reports will be saved. Default to /var/lib/pgcluu/report/. Thanks to Jean Pierre Huart for the feature request. - Fix error "Use of uninitialized value $pidstat_file in -e at ..." when generating report for sar file only. - Fix possible bad format issue in pidstat parsing du to command with a space character. Thanks to Christophe Courtois for the report. - Subtract the lock_timeout setting when looking at pg_settings to have the right values outside pgcluu. - Fix README typo ; releases come from Github. Thanks to Christophe Courtois for the patch. - Add BEGIN_STAT/END_STAT variables dedicated to sar data on their timezone, to avoid filtering on the wrong hours. - Add command line option -t | --lock-timeout with default to 3 seconds to self terminate a SQL query that could be lock by the activity of an other session like a drop table for example. Thanks to Christophe Courtois for the feature request. - Limit output length of usage to 80 characters. - Fix for begin/end date filtering for sar data. Thanks to Christophe Courtois for the patch. - Fix failure of pidstat through ssh. Thanks to Christophe Courtois for the patch. - Verify that same statistic report is not built twice. - Remove fork for collect of PostgreSQL metrics and add autodetection of the main process pid when pid file is not found. - Remove copyright from html footer. - Split per database report in several HTML pages to avoid too large files. Thanks to Andreas Schrafl for the report. - Add collect of wait event if extension pg_wait_sampling is available in a database. This collect can be disabled with new command line option -w, --no-waitevent. - Set hamburger dropdown menu background from transparent to grey. Thansk to Andreas Schrafl for the patch. - Fix missing report of tables and indexes statistics. - Fix lookup at database list. - Add collect of unused trigger functions. Thanks to Adrien Neyrat for the feature request. - Fix load of database list from binary files in CGI mode. - Fix pidstat reports about CPU wait and iodelay with old systat version where these information was not available. iodelay < sysstat v10.2.1 and wait < sysstat v11.5.5 - Add CGI reports for pidstat statistics. - Add pidstat data caching. - Fix link to page cluster-archive.html. Thanks to cyberioio for the report. - Fix view of configurations reports that was hidden. Thanks to Lionel for the report. - Fix formatting of README file for command line option. - Add collect of PostgreSQL only system metrics through pidstat. - Fix pgcluu error on reading compressed data files. Thanks to Damien Clochard for the report. - Fix wrong detection of --output option. - Fix previous commit that was breaking GetOptions behaviors - Do not take care of the STATS_REPORT_CACHING value in config file if pgcluu is run in interactive mode with an ouptut directory (-o) and not caching enabled (-c). - Update year in copyright - Fix Storable.pm error when reading empty file. Thanks to Christian Roux for the report. - Make update of system information stat file sysinfo.txt atomic to prevent zero file size. - Fix a bgwriter URL. Thanks to Guillaume Lelarge for the patch. - Fix parsing of sar v12 header. Thanks to Christoph Berg for the report. - Fix INPUT_DIR in generated pgcluu.conf. The path here must not have DESTDIR prepended. Thanks to Christoph Berg for the patch. - Prevent look at pg_statistic_ext if major version is bellow 10.0. Thanks to Devrim Gunduz for the report. - Fix use of DESTDIR vs PREFIX installation process. Thanks to Christoph Berg for the report. 2019-10-29 - Version 3.1 released This release is a maintenance release that adds support to PostgreSQL v12.0. It also add a new report: * Add report of extended statistics in report Table Statistics per database. Example: Table | Extended Statistic public.cities | CREATE STATISTICS sch1.t1_stat (dependencies) ON city,country FROM public.cities; It also fixes some issues reported by users since last month: - Fix some typos in documentation. Thanks to Justin Pryzby for the report. - Replace obsolete pg_constraint.consrc in PostgreSQL v12 by a call to function pg_get_constraintdef(). Thanks to seadba and Devrim Gunduz for the report. 2019-09-18 - Version 3.0 released This release is a major release that adds lot of improvement and new reports. It also improve performances and fully support a CGI mode with dynamic content for realtime reports. New or enhanced reports/features: * The CGI mode to provide dynamic reports on time selection is now production ready and has left the beta stage. See documentation on how to enable it. Debian packaging enable this mode by default. * Installation and configuration is fully managed from Makefile.PL script. This will help distribution packaging with additional configuration directives to control the behavior of pgcluu scripts with generic systemd service and timer files. * Add incremental mode to pgcluu. That mean that you don't have to remove the content of the report directory anymore before running pgcluu again and when pgcluu detect that you are running in this mode. The incremental mode is detected from statistic directory when you have used the --rotate-daily or --rotate-hourly option to pgcluu_collectd. * Rewrite Stats I/O reports to display read/hit per table or indexes during the audit period reported. * Move Xlog (Wal bytes written) report under replication menu as data comes from pg_stat_replication. * Explode System and Cluster reports into several reports dedicated files to limit the size of the HTML files in incremental mode. * Add collect and report of vm.nr_overcommit_hugepages. * Add report on pages scanned to report number of pages scanned by the kswapd daemon and number of pages scanned directly per second (pgscank/s and pgscand/s). It also reports the number of pages the system has reclaimed from cache (pagecache and swapcache) per second to satisfy its memory demands (pgsteal/s). On a second axis %vmeffto reports the the efficiency of page reclaim. * All pages reports have been moved as sub menu of the Page menu and pgfree has been removed from the page fault report. * Limit system cache statistics report to pgpgin/s and pgpgout/s and a new report is dedicated to Page fault statistics with majflt/s, minflt/s and pgfree/s dataset. Dataset minflt/s is calculated from sar output as result of "fault/s - majflt/s". * All statistics about pg_stat_user_* and pg_stat_xact_user_* views are now stored as a snapshot at start and at end of the audit period and each hour if you are running in incremental mode. The old behavior was to append the snapshot at each running interval to the data file but this is not necessary and this can use huge disk space if you have lot of table and indexes in your database. Backward compatibility is preserved. * New report of json versus jsonb columns in each database. * Add keepalive kernel configuration in SysInfo report. * Separate bgwriter "allocated" buffers in a dedicated report for better reading. * Add new report about estimated memory workload based on the value of Committed_AS from /proc/meminfo. Committed_AS is the total amount of memory estimated to complete the workload. This value represents the worst case scenario value, and also includes swap memory. The report show the memory workload aver the time. New command line options: - Add option -x, --external-menu to pgcluu to save the common menu in menu.html and load it into each report using w3-include-html attribute from w3.js. This will only work if access to HTML report is through a Web server, not using the file:// protocol. - Add new option --retention to pgcluu to set number of rolling days to keep in report directory. Default is to store indefinitely. - Add option --retention NDAYS to pgcluu_collectd to set the number of rolling days to keep in data directory in incremental mode. Default is to store indefinitely. New configuration directives : * Add configuration directive STATS_REPORT_CACHING to pgcluu.conf file. This directive must be enabled in CGI mode. In this case the pgcluu script must be run periodically at least each 5 minutes to build and update cached statistics. In cache mode pgcluu generate binary file in the statistics directory and nothing in the report directory. The pgcluu.service and pgcluu.timer can be use, pgcluu will read the configuration file and switch automatically to cache mode if STATS_REPORT_CACHING is enabled. * Add MAX_RENDERED_DAYS configuration directive to set the maximum number of days in a graph. Using default pgcluu_collectd 1 minute interval each, daily graph will have 1440 