| | h1. LPAR CPU statistics and documentation tool for [IBM Power systems|http://www-03.ibm.com/systems/power/] servers |
| | |
| | The tool is capable to produce *historical CPU utilization graphs of LPAR's and shared CPU usage*. |
| | It also collects *complete physical (HW) and logical configuration* of all managed systems/LPAR's and all changes in their state and configuration. |
| | The rough idea comes from Ondrej Plachy. The tool itself is being developed by [Pavel Hampl|~pave1], IBM Czech Republic. |
| | There is not and will not be any license restriction. Just use it if you like it. Any hints or comments are welcome. |
| | {center} |
| | !lpar2rrd-1.jpg! !CPU pool.jpg! !CPU pool - yearly.jpg! |
| | {center} |
| | for more examples go to the [demo site|http://lpar2rrd.sourceforge.net/demo/] with 2.32 version (15-Sep-2008). It is a snapshot of a live production environment with 6 HMCs, 43 servers (p525 - p595) and 154 LPAR's (Thanks to Anoop Sood for providing the data). Unfortunately there are not all the latest features like lpars aggregation and historical graphs. |
| | |
| | *Content* |
| | {toc:minLevel=1} |
| | |
| | h2. Features |
| | |
| | * intended only for HMC based *micro-partitioned systems with shared CPU pool* |
| | * creates charts based on utilization data collected on HMC's (lslparutil) |
| | * *it is agent-less*, no agents need to be installed on LPAR's |
| | * it uses ssh-keys based access to HMC servers to get all the data, so it does not cause *ANY load* on monitored LPAR's |
| | * it supports all types of LPAR's and OS'es : *[AIX|http://www.ibm.com/servers/aix] , [VIOS|http://www14.software.ibm.com/webapp/set2/sas/f/vios/documentation/home.html] , [Linux on Power|http://www-03.ibm.com/systems/power/software/linux/index.html], i5/OS* on *IBM Power Systems* (formerly [pSeries|http://www-03.ibm.com/systems/p/index.html]/[iSeries|http://www-03.ibm.com/systems/i/index.html]) |
| | * it creates automatically a menu based [WWW front-end|^front-end.jpg] for viewing charts, configuration and logs |
| | * it creates physical (HW) and logical [configuration inventory|^config.html] of all managed systems and their lpars (once a day) |
| | * it shows [100 last changes in configuration|^changes.html] and [100 last changes in the state|^state.html] of all managed systems and their lpars |
| | * it shows total [memory usage|^memory.png] for each managed systems |
| | * simple to install, configure and use (initial install & configuration together with supporting tools like Apache/Perl/SSH should not take more than half an hour\!) |
| | * no any additional management when any *change/adding/remove/rename* of LPAR or managed system (the tool discovers everything automatically for all configured HMC's) |
| | * default graphs are created 1 year back if historical utilization data on HMC's is available (note the HMC keeps 60sec sample rate data for 2 days, hourly data for last 2 months and daily data for last 2 years if utilization data collection is enabled) |
| | * for viewing older data than 1 year you have to use [Historical reports|^lpar2rrd-hist1.jpg] if the data is already present in LPAR2RRD |
| | * initially the tool loads all historical data already collected on the HMC, after that the tool loads only new data every hour (depends on cron schedule), saves it in RRDTool databases and re-draws the graphs |
| | * it can be hosted on any \*NIX (theoretically even WIN) platform, it just needs a web server, SSH, Perl and RRDTool installed. (check prerequisites below) |
| | * by default it creates 4 kind of graphs for each LPAR, CPU shared pool and memory usage. First 3 ([last day|^day.png], [week|^week.png] and [month|^month.png]) are based on *60secs* averages, [yearly|^year.png] chart is based on daily averages, all provided by the HMC |
| | |
| | h2. How it works |
| | !architecture2.gif! |
| | |
| |  | h2. {color:red} Mailing list is available {color} |
| | | h2. Mailing list is available |
| | * To post on the mailing list, simply send an email to *lpar2rrd@freelists.org*. |
| | * You can subscribe to the list by sending an email to lpar2rrd-request@freelists.org with 'subscribe' in the Subject field or by visiting the list page at http://www.freelists.org/list/lpar2rrd . The web-based administration suite allows you to add yourself manually. |
| | * Online, searchable archives of your list are available at http://www.freelists.org/archive/lpar2rrd |
| | |
| | h2. Major news in 2.XX |
 | | * 2.44 : added LPAR CPU aggregated graphs (*all LPARs CPU usage in one graph*) ([example|^aggregte.png]) |
| | | * 2.49 : added LPAR search option ([example|^search.htm]) |
| | * 2.49 : added export to CSV in historical graphs ([example|^pool.xls]) |
| | * 2.44 : added LPAR CPU aggregated graphs (*all LPARs CPU usage in one graph*) ([example|^aggregated-p595.png] [example1|^aggregte.png]) |
| | * 2.44 : added possibility to change resolution of "Historical report" graphs in the web |
| | * 2.40 : is available detailed *historical reporting* from all data stored LPAR2RRD ([example-pre|^lpar2rrd-hist1.jpg], [example-post|^lpar2rrd-hist2.jpg]) |
| | * 2.40 : is available log of last 100 changes in state for each managed system ([example|^state.html]) |