points. Having too much data to render is not a good point for performances with the current code. The user can set is own graph period, default is to limit to the first seven days of the selected period. This mean a maximum of 10080 points for one week, this seems a safe limit. It also fixes some issues reported by users since last five months: - New report on memory workload based on values of Committed_AS from /proc/meminfo. Thanks to Jehan-Guillaume de Rorthais for the feature request. - Separate bgwriter "allocated" buffers in a dedicated report. Thanks to Guillaume Lelarge for the report. - Use MASK to 0022 for directory and file creation and add a note about securing the stats repository in the source code. It can be changed to 0027 if you want to secure the data directory Note that in CGI mode the www-data user or group need read access to all files in the stat directory. The systemd service files take care of setting the privilege for the postgres and www-data users. - Make perlcritic more happy. - Try to handle automatically systemd RPM specific path in install process (/lib/systemd/system/). - Update resources files minimization. - Fix declaration of config file. - Change Perl installation paths to be vendor paths by default and adapt path replacement by DESTDIR. - Keep pgcluu.conf after a make clean to be able to see the content. - Update documentation about new installation behavior. - Fix perl replacement issues on service file - Update MANIFEST and META.yml files - Update path to man page in documentation. - Replace DATADIR by STATDIR and HTMLDIR by RSCDIR in Makefile.PL. Add missing REPORTDIR variable, default /var/lib/pgcluu/report. - Change default PIDDIR to /var/run/postgresql. - Replace path in services files following environment variables defined with Makefile.PL. - The configuration file pgcluu.conf is now auto-generated by script Makefile.PL - Add support to some new environment variables to fully customize the installation paths (HTMLDIR,CGIDIR,CONFDIR,PIDDIR,DATADIR, APACHECONF,MANDIR,DOCDIR,SYSTEMDDIR,RETENTION). Default are: DESTDIR => /usr/local INSTALLDIRS => site CONFDIR => DESTDIR/etc PIDDIR => /var/run DATADIR => /var/lib/pgcluu/data HTMLDIR => DESTDIR/share/pgcluu CGIDIR => /usr/lib/cgi-bin APACHECONF => /etc/apache2/conf-available MANDIR => DESTDIR/share/man DOCDIR => DESTDIR/share/doc SYSTEMDIR => DESTDIR/lib/systemd/system RETENTION => 0 If INSTALLDIRS is set to 'vendor': CONFDIR => /etc DESTDIR => /usr The configuration file is auto-generated by the Makefile.PL script and saved into CONFDIR/pgcluu.conf. If the destination file exists it is not overridden. The file is saved as example in directory DOCDIR/pgcluu/examples/pgcluu.conf.dist The man page is saved as DESTDIR/share/man/pgcluu.1p.gz and a symbolic link pgcluu_collectd.1p.gz is created to this file. The documentation, README, changelog.gz, LICENSE files are saved under DESTDIR/share/doc/. For the CGI mode, resources (css & js files from the cgi-bin/rsc) are saved under the DESTDIR/share/pgcluu/rsc directory. The CGI script is saved under /usr/lib/cgi-bin/pgcluu.cgi. The Apache configuration file under /etc/apache2/conf-available/pgcluu.conf with a symbolic link /etc/apache2/conf-enabled/pgcluu.conf created to this file. Its content: Alias /pgcluu HTMLDIR/ Options FollowSymLinks MultiViews AllowOverride None Require local #Require ip 192.1.168.0/24 The systemd files (pgcluu_collectd.service, pgcluu.service and pgcluu.timer) are saved as examples into DOCDIR/pgcluu/examples/ and into the systemd directory SYSTEMDDIR/ The right path to the configuration file is set into all scripts pgcluu, pgcluu_collectd and pgcluu.cgi. The path where the pid file must be saved is replaced into pgcluu_collectd with the value of PIDFILE variable. - Enable configuration reading to pgcluu_collectd. The configuration directive supported by this script is STATS_COLLECTD_RETENTION. It is used to set the retention limit in the statistics directory expressed in number of days. This directive is only used by script pgcluu_collectd in incremental mode. Only the last number of days will be preserved, obsolete directories will be removed. It can be used to preserved disk space. Default store indefinitely. - Enable configuration reading to script pgcluu. The configuration directive supported by this script is STATS_REPORT_RETENTION. It is used to set the retention limit in static html report directory expressed in number of days. This directive is only used by pgcluu in incremental mode. Only last number of days will be preserved, obsolete directories will be removed. It can be used to preserved disk space. Default store indefinitely. - Fix some uninitialized variables during report generation of pg_stat_user_tables and pg_stat_user_indexes. - Set default configuration file path to /etc/pgcluu.conf and no more /usr/local/etc/pgcluu.conf - Change $CONFIG_FILE declaration in CGI script. - Update copyright year. - Limit number of days to process for a single time period controlled by configuration directive MAX_RENDERED_DAYS. - Disable Year and Month time selector for the moment, this is too much data to render with the actual implementation of pgcluu.cgi - Separate cluster / database actions. - Only display latest information in CGI mode for all database table and index reports that reports a list of statistics. These informations can not be cumulated over time selection. - Add map between actions and binary file to only load relevant files - Improve report generation by reading only required data. - Prevent reading sar stat from data file when already read from binary file. - Fix multiple call to set_device_list() and compute_sarfile_stats() - Only display latest information in CGI mode for menu Home, SysInfo and Database. - Fix offset caching for incremental mode. - Fix missing call to dump_sar_binary() when build cache from hourly rotation. - Fix missing System menu on some condition. - Submit immediately the time selector when day/week/month/year is selected. - Add more information on incremental mode to documentation. - Fix CGI Time selector behavior. - Fix calendar output for days that have already been processed. - Fix issue in CPU parsing from sar output since addition of Softnet stats in version 11.7.2. - Update minified javascript files and add file w3.js to the list of resources files. - Add resources/w3.js Javascript library to be able to include a HTML file into an other. Used to not include the menu in each HTML file from menu.html. - Explode cluster.html output into per report dedicated files to limit the size of a HTML file in incremental mode. This should be applied to other huge files (system.html and database-DB.html). - Fix some HTML tags. - Add documentation about Dynamic Reports generation using the CGI mode. Thanks to Maiken Saveljev for most part of the installation guide. - Add information about incremental mode and --retention option at end of pgcluu help. - Add example for incremental mode at end of help message. - Add a warning about storing indefinitely statistics with script pgcluu_collectd that can fill disk space in the short or medium term. Obsolete statistics can be removed using new command line option '--retention ndays' with pgcluu_collectd. - Update documentation about pgcluu_collectd --retention option. Thanks to Christoph Courtois for the feature request. - Document sorttable license. The license is called X11 in the file, but MIT on the website. Stick with MIT because that's what the other js files are using. Thanks to Christoph Berg for the patch. - Add pages reports changes to CGI. - Add report on pages scanned to report number of pages scanned by the kswapd daemon and number of pages scanned directly per second (pgscank/s and pgscand/s). Thanks to Jehan-Guillaume de Rorthais for the feature request. - Additional fixes and improvements on CGI devices detection to display system menu. - Fix unwanted "no dataset" in table stats. - Modifiy pgcluu to support new changes in pgcluu_collected about pg_stat_(user|all)_* statistics storage. This patch also remove use of global variable %all_vacuum_stats that was duplicating storage with %all_stat_user_tables that already contain vacumm and analyze statistics. - Add description of %vmeff to reports. - Fix missing wait of compression child process resulting in zombi process. Thanks to Jehan Guillaume de Rorthais for the report. - Suppress which command error in some distribution. Thanks to Michael Vitale for the report. 