| | * 2.40 : is available log of last 100 changes in configuration for each managed system ([example|^changes.html]) |
| | * 2.30 : it shows *CPU donation of partitions with dedicated CPUs* in "allowed processor sharing" mode |
| | It works only for LPARs with following configuration: |
| | |Processing mode|Dedicated| |
| | |Processor Sharing|Allow when partition is active| |
| | |Compatibility mode|POWER6| |
| | * 2.22 : the tool now collects complete *physical and logical info* about managed systems (once a day) and shows that on the web |
| | * 2.01 : added support for *more POWER6 CPU shared pools* ([example|^shared CPU pools.jpg]) |
| | * 2.00 : support for *all Utilization data collection sample rates* (HMC 7.3.3\+ only), actually it is tested and I recommend *60 seconds* (there is no reason to use another once) |
| | |
| | h2. What it is not |
| | |
| | * it is not intended to be real-time monitoring tool, check the FAQ section for more |
| | * it is not a tool for partitions using dedicated CPUs, it always shows 100% CPU utilization for them (unfortunately HMC does not provide detailed CPU stats usage for LPARs with dedicated CPUs) |
| | Well, for POWER6 and "allowed processor sharing" mode LPAR2RRD shows CPU donation of those partitions to the CPU shared pool |
| | * it is not a complex tool with many features, options etc. It a simple tool which does *mainly CPU utilization monitoring on micro-partitioned systems*. |
| | * it does not have perfect web front-end, no nice menus ... I prefer functionality and simplicity. Anyway I welcome anyone who would be able to write professional HTML/Java front-end for the tool. |
| | |
| | h2. Future |
| | |
| | * Support *ASM* (Active Memory Sharing) if there is relevant data on the HMC, will check it as soon as I upgrade my env to the required level |
| | * Support of IVM based systems, actually only HMC based systems are supported (will include them as soon as I get access to any running IVM based managed system) |
| | * Will include global *HEA/IVE network throughput statistics* per managed system/LPAR (HMC provides such data, so it should not be a problem ... ) |
| | * Will re-write report of physical and logical configuration from text based formatting into pure HTML |
| | * Will create an option for "dumping" row data from RRDTool DB's withing "Historical reporting" in txt/xls/cvs format, for easy importing and analyzing it in an other tools |
| | * Create "fast start" bundle which would include everything what LPAR2RRD needs for its run on different platforms (could be useful either for people doing performance tuning on the customer side or for people which just want to test LPAR2RRD) |
| | * Add a "find" option for finding a lpar within the GUI |
| | * Add license info into if it is available in the HMC |
| | * Will update colors in aggregate graphs in the next version to make first a few first lpars better recognizable |
| | * Any other ideas??? |
| | |
| | h2. FAQ |
| | |
| | * what borrowed CPU means (yellow part in CPU pool utilization graphs)? |
| | -* borrowed processing units are processing units not being assigned to any partition (based on my findings) |
| | -* borrowed processing units are being borrowed from powered off partitions with dedicated processors (official description) |
| | -* Basically, if you have 8 CPUs box and 2 LPARs each with 3 CPUs assigned (no matter if capped, dedicated or uncapped ), then 2 CPUs are left and they are considered as borrowed. |
| | If you switch off one LPAR (it must have dedicated CPUs) then there appears 5 borrowed CPUs. |
| | * why there are displayed 3 different CPU pools when I use only 1 additional CPU pool except the default one? |
| | -* Default pool is SharedPool0. SharedPool1 is your additional CPU pool. "Total in CPU pools" means SharedPool0 & SharedPool1 together. It always shows total utilization per all CPUs in all shared pools. It is at least my understanding of the data which lslparutil on HMC provides |
| | * How much disk lpar2rrd (or maybe I should say lslparutil) would use on HMC? |
| | -* lpar2rrd itself does not use any disk space on the HMC, it just uses lslparutil and coule of other standart HMC cmds, and stores everything on the hosted server. |
| | * How about load on the HMC servers? |
| | -* Load on HMC servers is minimal as LPAR2RRD tool once an hour runs lslparutil HMC command which provides the data from the past hour. The data is already collected on the HMC itself and it only sends it out to LPAR2RRD hosted server. In fact lslparutil reads a few binary files, re-formats the data into text fields and LPAR2RRD send it out via ssh, it is very simple task for the OS of the HMC itself. |
| | * Would HMC still be supported by IBM if the you decided to use the LPAR2RRD? |
| | -* If anything goes wrong with lslparutil (it happens time to time, check bellow for examples ...) and you have the HMC under support then you can ask IBM for fixing it. But you cannot ask for support of LPAR2RRD itself. |
| | -* HMC is still supported even with using LPAR2RRD as it uses only commands provided by the HMC itself (which are obviously supported), there is nothing new installed on HMC's except the SSH key for access without the password prompt. |
| | * Can somehow force the graphs to be updated with the most latest information? |
| | -* You can anytime run ./load.sh from the command line under lpar2rrd user identification to get the actual data in the charts. |
| | -* Basically LPAR2RRD is not intended to be *real-time monitoring* tool. To make it more *real-time* I would suggest to schedule it run more frequently from cron. First of all check how long LPAR2RRD usually runs (it depends on number of HMC's, managed systems, LPAR's, hosting HW, CPU available ... I saw 30mins in a huge environment like on the demo site, but it might even take up to couple of minutes in a simple env). So if it takes for example 10mins (check the time stamp of load.sh when LPAR2RRD finishes "$ ps -ef |grep lpar2rrd") in your environment, then you might schedule it every 15mins from cron to be on the safe side. Just assure that there is not and will not be running 2 or more LPAR2RRD instances concurrently. It would not harm anything except of CPU and other resources on the hosting server. |
| | |
| | h2. Reference |
| | * [Presentation : VIOS "best practices"|http://www.ibm.com/developerworks/wikis/download/attachments/53871900/VIOS_bestpractices_VUG_062608_1.1.pdf] |
| | * [Redbook : SAP Applications on a Virtualized IBM Power Systems Environment|http://www.redbooks.ibm.com/redpieces/pdfs/sg247564.pdf] |
| | * [Presentation : Topas Recording Features|http://www.ibm.com/developerworks/wikis/download/attachments/53871900/VUG_Sept252008_TopasRecording.pdf] |
| | * [IBM Wiki : Other similar free Performance tools for AIX|http://www.ibm.com/developerworks/wikis/display/WikiPtype/Performance+Other+Tools] |
| | * [IBM Wiki : Virtual I/O Server Performance Monitoring Tools|http://www.ibm.com/developerworks/wikis/display/WikiPtype/VIOS_Monitoring] |
| | |
| | h1. Download |
| | |
| | * [LPAR2RRD tool download | https://sourceforge.net/project/showfiles.php?group_id=207954] from !http://sflogo.sourceforge.net/sflogo.php?group_id=207954&type=2! , current version : *2.46* (it is platform independent) |
| | | * [LPAR2RRD tool download | https://sourceforge.net/project/showfiles.php?group_id=207954] from !http://sflogo.sourceforge.net/sflogo.php?group_id=207954&type=2! , current version : *2.49* (it is platform independent) |
| | * AIX tested packages : (tested on AIX 5.3, but generally it should work even on AIX 5.1, 5.2 and 6.1) |
| | ||Package||IBM Wiki download|| |
| | |[RRDTool | http://www.inet.hr/zmp/ibm/rrdtool/]|[^rrdtool-1.2.13-1.perl58.aix5.2.ppc.rpm]| |
| | |[Perl | http://www-03.ibm.com/systems/p/os/aix/linux/toolbox/download.html]|[^perl-5.8.2-1.aix5.1.ppc.rpm]| |
| | |[TimeDate| http://search.cpan.org/~gbarr/TimeDate-1.16]|[^TimeDate-1.16.tar.gz]| |
| | |[Apache| http://www-03.ibm.com/servers/aix/products/aixos/linux/download.html]|[^apache-1.3.31-1.aix5.1.ppc.rpm]| |
| | |[Expact| http://www-03.ibm.com/servers/aix/products/aixos/linux/download.html]|[^expat-1.95.7-4.aix5.1.ppc.rpm]| |
| | |[Freetype2 | http://www-03.ibm.com/servers/aix/products/aixos/linux/download.html]|[^freetype2-2.1.7-5.aix5.1.ppc.rpm]| |
| | |[Libart_lgpl | http://www-03.ibm.com/servers/aix/products/aixos/linux/download.html]|[^libart_lgpl-2.3.17-4.aix5.1.ppc.rpm]| |
| | |[Libpng| http://www-03.ibm.com/servers/aix/products/aixos/linux/download.html]|[^libpng-1.2.8-6.aix5.2.ppc.rpm]| |
| | |[Zlib| http://www-03.ibm.com/servers/aix/products/aixos/linux/download.html]|[^zlib-1.2.3-4.aix5.2.ppc.rpm]| |
| | * Linux (PPC or x86): just install Apache, Perl with Time-Date module as is described below and RRDTool (apart of Time-Date module all should be on the installation medias) |
| | Do not use RRDtool 1.2.12 on Linux. Use at least 1.2.27, [here|http://www.express.org/~wrl/rrdtool/] are for example compiled RRDtool packages for Linux (do not forget for rrdtool-perl package) |
| | |
| | * Windows : install Perl, Cygwin shell, ISS (Apache) and RRDTool (if it is available for Windows). Basically it is not tested so I do not recommend it. Anyway it should work if anyone wants to play with it .... |
| | |
| | h1. Installation |
| | |
| | |
| | h2. Prerequisites |
| | |
| | * LPAR2RRD can be hosted on any Unix or even Windows server, so far it has been tested on AIX 5.3/5.2, SuSE Linux ES 9 SP3 for PPC, RedHad and Ubuntu on Intel |
| | * it is not necessary to have a dedicated box for it, good idea is use some management server like NIM server |
| | * disk space requirements : each monitored LPAR needs roughly 2MB of the disk space, if you use dual HMC setup then 4MB |
| | * Perl : arbitrary version (tested with 5.8.0, 5.8.2), it just requires [TimeDate| http://search.cpan.org/~gbarr/TimeDate-1.16] module (installation procedure is below) |
| | * Apache or any other Web server |
| | * SSH : OpenSSH or any other commercial SSH (Tested with OpenSSH & F-Secure SSH) |
| | * RRDTool, tested with 1.0.33, 1.0.42 and 1.2.15 , generally should work with all versions |
| | * HMC >= V5R2.1 (it must support : Utilization data collection) |
| | * p5/i5 & p6/i6 servers (actually it is called Power systems) need to be at least at SF240_201 firmware level, since then is supported utilization data collection. (all POWER6 are ok) |
| | Check this on a running lpar for POWER5 systems |
| | {noformat} |
| | # lsmcode |