2018-12-13 - Version 2.9 released This release is a minor release that fix a major issue in the interval detection. This This led to a bad scale in bg_writter, replication and most database statistics reports. Everybody must upgrade immediately. It also fixes some issues reported by users since last release: - Improve devices list retrieving and fix several menu issues. - Fix checkpoint counter legend. - Move query in pg_stat_statements at the right of the table for better readability. - Fix report of disk use, mounted devices, running processes, crontab and installed package to avoid duplicate entries. - Empty SysInfo arrays before reading sysinfo file, this prevent duplicate data. - Improve parsing of sar file. - Fix interval that was set in milliseconds instead of second. This conduct to wrong scale (ex: Kb instead of Gb) in bg_writter, replication and database statistics reports. Thanks to Guillaume Lelarge for the report. 2018-11-28 - Version 2.8 released This release is a minor release that fix some issues reported by users over past months, it also adds some new reports: - Add disk space utilization report over time using df command for sysstat version older than 11.1.4. - Add report of percentage of disk space and inode used on each file system. The information is available in sar output since 11.1.4 version. Thanks to Guillaume Lelarge for the feature report. - Add PageTables information in SysInfo memory report. - Show database information about table/indexes in capture mode. - Add report of number of tasks currently blocked, waiting for I/O to complete "Run queue" report. - Add new report about system dirty memory that need to be written to disk as well as amount of active/inactive memory. - Show highest dirty memory to write and highest number of processes blocked in overall stats page. - Add report of crontab entries for the user running pgcluu. - Add report of installed PostgreSQL packages in main page. New pgcluu_collectd command line option: - Add command line option --package-list to be able to set a custom command to list PostgreSQL packages. Default is to auto-detect package type between rpm and dpkg, using command 'rpm -qa' or 'dpkg -l'. If you have an other system you can use this option to set a custom command. A filter on keyword 'postgres' is appended to the command: ' | grep postgres'. Here are the complete list of bug fixes in this release: - Change the way sysstat version is checked and save the version into to sysinfo.txt file for use in pgcluu script. - Remove extra new line at end of data file generated by the patch on search_path securing. - Add collect of disk space and inode used over time stored in file fs_stat_use.csv. Need more work to avoid duplicate data with new versions of sar that already report fs space use. - Replace all call to Perl ternary operator with usual if-then-else statement. In some condition this operator do not works and makes pgcluu return negative values in overall stats. This was happening only when pgcluu encounter a stat reset in the data files. Thanks to Guillaume Lelarge for the report. - Remove /dev/loop from devices statistics report. - Add PageTables information in SysInfo memoryb report. Thanks to Adrien Neyrat for the patch. - Remove some unwanted metrics displayed in capture mode. - Show database information about table/indexes in capture mode. - Exclude /dev/loop from result of "df" command and squashfs from "mount -l" command. - Prevent pgcluu_collectd to stop before that all metrics collected from the current loop have been written to disk. When it receives the terminate signal (-k option) it previously stops immediately, the problem was that this can conduct to empty or incomplete data file. Now it waits until the loop is terminated before exiting. - Fix compression of hourly directory when -R | --rotate-hourly and -z | --compress options are enabled. The compression was called each loop and not at each hour rotation. - Add warning about mandatory use of -d option when pg_buffercache reports are enabled (-B) but pg_buffercache extention is not found in the connection database. - Fix missing buffercache statistics since secure path has been included in commit b61bf44. This required that pg_buffercache table call was prefixed by the public schema. Thanks to lobojohnson for the report. - Fix unwanted print of tablespace path in cluster reports. - Remove not useful information in buffercache graph legends. - CSS: Increase size of the sysinfo panel. - Adapt automatically pgCluu to changes in sar sysstat tool since version 11.5.7. As per sysstat commit 8d635e0: Replace "rd_sec/s" and "wr_sec/s" fields with "rkB/s" and "wkB/s". These fields are now expressed in kilobytes instead of sectors. Replace "avgrq-sz" field with "areq-sz". This field is now expressed in kilobytes instead of sectors. Rename "avgqu-sz" field to "aqu-sz" to make it consistent with iostat's output. - Add report of number of tasks currently blocked, waiting for I/O to complete to "Run queue" report. This information is available with sysstat >= 9.1.7 - Show also highest dirty memory to write and highest number of processes blocked in main page. - Add new report about system dirty memory that need to be written to disk as well as amount of active/inactive memory. These metrics are available in recent sar/sysstat versions. - Add report of crontab entries. - Change icon for packages report. - Add report of PostgreSQL installed package in main page. - Add storage information about PostgreSQL packages installed on the system. - Add command line option --package-list to be able to set a custom command to list PostgreSQL packages. Default is to auto-detect package type between rpm and dpkg, using command 'rpm -qa' or 'dpkg -l'. If you have an other system you can use this option to set a custom command. A filter on keyword 'postgres' is appended to the command: ' | grep postgres'. 2018-07-09 - Version 2.7 released This release is a minor release that fix some issues reported by users over past year, it also adds some new reports: - Add reports of pgbouncer wait for server statistics. - Make pgCluu more compatible with PG10 and pgBouncer 1.8 - Add kernel vm.overcommit_kbytes configuration to report. Here are the complete list of bug fixes in this release: - Fix Prepared Transaction menu not disabled with no stat data. - Fix typos and use proper measure in the graph. Thanks to Fabio Porta for the patch - Add vm.overcommit_kbytes to kernel report. Thanks to Adrien Neyrat for the patch. - Since version 1.8 of pgbouncer, the command "show stats" changed output. Make pgcluu compatible. Thanks to Fabio Pardi for the report and the mapping of changes. - Fix issue in partitionning lookup. - Secure search_path before executing SQL queries. - Fix several use of unintialized variable case. - Add missing function IsLeapYear. Thanks to Emmanuel Boucle for the patch. - Fix report for network and disk devices in incremental mode. - Fix unwanted Network menu when no network device are found. - Fetch devices informations. Thanks to Adrien Neyrat for the patch. - Dockerise pgcluu, the container helps run both the collection and the report generator. Thanks to Roy Golan for the patch. - Disable System submenu when the corresponding report is not available. - Append %idle to cpu report. - Fix list of resources files. Thanks to Bosstek Consulting for the report. - Change fetch_version() to be closer with the new versioning politic in 10.0 and above. Thanks to Julien Rouhaud for the patch. - Makes pgcluu compatible with PostgreSQL 10.0. 