| | The current permanent system firmware image is SF240_320 |
| | The current temporary system firmware image is SF240_320 |
| | {noformat} |
| | * there is a known bug in HMC 7.3.3 without any service pack (newer should be ok) after upgrade from HMC 6.1 and lower, utilization data collection does not work, check the problem section below for a workaround. |
| | * Similar problem causing patch MH01102, details are below as well. Ignore it if you have already HMC 7.3.3 SP1+ |
| | |
| | h2. Installation procedure |
| | |
| | * check prerequisites at first, especially HMC version and server firmware version (for POWER5) |
| | * create preferably a dedicated user (lpar2rrd) on a server which will own the tool and will have access to hmc servers via ssh-keys |
| | * add a user on Unix under root account |
| | {noformat} |
| | # useradd -m lpar2rrd |
| | {noformat} |
| | * allow utilization data collection (either via HMC GUI or globally via following cmd as hscroot) |
| | {noformat} |
| | # ssh hmc1 -l hscroot |
| | hscroot@hmc1:~> chlparutil -r config -s 60 |
| | hscroot@hmc1:~> lslparutil -r config -F name,sample_rate |
| | PWR6A-9117-MMA-SN103A4B0,60 |
| | PWR6B-9117-MMA-SN103B5C0,60 |
| | {noformat} |
| | use 3600 for HMC older than 7.3.3 |
| | * create an HMC account with restricted rights *hmcviewer* (I strongly recommend it) on a HMC server and allow ssh-keys access from a LPAR2RRD hosted server |
| | login to HMC as hscroot |
| | {noformat} |
| | # ssh -l hscroot hmc1 |
| | $ mkhmcusr -u lpar2rrd -a hmcviewer --passwd abc1234 |
| | {noformat} |
| | * create the user ssh-key on a LPAR2RRD hosted server for lpar2rrd user: |
| | {noformat} |
| | $ su - lpar2rrd |
| | $ ssh-keygen -t dsa |
| | Generating public/private dsa key pair. |
| | Enter file in which to save the key (/home/lpar2rrd/.ssh/id_dsa): |
| | Enter passphrase (empty for no passphrase): |
| | Enter same passphrase again: |
| | Your identification has been saved in /home/lpar2rrd/.ssh/id_dsa. |
| | Your public key has been saved in /home/lpar2rrd/.ssh/id_dsa.pub. |
| | The key fingerprint is: |
| | c0:05:85:ff:9c:90:39:da:d7:48:33:35:4f:cf:1a:21 lpar2rrd@p550-2 |
| | $ cat /home/lpar2rrd/.ssh/id_dsa.pub |
| | ssh-dss AAAAB3NzaC1kc3MAAACBALvbupPLnqy6UfJjgMG5SRnnFDfD013OtBxFv8P7qoHfGKCG0Vu2IhNCYttpYMmsMR+BWADo4c9oT |
| | 7r92raLfVDjNW8uO5C5fatK305+sNqazbT91HDDNDKQnbpoKUqybVRC2BQbPR8ESh+ws3uHdxXSQKwOSyrHO6Nwkkx8/h4TAAAAFQDZgT |
| | 9MxrEUQg4uOwxhciwwdoOO8QAAAIAH37x06Ia4FiCdlk3U9vuOI0QEvCKMXL+ZPFNoRcgiqiCnR2WeiaG5qM+odYWk/F/owV0Y/Dmkmrr |
| | jPrgDADjS1uKoSy+NMg803+4cS1B06EMtia+RmMbWkr+kQZ90WvB8C5cxofzcllnQjztQxVrZnZPmMI73/SQdS7QBU9a9WwAAAIA3VAdM |
| | vP+ZO3zs57FaznMySoszxgNWxuINlcP61TQuyuCp5Bq3mFXwxFwJWPpqpbW5yHj9N1+3pscSsFxEQb9YVvb1oL9c5QdPDnmXZK8BHHmNZ |
| | Mn+ftuzJKHgR2lynaxQ4hxTnz+xT0ywGwokeuuCnr4G3Bftc70Yn9vCNjuzGQ== lpar2rrd@p550-2 |
| | {noformat} |
| | * paste the pub key just created to mkauthkeys command on HMC (as hscroot, all in one line, without carriage returns): |
| | {noformat} |
| | hscroot@hmc1:~> mkauthkeys -u lpar2rrd -a 'ssh-dss AAAAB3NzaC1kc3MAAACBALvbupPLnqy6UfJjgMG5SRnnFDfD013OtB |
| | xFv8P7qoHfGKCG0Vu2IhNCYttpYMmsMR+BWADo4c9oT7r92raLfVDjNW8uO5C5fatK305+sNqazbT91HDDNDKQnbpoKUqybVRC2BQbPR8 |
| | ESh+ws3uHdxXSQKwOSyrHO6Nwkkx8/h4TAAAAFQDZgT9MxrEUQg4uOwxhciwwdoOO8QAAAIAH37x06Ia4FiCdlk3U9vuOI0QEvCKMXL+Z |
| | PFNoRcgiqiCnR2WeiaG5qM+odYWk/F/owV0Y/DmkmrrjPrgDADjS1uKoSy+NMg803+4cS1B06EMtia+RmMbWkr+kQZ90WvB8C5cxofzcl |
| | lnQjztQxVrZnZPmMI73/SQdS7QBU9a9WwAAAIA3VAdMvP+ZO3zs57FaznMySoszxgNWxuINlcP61TQuyuCp5Bq3mFXwxFwJWPpqpbW5yH |
| | j9N1+3pscSsFxEQb9YVvb1oL9c5QdPDnmXZK8BHHmNZ{{Mn+ftuzJKHgR2lynaxQ4hxTnz+xT0ywGwokeuuCnr4G3Bftc70Yn9vCNjuzG |
| | Q== lpar2rrd@p550-2' |
| | {noformat} |
| | * test the connection between the LPAR2RRD hosted server and HMC (should not request password any more) as lpar2rrd user: |
| | {noformat} |
| | |
| | $ ssh -l lpar2rrd hmc1 |
| | The authenticity of host 'hmc1 (9.138.236.97)' can't be established. |
| | RSA key fingerprint is 97:e1:a5:c2:58:3c:c6:0c:d1:6f:4d:6e:62:9c:68:4b. |
| | |
| | Are you sure you want to continue connecting (yes/no)? yes |
| | Warning: Permanently added 'hmc1,9.138.236.97' (RSA) to the list of known hosts. |
| | hmc1:~> |
| | {noformat} |
| | * when you use a commercial SSH distribution (like F-Secure ssh) which supports only SSH2 format (above example was for OpenSSH format), then you need to convert public keys onto OpenSSH format before uploading the keys into HMC. Conversion SSH2 to OpenSSH format and after that use it in mkauthkeys: |
| | {noformat} |
| | $ ssh-keygen -i -f your-ssh2.pub > your-openssh.pub |
| | {noformat} |
| | * install RRDTool prerequisites : freetype2, libart_lgpl, libpng, zlib as root |
| | {noformat} |
| | # rpm -ivh freetype2-2.1.7-5.aix5.1.ppc.rpm libart_lgpl-2.3.17-4.aix5.1.ppc.rpm libpng-1.2.8-5.aix5.1.ppc.rpm zlib-1.2.3-3.aix5.1.ppc.rpm |
| | | # rpm -ivh freetype2-2.1.7-5.aix5.1.ppc.rpm libart_lgpl-2.3.17-4.aix5.1.ppc.rpm libpng-1.2.8-6.aix5.1.ppc.rpm zlib-1.2.3-4.aix5.1.ppc.rpm |
| | {noformat} |
| | * install RRDTool as root |
| | {noformat} |
| | # rpm -ivh rrdtool-1.2.13-1.perl58.aix5.2.ppc.rpm |
| | {noformat} |
| | \-> make sure RRDTool is compiled with Perl support and the version of Perl is the right one |
| | if not and LPAR2RRD complains that cannot find RRDp.pm then adjust PERL5LIB parameter in lpar2rrd.cfg after installation |
| | (ignore it now, just do it in case of an error) |
| | {noformat} |
| | # ls -l /opt/freeware/lib/perl/5.8.0 |
| | total 16 |
| | -r--r--r-- 1 root system 5048 May 04 2006 RRDp.pm |
| | drwxr-xr-x 3 root system 256 Jan 23 2007 aix-thread-multi |
| | {noformat} |
| | when RRD.pm is not there then you will need to change default PERL5LIB setting in lpar2rrd.cfg afterwards to point to the right directory |
| | * install Perl as root |
| | {noformat} |
| | # rpm -ivh perl-5.8.2-1.aix5.1.ppc.rpm |
| | {noformat} |
| | * install TimeDate Perl module as root |
| | test whether is already installed or not: |
| | {noformat} |
| | # perl -MDate::Parse -le 'print (str2time("12/24/2006"))' |
| | 1166914800 |
| | {noformat} |
| | \-> means it is installed, if there is no output or an error then install it as per below |
| | {noformat} |
| | # gzip -d TimeDate-1.16.tar.gz; tar xvf TimeDate-1.16.tar; cd TimeDate-1.16 |
| | # perl ./Makefile.PL |
| | # make test |
| | # make install |
| | {noformat} |
| | * install and configure Apache or any other web server as root |
| | {noformat} |
| | # rpm -ivh expat-1.95.7-4.aix5.1.ppc.rpm apache-1.3.31-1.aix5.1.ppc.rpm |
| | # vi /etc/opt/freeware/apache/httpd.conf |
| | {noformat} |
| | -* it might be in different locations like /opt/freeware/etc/httpd/httpd.conf, /etc/httpd/httpd.conf, /etc/httpd/conf/httpd.conf ...) |
| | -* append at the end following: |
| | {noformat} |
| | # NB: The "Alias" line below must NOT be used if you have |
| | # the LPAR2RRD webfiles as the root URL. In that case, |
| | # you should instead set this: |
| | # |
| | # DocumentRoot /home/lpar2rrd/lpar2rrd/www/ |
| | |
| | Alias /lpar2rrd/ "/home/lpar2rrd/lpar2rrd/www/" |
| | <Directory "/home/lpar2rrd/lpar2rrd/www/"> |
| | Options Indexes FollowSymLinks Includes MultiViews |
| | Order allow,deny |
| | Allow from all |
| | </Directory> |
| | |
| | # CGI-BIN |
| | ScriptAlias /lpar2rrd-cgi/ "/home/lpar2rrd/lpar2rrd/lpar2rrd-cgi/" |
| | <Directory "/home/lpar2rrd/lpar2rrd/lpar2rrd-cgi"> |
| | AllowOverride None |
| | Options ExecCGI Includes |
| | Order allow,deny |
| | Allow from all |
| | </Directory> |
| | {noformat} |
| | {noformat} |
| | # /opt/freeware/apache/sbin/apachectl start |
| | {noformat} |
| | If you append there exactly this, then you will find LPAR2RRD here : httpd://your_web_server/lpar2rrd/ |
| | Do not forget to assure that Apache starts after the OS reboot (there must be a starting script in /etc/rc.... ) |
| | * assure that the Web server user ("nobody" in following example) has read and execution rights for LPAR2RRD CGI-BIN and the data directory in /home/lpar2rrd/lpar2rrd |
| | {noformat} |
| | $ ps -ef|grep apache|head -2 |
| | nobody 233678 348372 0 Dec 30 - 0:00 /opt/freeware/apache/sbin/httpd |
| | nobody 319670 348372 0 Dec 30 - 0:00 /opt/freeware/apache/sbin/httpd |
| | $ ls -ld /home/lpar2rrd/lpar2rrd/lpar2rrd-cgi /home/lpar2rrd/lpar2rrd /home/lpar2rrd/lpar2rrd/* |
| | drwxr-xr-x 2 lpar2rrd staff 256 Dec 30 11:13 /home/lpar2rrd/lpar2rrd/lpar2rrd-cgi |
| | ... |
| | {noformat} |
| | * install LPAR2RRD |
| | {noformat} |
| | # su - lpar2rrd |
| | $ uncompress lpar2rrd-dist-2.XX.tar.Z (also gunzip can uncompress it) |
| | $ tar xvf lpar2rrd-dist-2.XX.tar |
| | $ mv dist lpar2rrd |
| | $ cd lpar2rrd |
| | {noformat} |
| | * configure LPAR2RRD parameters in lpar2rrd.cfg (all what needs to be done is described inside the file, it is just about a few params), here is the list of parameter which need to be reviewed: |
| | {noformat} |
| | $ vi /home/lpar2rrd/lpar2rrd/lpar2rrd.cfg |
| | WEBDIR=/home/lpar2rrd/lpar2rrd/www |
| | HMC_USER=lpar2rrd |
| | HMC_LIST="hmc1 hmc2" |
| | PERL=/usr/bin/perl |
| | RRDTOOL=/opt/freeware/bin/rrdtool |
| | {noformat} |
| | * for LPAR2RRD 2.XX and HMC 7.3.3+ you might check sample rates of Utilization data collection on managed systems (must not be 0!) |
| | {noformat} |
| | $ ./sample_rate.sh |
| | Going to check HMC as user lpar2rrd, will ask for a password if there is not allowed automatic access |
| | hmc1:PWR6A-9117-MMA-SN103A4B0 sample rate : 60, OK |
| | hmc1:PWR6B-9117-MMA-SN103B5C0 sample rate : 60, OK |
| | {noformat} |
| | * now, you can start the tool itself (note that the first run might take a while especially if you had already allowed utilization data collection for a long time) |
| | {noformat} |
| | $ ./load.sh |
| | {noformat} |
| | * check errors in error.log if any |
| | * check graphs via web (point the web browser on the page which was chosen as the target one http://your_web_server/lpar2rrd/) |
| | * schedule to run LPAR2RRD from lpar2rrd crontab (once an hour seems to be reasonable time, do not use less as LPAR2RD might run 1/2 an hour if you have a big environment) |
| | {noformat} |
| | $ crontab -e |
| | 0 * * * * /home/lpar2rrd/lpar2rrd/load.sh > /home/lpar2rrd/lpar2rrd/load.out 2>&1 |
| | {noformat} |
| | You might need to add lpar2rrd user into /var/adm/cron/cron.allow if crontab command fails as root user. |
| | {noformat} |
| | # echo "lpar2rrd" >> /var/adm/cron/cron.allow |
| | {noformat} |
| | |
| | h1. Knows issues |
| | * there is a problem with a comma within the lpar name, it will be fixed in the next release 2.47 |