2017-07-09 - Version 2.6 released This release is a minor release that fix some issues reported by users over past year but it also adds some new interesting reports: * Add report on prepared transaction and oldest one in seconde per database. * Detect partitions and summarize information in a dedicated report. * Add kernel scheduler configuration for sched_autogroup_enabled and sched_migration_cost_ns to sysinfo report. * Add report of configuration files changes in incremental mode. * Report on cancelled queries due to conflicts is now a time based graph instead of a pie chart. * bgwritter buffer clean, checkpoint and backend statistics are now reported as bytes using size of 8192 per buffer. * Add report of allocated buffers with bgwritter buffer statistics. * Add report of transaction throughput per second. * Show data checksum status. * Add report of unlogged tables. Database with unlogged tables will be listed in cluster view. * Add hourly index when --rotate-hourly is enabled. and useful features: * Compatibility with PostgreSQL 10.0 * Add static index on main directory with incremental report to link to the different days. * Use bootstrap modal dialog windows to download graph as png. * Autodetect interval between collected data to support interval change during stats collect. * Replace javascript call to dateToDisplay.toGMTString() with dateToDisplay.toString(). Please note that this could not be backward compatible with your previous timezone settings. See pgcluu.js to revert the function call. * Create DDL of missing index concurrently. Here are the complete list of bug fixes in this release: - Finalize systemd unit files. pgcluu_collectd.service is used to start pgcluu_collectd in daemon mode. Other files, pgcluu.service and pgcluu.timer to execute pgcluu periodically to generate reports. - Add import of missing jqplot.canvasAxisLabelRenderer.min.js file used to render axis labels in graphs. - Set logo and icon on a single line in an url attribute as chrome complain that it will be obsolete in M60. - Add partitioning report and partition information into cache. - Add hourly index when --rotate-hourly is enabled. - Add support of partition information to cache mode. - Fix unwanted exit in cache mode - Cosmetic change in tooltip and download button. Main menu font size have also been improved. - Update generated html to use Bootstrap 3 glyphicons. - Add javascript and CSS sources, licences and download information for packaging. Add a tool to minified and embedded the script and css into pgcluu Perl script and copy the minified files into cgi-bin/rsc/ - Add minified resources file for CGI into cgi-bin/rsc - Fix redundant index query. Thanks to Julien Rouhaud for the patch. - Detect partitions and summarize information in a dedicated report. Thanks to Julien Rouhaud for the patch. - Refactor pidfile unlink handling. Thanks to Julien Rouhaud for the patch. - Limit call to pg_relation_size() when we are in capture mode. Thanks to Guillaume Lelarge for the report. - Add information about recheck of redundant index with primary key and index on a column referencing a foreign key. - Fix exclusion of UNIQUE index in redudant indexes report. - Fix incremental global index on resize. - Update year in copyright. - Add index on main directory with incremental report to link to the different days. Thanks to Heath Yob for the feature request. - Add verification that pg_stat_statement is loaded from shared_preload_libraries. - Fix incomplete per-database-statistics. Thanks to Markus Braunig for the report. - Add kernel scheduler configuration for sched_autogroup_enabled and sched_migration_cost_ns to sysinfo report. Thanks to Adrien Nayrat for the patch. - Fix broken main menu when no disk devices was present in sar report. - Allow to set label for y2axis in create_linegraph() parameters - Change parameters and return of the get_diff() method to support incremental mode. - Create function get_diff() and shows_diff() to report configuration file change. - Append new color for line graph. - Set missing database list with sysinfo and about menu. - Fix Not a HASH reference in function reporting information about indexes. - Rename diff storage variables and stores them in binary file. Initialyze storage variables for statistics that must be read from file each time. - Fix detection of working directories with dates over two months. - Fix runqueue size report. Thanks to Thomas Reiss for the report. - Create function to reuse the look for sysinfo file. - pgcluu CGI now append new csv statistic to cached binary files. - The menu is now built at end to avoid reading csv files first to look for database, disk device and network interface. - The CGI home page is now just built from the binary files to speed up the first screen. As we do not append last data collected from the csv files we indicate at which time the cache was last built. This is to give a quick snapshot of the dashboard. When looking at the statistics binary and csv files are both read. If the cache have not been run yet, pgcluu will read dashboard statistics from data fileis which takes longer to display. - Allow incremental cache. When running in cache mode pgcluu will append new csv statistics to old cached binary files. Then when building reports, pgcluu will first load statistics from binary files and complete statistics with data from csv file collected after the last cache mode run. This mean that if cache mode is run each minute on an incremental statistics collect (see option -r or -R of pgcluu_collectd), pgcluu is able to create report very quickly at any time. This is especially useful with CGI mode where a full day statistics report can be displayed in few seconds. - Fix diff of configuration files, old and new files was inversed. Thanks to Adrien Nayrat for the report. - Fix display of message no dataset on empty graphs. - Update description of temporary files report and statistics on checkpoints. - Rewrite use of interval between collected data to support change during collecting. Interval is now always calculated from the difference between the current line end the previous line. This mean that the first line is never present in reports with per second statistics. - Fix formatting of bytes in pretty number. - Improve storage of database list with huge number of database. - Fix timezone on all reports and remove graph per tablespace and replace them with a list of tablespace and their location. - Fix timezone on start and end timestamp of collect. Remove graph per tablespace and replace them with a list of tablespace and their location. - Clear Start/End input box at startup in CGI. - Comment and include REVERT_DATE toogle about sar date format in CGI configuration file. - Fix timezone and time selector in javascript menu. - Fix use of timezone and date detection in sa file. - Fix date parsing from sa text file. - Fix some warning about use of undefined variables. - Add link to last known statistics in CGI front page when no data are found. - Fix a call to timegm_nocheck() - Split set_overall_stat_from_binary() in two function, one for database orverall statistics and the other for system statistics. - Fix parsing of sa file. - Do not load statistic from the last day when hour:min is 00:00. - Fix some bugs when there is just sar cached binary files. - Fix two digit in date parts on start/end date. - Fix cache mode when there is just sar file in entry. - Add INCLUDE_DEV configuration directive to be able to filter disk device. Allow comma as list separator in configuration file. - Add information about how to parse sysstat sa binary and text file. - Fix remove of unlogged tables following pg version. - Fix colspan on unlogged table report. - Add report of transaction throughput per second. - Use psql -X option so .psqlrc doesn't get in the way. If .psqlrc contains for example \timing, pgcluu_collectd will get confused. Thanks to Christoph Berg for the patch. - Separate svctm/await dataset on two different axes. - Remove unwanted bgwriter stats prior pg 9.1 - Add data checksum status to CGI report. - Fix autodetection of timezone and set default timezone for sar statistics to the result of perl -MPOSIX -e "print substr(strftime(\"%z\", localtime()), 0, 3);". - Replace javascript call to dateToDisplay.toGMTString() with dateToDisplay.toString(). Please note that this could not be backward compatible with your previous timezone settings. See pgcluu.js to revert the function call. - Fix an issue with --from-sa-file where date was still incremented. Thanks to Flavie Perette for the report. - Replace sadf -D option by -d to obtain a compatible output. - Separate svctm/await dataset on two different axes to be able to see both dataset. - Fix kernel parameter hugepage report when not available. - Fix all call of psql that use both -f - and -c that doe not have the same behavior in pg 9.6. Previous version does not care of the -f stdin input when a command was provided with -c. This not compatible anymore. Thanks to Vincent Laborie for the report. - Fix collect of lock in pg_stat_activity with PostgreSQL 9.6. Thanks to Vincent Laborie for the report. 