| | * there is a problem on Linux servers which hosts LPAR2RRD with aggregation graphs (VRULE works strange in RRDTOOL), will try to fix it in 2.47 |
| | |
| | h1. Upgrade |
| | * first of all check the release note section if make sense to upgrade the tool |
| | * unfortunately you will lose daily/weekly/monthly data if you upgrade from LPAR2RRD 1.XX and use less than 1h sample rate on the HMC, but yearly data will be kept, so you will not loose long term trends. |
| | * stop the tool (assure it is not running and will not be started from cron in next couple of minutes) |
| | * install new version via update.sh script |
| | {noformat} |
| | # su - lpar2rrd (if the tool is running under lpar2rrd user) |
| | $ cd /home/lpar2rrd |
| | $ uncompress lpar2rrd-dist-2.XX.tar.Z (also gunzip can uncompress it) |
| | $ tar xvf lpar2rrd-dist-2.XX.tar |
| | $ cd dist |
| | $ ./update.sh |
| | $ cd ../lpar2rrd |
| | {noformat} |
| | * optionaly delete all LPAR2RRD web content, all will be re-created automatically, do not worry :) |
| | {noformat} |
| | $ rm -rf /_your_web_home_/lpar2rrd/* |
| | {noformat} |
| | * remove memory RRD DB as there was changed format (just in case of 1.XX to 2.XX upgrade) |
| | {noformat} |
| | $ find data -name "mem*rr*" -exec rm {} \; |
| | {noformat} |
| | * update your web server configuration to add CGI-BIN section only if you upgrade from LPAR2RRD version < 2.40. Attach following to httpd.conf for Apache web server : |
| | {noformat} |
| | ScriptAlias /lpar2rrd-cgi/ "/home/lpar2rrd/lpar2rrd/lpar2rrd-cgi/" |
| | <Directory "/home/lpar2rrd/lpar2rrd/lpar2rrd-cgi"> |
| | AllowOverride None |
| | Options ExecCGI Includes |
| | Order allow,deny |
| | Allow from all |
| | </Directory> |
| | {noformat} |
| | * restart your web server just in case of upgrade from < 2.40 |
| | {noformat} |
| | # /opt/freeware/apache/sbin/apachectl restart |
| | {noformat} |
| | * assure that Web server user ("nobody" in following example) has read and execution rights for LPAR2RRD CGI-BIN and data directory, also only from upgrade from < 2.40 |
| | {noformat} |
| | $ ps -ef|grep apache|head -2 |
| | nobody 233678 348372 0 Dec 30 - 0:00 /opt/freeware/apache/sbin/httpd |
| | nobody 319670 348372 0 Dec 30 - 0:00 /opt/freeware/apache/sbin/httpd |
| | $ ls -ld /home/lpar2rrd/lpar2rrd/lpar2rrd-cgi /home/lpar2rrd/lpar2rrd /home/lpar2rrd/lpar2rrd/* |
| | drwxr-xr-x 2 lpar2rrd staff 256 Dec 30 11:13 /home/lpar2rrd/lpar2rrd/lpar2rrd-cgi |
| | ... |
| | {noformat} |
| | * start the tool and check whether there are errors |
| | {noformat} |
| | $ ./load.sh |
| | $ cat error.log |
| | {noformat} |
| | * set schedule time in cron to once an hour. Run time might take more than 1/2 an hour for LPAR2RRD 2.xx if you have similar configuration as is on the [demo site|http://lpar2rrd.sourceforge.net/demo/] |
| | * if you have any problem with "Historical report" then check Apache /var/opt/freeware/apache/logs/error_log and access_log and assure that all files and dirs in /home/lpar2rrd are readable for the Apache user |
| | |
| | h1. Problems |
| | |
| | check below, if you find your problem there, otherwise feel free to contact me [Pavel Hampl|~pave1]. |
| | |
| | h2. Known issues |
| | * first of all if there is a problem with your LPAR2RRD, update to the latest LPAR2RRD as your error might be already fixed |
| | * LPAR2RRD might hang in lslparutil/lssyscfg after HMC upgrade from 6.X to 7.X. It is known problem and is necessary to remove old utilization data. You need to contact support to be able to fix it. |
| | * Do not use RRDtool 1.2.12 on Linux, it is causing several problems. Use at least 1.2.27, [here|http://www.express.org/~wrl/rrdtool/] are for example compiled RRDtool packages for Linux (do not forget for rrdtool-perl package) |
| | * 2.49 problem on Linux. When there is not XML support in Perl then even historical graphs fails. |
| | Solution is installation Perl XML::Simple module. |
| | * 2.49 problem with excluded managed systems, error: |
| | {noformat} |
| | The requested URL /lpar2rrd/10.180.37.13/MPPEBC-9119-FHA-SNXXXXXX/Excluding : MPP-NEW-AT-9119-595-SNXXXXXX/pool/index.html was not found on |
| | this server. |
| | {noformat} |
| | This will be fixed in 2.50, anyway you can fix it by removing following line 837 from install-html.sh: |
| | {noformat} |
| | if [ $DEBUG -eq 1 ]; then echo "Excluding : $mmb"; fi |
| | {noformat} |
| | * HMC 7.3.2 and less does no show additional CPU pools. It will be fixed in 2.50 |
| | * LPAR2RRD reports "time issue" error: |
| | {noformat} |
| | Tue Mar 10 08:33:40 2009:time issue : |
| | epfibm01hmc:epfibm01-9119-FHA-SN021D292 -0 - 1236696294 - 1236695623 |
| | {noformat} |
| | check the time on the HMC, it is probably wrong (out of sync) |
| | * contact the IBM representative when following happens (HMC v7.3.3.): |
| | {noformat} |
| | hscroot@hmc1:~> lslparutil -r pool -s s -m 9117-MMA*XXXXXX |
| | HSCL800E An I/O exception occurred reading the file named: {0} |
| | {noformat} |
| | * ssh on HMC sometimes refuses new connection (you can notice that some managed systems are not being updated regularly) |
| | {noformat} |
| | $ ssh -l lpar2rrd 9.26.51.203 "lssyscfg -r sys -F name" |
| | ssh_exchange_identification: Connection closed by remote host |
| | {noformat} |
| | You will see just en error message : "ssh_exchange_identification: Connection closed by remote host" |
| | It is a HMC issue (do not reboot HMC as this could temporarily fix it), contact IBM support and ask for a pernament solution (Ive seen it in HMC 7.3.3. & 7.3.3 SP1) |
| | ssh works couple of times, then it refuses new connections and after some time it works again. It looks like an identical issue fixed in 6.x but seems it is again a in place, no idea how to fix it ... |
| | check $LPAR2RRD_HOME/error.log (for lpar2rrd 2.22+) |
| | Ive just heard it should be fixed in HMC 7.3.3.0 SP2. |
| | * can happen that you run of the space on HMC (no lpar2rrd mistake!), then you got following |
| | {noformat} |
| | No valid mem data got from HMC : time = No results were found. |
| | {noformat} |
| | this confirms it: |
| | {noformat} |
| | hmc1:~> monhmc -r disk -n 0 |
| | Filesystem 1K-blocks Used Available Use% Mounted on |
| | /dev/sda2 16121184 7589780 7712492 50% / |
| | /dev/sda3 6040320 5917892 0 100% /var |
| | {noformat} |
| | call IBM support for fix it |
| | * can happen you run out of the space on LPAR2RRD hosted server, this error can appear: |
| | {noformat} |
| | # ./load.sh |
| | Use of uninitialized value in string eq at /opt/freeware/lib/perl/5.8.0/RRDp.pm line 165, <FH> line 1.ERROR: |
| | '/home/lpar2rrd/lpar2rrd/data/Server-9117-570-SN1094E4A/vnshmc02/PUDRDB1.rrm' is not an RRD file at /home/lpar2rrd/lpar2rrd/lpar2rrd.pl line 384 |
| | {noformat} |
| | free up the space, remove /home/lpar2rrd/lpar2rrd/data/Server-9117-570-SN1094E4A/vnshmc02/PUDRDB1.rrm and re-run the tool |
| | * utilization data is not being collected although it is allowed (saw that on one HMC 7.3.3 SP1, affected was only one managed system, others were ok) |
| | check that via this cmd after you allow it globally |
| | {noformat} |
| | hscroot@hmc1:~> chlparutil -r config -s 60 |
| | hscroot@hmc1:~> lslparutil -r config -F name,sample_rate |
| | p550-SN65987DD-test1,60 |
| | PWR6A-9117-MMA-SN103A4B0,60 |
| | PWR6B-9117-MMA-SN103B5C0,60 |
| | hscroot@hmc1:~> lslparutil -r pool -m PWR6A-9117-MMA-SN103A4B0 |
| | No results were found. |
| | {noformat} |
| | If it still reports "No results were found." even after couple of hours then contact IBM support |
| | Sometimes helps following sequence of commands: |
| | {noformat} |
| | hscroot@hmc1:~> chlparutil -r config -s 0 |
| | hscroot@hmc1:~> rmlparutil -m PWR6A-9117-MMA-SN103A4B0 |
| | hscroot@hmc1:~> chlparutil -r config -s 60 |
| | {noformat} |
| | Check if all dirs and files on the HMC where utilization data is collected are owned by : ccfw:ccfw, if not then contact IBM support |
| | {noformat} |
| | ls -lRa /opt/hsc/data/utilization |
| | {noformat} |
| | If nothing works then contact IBM support (btw I know an example where I was not able to fix it, must be a HMC bug) |
| | * error when run load.sh |
| | {noformat} |
| | connect: Connection refused |
| | A connection to the Command Server failed. |
| | {noformat} |
| | RMC on the HMC seems to be down, reboot the HMC (it happens often when you change sample rates of Utilization data collection) |
| | * mandatory patch MH01102 (HMC 7.3.3) stops working utilization data collection on the HMC, therefore lpar2rrd does not work\! |
| | Workaround seems to be one additional reboot after apply HM01102. Go for it, it should sort it out. |
| | * wrong daily/weekly/monthly data being displayed for one or more managed systems if you have more managed systems on 1 hmc |
| | \-> update LPAR2RRD to version 1.8 |
| | * wrong path to rrdtool perl class causes abort with this error: |
| | {noformat} |
| | Can't locate RRDp.pm in @INC (@INC contains: /usr/local/lib/perl5/5.8.8.) at /home/lpar2rrd/lpar2rrd/lpar2rrd.pl line 19. |
| | BEGIN failed--compilation aborted at /home/lpar2rrd/lpar2rrd/lpar2rrd.pl line 19. |
| | error "2" occurred, exiting before web section |
| | {noformat} |
| | \-> modify following row in lpar2rrd.pl to reflect your installed rrdtool |
| | {noformat} |
| | use lib qw (/usr/local/rrdtool-1.2.15/lib/perl); |
| | $ ls -l /usr/local/rrdtool-1.2.15/lib/perl/5.8.2/RRDp.pm |
| | -r--r--r-- 1 root system 5048 Jul 14 2006 /usr/local/rrdtool-1.2.15/lib/perl/5.8.2/RRDp.pm |
| | {noformat} |
| | * error when creating a user on a HMC with role hmcviewer |
| | {noformat} |
| | $ mkhmcusr -u lpar2rrd -a hmcviewer --passwd abc1234 |
| | {noformat} |
| | when role hmcviewer does not exist then it is most probably due to old HMC version, pls check whether you are running 5.x at least |
| | |
| | h2. Error reporting |
| | |
| | provide me following: |
| | * error log : $LPAR2RRD_HOME/error.log (only for lpar2rrd 2.22+) |
| | * output from "$ ./load.sh" or a file where is redirected output from cron (/home/lpar2rrd/lpar2rrd/load.out) |
| | * version of HMC from HMC : |
| | {noformat} |
| | # lshmc -V |
| | {noformat} |
| | |
| | h1. Release notes |
| | I prefer to make a new release immediately after a fixing one or a few bugs, so if you have no problem and there is no new feature noted in *Major news* section above then you probably do not need to follow each version. |
| | * 2.49: |
| | -* was added LPAR search option |
| | -* was added export to CSV in historical graphs |