2016-04-27 - Version 2.5 released This release is a major release that fix some issues reported by users over past year and a full replacement of the flotr2 javascript chart library with jqplot. There's also some new interesting reports: * Add report for hash indexes. * Keep track of pg_settings and database/roles settings changes and show diff in the the reports. * Add report of pending restart in pg settings view. * Add information about percentage of timed against requested checkpoints. * Split database menu in submenu per set of 10 databases. * Add report of tables without indexes and tables with more than five indexes. * Add report about invalid index after concurrency build. * Add report of system and PostgreSQL uptime. * Add report of last statistics reset per database and report of last autovacuum and autoanalyze. * Add report of bgwriter last statistics reset in Home/Cluster view. * Add a non default configuration settings report. and useful features: * Autodetect timezone from csv data files and automatic adjustment of chart axis. * Add systemd start script. * Add collect of crontab information. * Add support to daylight saving. * Add compatibility with PostgreSQL 9.5 * Add ablity to export plots data as csv * Set legend table outside the graph. * Display specific titles and description for overall graphs. Note that a CGI script have been added to be able to perform temporal lookup in incremental pgCluu statistics, with predefined year, month, day and hour views. This is a work in progress, it will be available in next major release. Upgrade: you can safely override previous installation, backward compatibility with 2.4 version is preserved. Here are the complete list of bug fixes in this release: - Add report for hash indexes. Thanks to Adrien Neyrat for the patch. - Fix detection of non official release like EnterpriseDB or devel. Thanks to Piotr Gbyliczek for the patch. - Add use of IO::File, some system was complaining about that. Thanks to microtodd for the report. - Only collect stored procedures count instead of name list. - Refresh sysinfo.txt at each loop and order output per database name. - Fix capture mode with PostgreSQL version 8.4 - Fix some bug in capture report. - Fix use of timezone in cluster/database related charts. - Autodetect timezone from csv data files. - Fix some documentation typo about systemd use and apply change to pgcluu.pod. - Fix systemd unit files. Thanks to gabx for the patch. - Update README with systemd unit files usage - Add collect of crontab information. It uses the current logged username or postgres as default and it can be overriden with the --cron-user command line. - Do not collect metrics if previous time is upper than current time This is to prevent storing statistics twice when hour change during daylight saving. Thanks to brownbh3 for the report. - Keep track of pg_settings and database/roles settings changes and show diff in the the reports. - Update comment on pg_settings columns. - Add report of pending restart in pg settings view. - Exclude from non default settings collect case where setting equal boot_val. - Add collect of pending_restart column from pg_settings (pg 9.5). - Remove obsolete css file, it is replaced by font-awesome.min.css - Keep track of configuration files changes by storing diff to files Changes are sorted by date at bottom of the configuration reports. - Fix range of icons in embedded fontawesome css. - Fix report of number of xlog data written per second. Thanks to Ronan Dunklau for the report. - Add ablity to export plots data as csv. Thanks to Julien Rouhaud for the patch. - Fix/change some fontawesome icons. - Fix menu height in pgcluu css. - Add ablity to export plots data as csv - Fix scale on yaxis sticks in javascript graphs. - Report percentage between checkpoint timed and requested in legend - Add information about percentage of timed against requested checkpoints. - Update with resources directory move. - Move resources directory into cgi-bin/ to be less confusing. - Update list of file in MANIFEST for pgcluu package. - Fix top menu height and color in start/end date input box of CGI. - Update html header with new .min.css and .min.js files - Use minimalist version of CSS and Javascript and remove call to sourceMappingURL - Replace external font file fontawesome-webfont.ttf by the base64 embeded version. - Removed by their optimised min files. - Fix unparsed section in sysinfo.txt when file is generated from a Windows server. - When system information are missing, display a message. - Remove Deadlocks end Temporary files menu when PostgreSQL version id lower that 9.2. - Fix y2axis on hit ratio and commit vs rollback reports. Thanks to Guillaume Lelarge for the report. - Fix report of huge page information. - Fix an additional series 4 in background writer clean stats report Thanks to Guillaume Lelarge for the report. - Disable mouse click on disabled menu. - Do not display empty deadlocks and temporary files informations if PostgreSQL version is lower than 9.2. Thanks to Guillaume Lelarge for the report. - Prevent display of SysInfo menu when there is no such information. Thanks to Guillaume Lelarge for the report. - Prevent collect of local system information when option -h is used but --enable-ssh is disabled. Thanks to Guillaume Lelarge for the report. - Set daily rotation instead of hourly. - Add pgcluu.service file to start/stop pgcluu_collectd with systemd Thank to Arnaud Gaboury fot the file. - Fix call to current_setting(checkpoint_segments) in dump_xlog_stat It has been replaced by max_wal_size in PostgreSQL 9.5. - Renamed option --notablespace by --no-tablespace. - Split database menu in submenu per set of 10 databases. - Remove border and shadow from all graphs. - Set legend table outside the graph. - Rewrite call to pgcluu_slide.js - Reduce font size in legend and tooltip chart. - Forgot to include uptime with PostgreSQL version in main page. - Fix homepage statistics when data are loaded from cache file. - Add subtitle in homepage panels. - Fix number of connection in home page. - Fix wrong count of database and tuples returned in overall stats. - Fix report of bgwriter stats_reset. - Make some changes in SysInfo view. - Force default date before looking to working dir. Only look for incremental repositiory with CGI. - Replace flotr2 javascript chart library with jqplot. - Add time navigation in pgcluu CGI - Add a configuration file to CGI - Update resources/ directory with new javascript libraries - Remove temporary debug line. - Add margin to panel buttons and center the graph. Thanks to Tomer Steinfeld for the patches. - Fix report of overcommit_memory. - Add configuration file for CGI initialization. - Fix two bugs on indexes report - Add report about invalid index after concurrency build - Fix report on most indexed tables - Add report of tables without indexes and tables with more than 5 indexes. Thanks to julien Rouhaud for the feature request. - Add statistics collect about number of indexes by table. Stats are stored in pg_stat_count_indexes.csv - Add postmaster start time. Thanks to Julie Rouhaud for the feature request. - Add system uptime information. Thanks to Julien Rouhaud for the feature request. - Add report of last statistics reset per database and report of last autovacuum and autoanalyze. Thanks to Julien Rouhaud for the feature request. - Add report of bgwriter last statistics reset in Home/Cluster view Thanks to Julien Rouhaud for the feature request. - Add collect of invalid indexes. - Fix pg 8.4 compatibility. Thanks to Julien Rouhaud for the patch. - Check indisprimary instead of name starting by "pk", and also exclude index supporting an exlucsion constraint. Thanks to Julien Rouhaud for the patch. - Fix colspan for pg_settings reports. Thanks to Julien Rouhaud for the patch. - Add a non default configuration settings report. Thanks to Julien Rouhaud for the patch. - Fix missing global title/description for connections and remove call to normalize_line on index report. Thanks to Julien Rouhault for the report. - Display specific titles and description for overall graphs. Thanks to Julien Rouhaud for the patch. - Fix wrong call to pg_current_xlog_location() on secondary host. Thanks to Julien Rouhaud for the report. - Fix error when pgcluu is just used to parse a sar output file. Thanks to Euler Taveira de Oliveira for the report. - Fix wrong calculation of pg_stat_replication lag reports. - Fix error: "Invalid offset: 0" when parsing pg_stat_replication statistics. Thanks to vscherbo for the report. - Fix regexp in previous commit to be more strict. - Fix creation of wrong database menu using function name. Thanks to spritchard for the report. - Fix psql error when using --list-metric option. Thanks to Fabio Pardi for the report. - Fix wrong storage of previous value in pg_stat_user_functions which result in counts on the database-functions page completely wrong. Thanks to Steve Pritchard for the patch. - Remove redundant regular expressions. - Update CSS and JS files in resources directory and datetimepicker ressources files for cgi. 2015-07-25 - Version 2.4 released This release is a maintenance release that fix some issues. There's also some new interesting reports: * Transfers per second (read/write/both) on all devices from sar -b * Transfers per second for each device from sar -d . * Number of tasks created per second * Number of context switches per seconds. * Improve pg_stat_statement report by adding all shared block stats and read/write I/O timing per query when track_io_timing is enabled. * Add device with highest tps on overall system information. and useful features: * Add --capture mode to pgcluu_collectd to be be able to build a snapshot of the PostgreSQL instance and exit. pgCluu will automatically adapt the report to this capture mode. It will use a temporary directory /tmp/pgcluu_capture to generate a tarball /tmp/pgcluu_capture.tar.gz containing the capture. * Add pgCluu logo and ico to the html output. * Add --charset option to be able to change the html charset, default: utf8. * Allow regular expression in database list available in reports, for example: with --db-only "p.*", only database beginning with p will be reported. * Allow pgcluu to parse and compute statistics from gzip compressed files. * pgcluu will not stop anymore if the sar file is not found, it will only show a warning message and continue. This release also adds -r | --rotate-daily and -R | --rotate-hourly options to pgcluu_collectd be able to rotate statistic files on a daily or hourly basis. You can use -z or --compress option to compress rotated data files. Thanks to Euler Taveira de Oliveira for the feature request. There also some code relative to next coming major release that will be used to allow a full incremental mode and a temporal navigation into the collected statistics from a CGI program. This code enable caching (option -C or --cache) by dumping statistics stored in memory into binary files. With those files, data files can be removed (automatically with option -c or --clean) and report can be build later from them. This is not really useful now but this s the first stage to build incremental and cumulative reports. Here are the complete list of bug fixes in this release: - Fix pretty print number format when units are blocks and not sizes. - Do not apply database list restriction on total cluster size calculation. - Fix Statistics about I/O on Indexes. - Rewrite some parts of the overall system statistics for better performances. - Fix bug in overall stat for Most read/written device, they was multiplied by 512 (size of a block) twice. - Fix system report of r/w and tps per devices. - Fix wrong report of pg_buffercache statistics. - Remove some global variables with local redeclaration. - Fix use of uninitialized value during build of pg_stat_statement report. - Fix Statements statistics not available because a local var is used during computation. Thanks to Michel Meyer for the patch. - Fix uninitialized value and wrong overall reports in Home menu. Thanks to Assem Bayahi for the report. - Remove garbage from last commit. - Fix replication lag statistics report and add new report about "Number of xlog data written per second." - Fix documentation about --included-db - Fix pgcluu_collectd crash when --no-database is used and the psql command is not available. Thanks to Ronan Dunklau for the report. - Fix call to cluster canceled queries statistics. - Fix menu when no database are found. - Fix start/end date and path to sar file in incremental and capture mode. - Separate load of statistics from report builder. - Make paths internally relative into output directory especially for incremental facilities. - Fix several issue that prevented configuration files to be copied into the output directory. - Allow white space in db's names. Thanks to Nicolas Thauvin for the report. - Fix wrongly disabled temporary files menu. - Fix database list with character - inside the name - Fix replication and checkpoint report and corresponding menus disabled. Thanks to Zsolt for the report. - Fix a print on undefined value when a device is mounted after the first execution of pgcluu. Thanks to Ezequiel Mina for the report. - Add timezone option for perl localtime() and fix mixed formated tabs. Thanks to David Cramblett for the patch. - Add missing information into usage about -w | --password option. - Fix for sar command when ssh not in use. Thanks to David Cramblett for the patch. - Make error message during first connection more explicit. Thanks to Gregoire Pineau for the report. - Be sure that interval is always != 0 to avoid illegal division by zero. Thanks to Gregoire Pineau for the report. - Fix control character in substitution regex. - Fix error message for $OUTPUT_DIR not being empty should not use $INPUT_DIR. Thanks to Matthew Musgrove for the patch. - Fix limit reports to the database when --only-db is used. Thanks to Bianca Santana Espichicoquez for the report. 2015-02-06 - Version 2.3 released This release is a maintenance release that fix some issues. There's also a new report about "Role Settings" per database. - Report default parameters values set with ALTER DATABASE and ALTER ROLE in new menu item: "Database/Role Settings". Thanks to Thomas Reiss for the feature request. - Fix detection of disk device in sar file. - Add boot value to the settings report - Add Unit and Reset value in pg_settings report to highlight parameters where values have been changes outside the configuration file. - Fix handling of sysinfo information, and enhance .gitignore. Thanks to Julien Rouhaud for the patch. - Fix kernel.* and transparent_hugepage display on Sysinfo section. Thanks to Julien Rouhaud for the patch. 2015-01-05 - Version 2.2 released This release is a maintenance release that fix several issues. There's also a simple report of transparent_hugepage from the system and a menu enhancement by dividing the device menu per 10 items which is helpful when there's plenty of disk devices. - Increase copyright year to 2015. - Fix bug in database list extraction. - Fix query to get missing foreign key indexes. Thanks to Ronan Dunklau for the patch. - Add collect of transparent_hugepage information into sysinfo.txt file. (read from /sys/kernel/mm/transparent_hugepage/-) - Fix overwriting of idle_in_xact for all database, should be idle. Thanks to Guillaume Lelarge for the patch. - Fix query for missing indexes on FKs. The previous query reported every index that SHOULD exist for FKs, but was not filtering out the already existing ones. Thanks to Ronan Dunklau for the patch. - Do not report redundant indexes when one is partial and not the other one. Thanks to Ronan Dunklau for the report. - Fix statement to search redundant index by not reporting index that has uniqueness and not the other one. The statement now also reports duplicate indexes on the same column and not only composite indexes. Thanks to Ronan Dunklau for the report. - Divide Devices menu by part of 10 devices. Useful when there is plenty of disk devices. - Add --from-sa-file to allow parsing of sar output coming from a sa daily file. - Fix call to local sar command and limit test on application_name for pg version >= 9.0. Thanks to Julien Rouhaud for the report. - Set application_name to pgcluu before collecting data and exclude from pg_stat_activity queries generated by pgcluu. 2014-09-25 - Version 2.1 released This release fix a lot of issue reported since six month and adds some improvments: Allow system information commands to be executed remotely Allow sar to be executed on a remote serveur using ssh connection. New options to pgcluu_collectd: -M or --max-size option to allow an output dir size limit. --no-database to avoid collecting statistics from a database. -C count to terminate program after collecting data for X times. -E or --end-after to terminate program after for some time. -V or --version to show version information. New or enhanced reports/features: Add report of connections waiting for a lock. Add average duration time in statement report. Add pg_stat_statements report. Add pg_default and pg_global to tablespace size report. Other changes/fixes: - Allow system information commands to be executed remotely. Thanks to Ahmad Iftekhar Rumman for the report. - Add --no-database to avoid collecting statistics from a database. Useful if you just want to collect sar statistics. - Fix query to collect information from pg_stat_activity that fail on postgres-xl and certainly postgres-xc. Thanks to xiao huan for the report. - Fix pgcluu_collectd to stop capturing activity when option -C is omitted. Thanks to Thomas Reiss. - Fix pg_settings query unit is available in >= 8.2 and boot and reset since 8.4. Thanks to Euler Taveira de Oliveira for the patch. - Fix unopen filehandle in printing system disk device statistics. Thanks to Ahmad Iftekhar Rumman for the report. - Cosmetic fixes in docs. Thanks to Euler Taveira de Oliveira for the patch. - Documentation update with the new -C option to pgcluu_collectd. - Terminate program after collecting data for X times. This is another option to end the program. Instead of doing some math with time counter, use a numeric counter. It can be combined with -i option too. Also, time (-E) has precedence over numeric (-C). Thanks to Euler Taveira de Oliveira for the patch. - Add report of kernel nr_hugepages report, 9.4+ now support huge page supported. Thanks to Euler Taveira de Oliveira for the patch. - Fix connection report enabled when a database has no connection. Thanks to Guillaume Lelarge for the report. - Add configuration report of recovery.conf and postgresql.auto.conf. Thanks to Guillaume Lelarge and Julien Rouhaud for the report. - Removed useless Type column in Indexes ans Tables size reports. Thanks to Guillaume Lelarge for the report. - Disable Conflict and Canceled queries menus when this is not a standby server. Thanks to Guillaume Lelarge for the report. - Fix some typo reported in issue #20. Thanks to Guillaume Lelarge for the report. - Change Returned and Fetched legend to Table (returned) and Index (fetched). Thanks to Guillaume Lelarge for the report. - Fix keep tick formatting when zooming. Thanks to Julien Rouhaud for the patch. - Fix non exclusion of devices in System Network and Devices submenu. Thanks to Thomas Reiss for the report. - Fix some error uninitialized value in concatenation. - Allow pgbouncer statistics collect+report only. Thanks to Eric Veldhuyzen for the report. - Add boot_val and reset_val to pg_settings output. - Fix multiline in df system information. Thanks to Nicolas Thauvin for the report. - Add -M or --max-size option to allow an output dir size limit. For example using -M 100MB will interrupt pgcluu_collectd when output dir will reach this size. Require the du system command. Thanks to Nicolas Thauvin for the feature request. - Add new option -E or --end-after to tell pgcluu_collectd to run for some time before terminating automatically. Thanks to Nicolas Thauvin for the feature resquest. - Fix issue with pg_stat_statements metrics when the extension was not installed in a schema searchable in the search_path connected user value. Thanks to Nicolas Thauvin for the report. - Failure on null value on replication report when pg_basebackup was running. Thanks to Emmanuel Vinet for the report. - Add/fix report of connections waiting for a lock. Thanks to Guillaume Lelarge for the report. - Replace label Waiting and Waiting in xact by Idle and Idle in xact in connection graph. Thanks to Guillaume Lelarge for the report. - Fix metrics that was not accepted if passed alone: pg_settings, lock_granted,lock_modes,lock_types. Thanks to Fabio Pardi for the report. - Add information about -v|--verbose and -V|--version to documentation. - Add -V | --version option. Thanks to Fabio Pardi for the feature request. - Fix several issues in disabling menu in cluster reports. - Fix temporary statistics report, menu was always disabled. Thanks to Julien Rouhaud for the report. - Fix wring number of CPU sockets that was reporting the total number of cores. Thanks to Thomas Reiss for the report. - Add collect of the extensions version. Thanks to Jehan-Guillaume de Rorthais for the request. - Add support to individual reports when the --metrics option have been used with pgcluu_collectd. - Fix some uninitialized variables and errors when pgcluu has only some metrics to parse. - It will not search for pg_database_size.csv anymore to contruct the list of database and read pgbouncer file to get list of database with pgbouncer. - Generation of database related reports (cluster, database, pgbouncer) have been review to use non global filehandle. It will not search for pg_database_size.csv anymore to contruct the list of database and prevent case where some reports was genereted twice. - Speed improvement. - Rename --os-info option into --sysinfo and add --no-sysinfo option to completly disable the generation of sysinfo.txt - Fix --metrics option that was not limiting the kind of metrics to use. Thanks to Fabio Pardi for the report. - Fix bug in device lookup in sar file that was generating undefined filehandle. Thanks to Toth Csaba for the help - Fix several part where pgcluu crash using undefined variable if the database in not found in pg_database_size.csv. - Fix error print on an undefined value when a database does not have lock defined. Thanks to Csaba Toth for the report. - Fix typo in System memory utilization report description. - Fix typo in Commits/Rollbacks report description. - Fix typo in Read Tuples report description. - Fix typo in locks per type report description. - Fix typo in Statistics about cache utilisation per relation. - Change description of Replication lag report. - Add sorttable.js to resources directory. - Fix position of index.html#home. - Fix parsing of --db-only, --dev-only ... comma separated list. - Fix typo in ssh related option. Thanks to Jehan-Guillaume de Rorthais for the report. - Fix code to set ssh command and allow collect of configuration files through ssh. - Allow sar to be executed on a remote serveur using ssh connection. Add --enable-ssh to activate this feature, the remote host is defined using -h option. There is several new options --ssh-* to be able to take full control of the ssh connection. - Add average duration time in statement report. - Add pg_stat_statements report. - Override statements statistics on each loop. - Replace multiple space by single one in pg_stat_statements queries. - Fix pg_stat_statements collect to replace semi-colon by SEMICOLON keyword to prevent breaking CSV format. - Add pg_default and pg_global to tablespace size report. Fix issue on spclocation display on per tablespace reports. 2014-03-22 - Version 2.0 released This major release adds more than twenty new reports. pgCluu now reports everything you want to know about your PostgreSQL server from a cluster, database, or operating system point of view. It is also able to collect statistics about pgBouncer, shared buffer utilization using pg_buffercache, most frequent and time consuming queries with pg_stat_statements and really much more, see full list bellow. New or enhanced reports/features: - Allow pgcluu_collected to collect sar and configuration files remotely together with PostgreSQL statistics using a ssh connection. That mean that pgCluu is fully able to audit a remote server. - Add pg_stat_statements report showing most frequent and time consuming queries. - Add reports for unused and redundant indexes per database. - Add report about archiver statistics with PostgreSQL 9.4 new pg_stat_archiver view. - Add report about wal files created vs recycled and the max number of wal files. - Add reports about network interfaces utilization and network errors. - Add reports about system cache page in/out from/to disk. - Add report of isdirty statistics. - Add --exclude-time to pgcluu_collectd options to stop collecting statistics during a period of time. Ex: --exclude-time "22:00-06:00" to stop collecting data from 22:00 to 06:00 am the next day. - Add report about percentage of shared buffer used per database - Add report about percentage of each databases loaded in shared buffer - Add report about usagecount distribution in shared buffer - Add report about usagecount distribution in dirty shared buffer - Add report of number of shared buffer/pages used by a relation - Add report number of buffers loaded in cache for a relation relation and the percentage of the relation loaded (1). - Add reports about configuration files (postgresql.conf, pg_hba.conf and pg_ident.conf), pgcluu_collectd must be run as postgresql user. - Add report of user functions statistics. - Add report about the pgbouncer settings in Configuration submenu. - Add copy of pgbouncer.ini file in Configuration submenu. - Add report of pg_settings in Configuration submenu.. - Add report with SQL orders to create missing indexes on foreign keys. - Allow HTML table sort with a modified version of the javascript library sorttable.js - Moved System info report from sar sytem menu to a dedicated main SysInfo menu. - Add the number and list of extensions used inside the PG cluster. - Add reports about Locks per types, Locks per modes and Locks granted or not. - Add report about I/O statistics per index. - Add report about I/O statistics per table. - Add last manual vacuum/analyze datatime on database info report. - Add per database index statistics. - Add report of size and number of tuples per table of a database. - Add statistics reports per table (idx_scan/seq_scan, vacuums/analyzes, Insert/update/delete/hot update and live/dead tuples). - Add count of user triggers in database information slide. - Add Database info submenu per database to display the general information about a database (installed extension, schemas, number of Stored procedures and of all kind of object. - Add last know size of the database in the Database info slide. - Add total number of databases on cluster key values. - Add report of tablespaces utilization. - Add report of pgbouncer statistics per dbname and per pool (dbname/username). (1) [Shared buffers statistics are collected automatically when pg_buffercache extension is installed and the -B | --enable-buffercache option is used.] Other changes/fixes: - Allow sar to be executed on a remote server using ssh connection. Add --enable-ssh to activate this feature, the remote host is defined using -h option. There is several new options --ssh-* to be able to take full control of the ssh connection. - Waiting interval is now calculated with the interval value minus the execution time of each loop. - Add output information when a kill signal is sent to pgcluu_collectd. - Option -p | --dev-only have change to -D | --device-only and add new option -N | --network-only to limit report on some network interfaces. - Pretty print system memory information. - Fix graphs when a new database is created or dropped during an audit. - Add -B | --enable-buffercache to activate pg_buffercache statistics. It is disable by default. - Reorder Database menu with submenu and prevent moving to non existent reports by disabling the submenu. - Reorder cluster menu, using submenu and fix title of replication lag report. - Add vm.dirty_background_bytes and vm.dirty_bytes to sysctl collect. - Add support to PostgreSQL 8.1 by using temporary table not COPY (sql). Thanks to kenstir for the patch. - Reopen stdin/out/err so that daemon mode works correctly. Thanks to kenstir for the patch. - Fix affected tuples per operation per database where tuples fetched was used as select instead of tuples returned. - Review code and use one method per report. - Update documentation and usage about --exclude-time - Fix detection of non official release, like devel or EnterpriseDB major version detection. - Change icon of CPU in SysInfo report. - Remove full path from sar and psql command to allow definition by environment variables. - Fix CPU report refresh when comming from a report in the same system.html page. - Change label of the 'Reset' button to 'To chart' to be less confusing. This button allow to switch from image to chart (opposite as 'To image') and not to reset a zoom in the graph. - Add -n, --top-number command line option to redefine the default top 10 tables or indexes I/O statistics. - Fix interval given to the sar command, should be one second and not the value of $INTERVAL. - Fix sar report with overlapped time. - Fix empty file pg_stat_user_tables.csv. - Add database name inside pg_class_size.csv output and add function get_proc_count(dbname) to retrieve function from a given database. - Add -T | --notablespace option to avoid printing an error message when the connection user is not superuser. - Remove pg_default and pg_global from tablespace report. - Change link on program name to pgcluu www site. - Move print of Postgresql full version at bottom of the home page. - Reduce font size of key value on home and information slides. - Update Flotr2 javascript library to fix redraw of crosshair. - Command line options that can be used multiple time can now also be used one time using a comma separated list of object names. 2014-01-28 - Version 1.1 released This release adds lot of report improvements and bug fixes. There is also several new features or reports. - Format mouse tracker on graphs to show all dataset values at a time. - Add run queue length report to system menu. - Add checkpoint write and sync times reports. - Add report of PostgreSQL version - Split background writer buffer and count statistics into separated reports. - Add report of maxwritten_clean into bgwriter reports. - Add report of kernel parameters to the system info page. - Add collect of system kernel tuning parameters. - Remove embedded CSS and javascript on each HTML page, resources are now automatically copied into the output directory if not already present. Thanks to Guillaume Lelarge for the suggestion. - Allow pgcluu to parse sar file generated from sa file, use commands like "sar -A -p -f /var/log/sysstat/sa*". Thanks to Julien Rouhaud for the feature request. - Split commit, rollback and backend graph by using a second yaxis for backend. Thanks to GUillaume Lelarge for the report. - Add System Information report. - Move Cache hit/miss ratio on second yaxis and change dataset colors. Thanks to Guillaume Lelarge for the report. - Add collect of OS release information. - Allow pgcluu_collectd to grab OS information (cpu, memory, etc.) and add --os-info option to only grab that information (for testing). - Reformat dashboard information. - Add -z | --timezone to set the hour(s) from GMT time to adjust times on sar report. Thanks to Bricklen for the feature request. and some more changes/fixes: - Add vertical crosshair on graph. - Update copyright date to 2014 - Disable database report of number of canceled queries when not on hot standby node. - Disable checkpoint write report following the pg version (< 9.2). - Disable temporary files and deadlocks reports following the postgresql version (< 9.2). - Add storage of pg version into sysinfo.txt - Fix missing legend of checkpoints_timed in checkpoint report. - Fix warning on META_MERGE for ExtUtils::MakeMaker < 6.46. Thanks to Julien Rouhaud for the patch. - Fix typo in pgcluu_collectd calls. Thanks to Jacky Rigoreau for the report. - Fix issue where information slide was not displayed when clicking on the information button. - Change documentation about resources files that are now autogenerated. - Fix grab of statistics from an 8.4 cluster (access to not-available-yet statistics catalogs). Thanks to Guillaume Lelarge for the report. - Fix issue on parsing CentOs release. Thanks to bricklen for the help. - Fix issue "Use of uninitialized value $val in substitution line 3312". Thanks to bricklen for the report. - Fix call method "print" on an undefined value at ./pgcluu line 1303. Thanks to Guillaume Lelarge for the report. - Fix sar dashboard report. - Remove decimal from hit cache ratio report. Thanks to Guillaume Lelarge for the report. - Dashboard review: cluster label rewrite, remove empty values from report, add start/end date of database stats and sar stats. Thanks to Guillaume Lelarge for the feature/change requests. - Fix some warning on uninitialized value on dashboard. Thanks to Julien Rouhaud for the report. - Fix Illegal division by zero at ./pgcluu line 1132. Thanks to Julien Rouhaud for the report. - Fix issue in building timestamps in sar data. Thanks to Bricklen for the report. - Fix sysstat version execution error with locale different to C or en_*. Thanks to forall for the report. - Fix broken sar charts when collect time is greater than 24 hours. - Add -z | --timezone option to documentation. - Fix error: print() on closed filehandle GEN9 at pgcluu line 1942. Thanks to Bricklen for the report. 2013-11-18 - Version 1.0 released This is the first public release of pgCluu, that is a packaging of the tools I use every day to collect statistics and build reports of PostgreSQL Clusters for performances auditing and troubleshooting. At this time it collect and report most of what is helpful for a PostgreSQL Cluster performance auditing. There's lot of others reports to be included: - Statistics reports concerning tables. - Statistics reports about pg_stat_statement. - More Sar statistiques reports. - ... This will comes in next release. The goal of this project is to provide a complete PostgreSQL auditing tool that do not need any dependency so that it can be run on any server. If you just have a sar output file, pgCluu can be use to draw graphs about the system utilization only. For more information take a look at http://pgcluu.darold.net/