| | -* lpar2rrd error log is now shown in web |
| | -* LPAR2RRD configuration (lpar2rrd.cfg) is also shown inweb |
| | -* minor fixies for lpar a managed names with a space inside its names (thx to Krzysztof Szczepura) |
| | -* fixed an issue with aggregated graphs on linux platform, when rrdtool crashed (thx to Elvio Prattico) |
| | -* new set of colors for aggregated graphs |
| | -* excluded managed systems are now excluded even from displaing (before they were excluded only from getting new data) |
| | -* when there is corrupted rrttool file then it is automatically removed |
| | -* fixed a bug when it did not show "LPARs aggregated" menu for servers with many lpars |
| | -* LPARs in aggregared graphs are now alphabetically sorted |
| | * 2.46 (11-Mar-2009) |
| | -* Aggregated graphs creation fails for more than 64 lpars in one server, it is fixed now. |
| | -* Fixed an issue with a dot within lpar names, now it works even for them (Thanks to James Mulcahy for spotting and fixing) |
| | * 2.45 (05-Mar-2009) |
| | -* small temp files /var/tmp/lpar2rrd.$PID are being removed now |
| | -* time range of "Historical reports" now includes possibility to choice hours |
| | -* fixed another bug in update.sh script (it appeared just on Linux) |
| | * 2.44 (15-Feb-2009) |
| | -* added LPAR CPU aggregated graphs (all LPARs in one graph) |
| | -* added possibility to change resolution of "Historical report" graphs in the web |
| | * 2.43 (01-Feb-2009) |
| | -* fixed install_html.sh for Linux platform (did not work "Historical reporting"), Thanks to Alessandro Pregarz for a fix. |
| | -* another small fix in install_html.sh on Ubuntu where missed a comma, AIX does not care about it . Thanks to Sveinn Gunnarsson for spoting and fixing it. |
| | * 2.42 (29-Jan-2009) |
| | -* fixed sample_rate.sh script, did not work since 2.40 |
| | -* support for showing alias names for CPU shared pools (POWER6), before names were always in "Shared CPU pool XY" format |
| | -* support for a hash in the managed names |
| | -* fixed update.sh when it did not work correctly after update from historical lpar2rrd versions |
| | * 2.41 (07-Jan-2009) |
| | -* just cosmetic changes in front-end |
| | * 2.40 (06-Jan-2009) |
| | -* Introduced "Historical reports" |
| | -* Introduced : Change log : state & Change log : config |
| | * 2.39 (01-Jan-2009) |
| | -* support for managed systems with a space inside a name |
| | * 2.34 (29-Dec-2008) |
| | -* "Memory resources" data in configuration inventory was not tabelated corectly |
| | -* fixed following error: Argument "" isn't numeric in numeric gt (>) at /backup/pave/rrd/lpar2rrd.pl line 321. |
| | Can't use string ("") as a SCALAR ref while "strict refs" in use at /backup/pave/rrd/lpar2rrd.pl line 371 |
| | * 2.33 (19-Sep-2008) |
| | -* fixed an minor issue with daily graphs when was not imported to RRD db shared CPU utilization of dedicated CPUs (affected was just 2.32 version, this appears in error.log : Argument "" isn't numeric in numeric lt (<) at /backup/pave/rrd/lpar2rrd.pl line 435.) |
| | * 2.32 (15-Sep-2008) |
| | -* fix for HMC V6 where LPAR2RRD fails since 2.30 |
| | * 2.31 (15-Sep-2008) |
| | -* added "Current CPU configuration" data into each lpar panel |
| | * 2.30 (10-Sep-2008) |
| | -* shows up CPU donation of partitions with dedicated CPUs with allowed processor sharing |
| | * 2.26 (09-Sep-2008) |
| | -* another fix for managed systems which are not being updated for 10 days (they are probably taken out of HMC), now it really works and they disappears from LPAR2RRD |
| | * 2.25 (08-Sep-2008) |
| | -* fixed a bug made in 2.24 when old not being updated managed systems are ignored by LPAR2RRD, after 10 days of inactivity (by a mistake there was 100days) |
| | -* update.sh script is ready, use it for the upgrade of LPAR2RRD |
| | * 2.24 (01-Sep-2008) |
| | -* physical and logical configuration is now human readable formated (thanks to Chris Ayres for great inspiration!) |
| | -* introduced script for automatic update : update.sh (just start update.sh and provide path to actual LPAR2RRD location) - well there is a bug if you have more HMCs, do not use it now, will fix it in next release |
| | * 2.23 (16-Aug-2008) |
| | -* LPARs are sorted alphabetically in the documentation part |
| | * 2.22 (15-Aug-2008) |
| | -* does not show old managed system (not being updated by utilization data more than 10days) |
| | -* support of renaming managed systems (it keeps all data after rename) the tool now collects all physical and logical info about managed systems (once a day) and shows that on the web |
| | -* improved error handling, all errors are directed to $LPAR2RRD_HOME/error.log |
| | * 2.17 (25-Jul-2008) |
| | -* all items in the web menu are now sorted alphabetically and initially is always showed the first HMC and its first managed system |
| | * 2.16 (24-Jul-2008) |
| | -* added support of "/" (slash) within lpar names |
| | * 2.15 (15-Jul-2008) |
| | -* fixed an error which showed incorrectly lpars which had a space within their names |
| | * 2.14 (10-Jul-2008) |
| | -* fixed an error during the load, details are in the error section above |
| | -* in some circumstancies could still appear error fixed in 2.12, now it should be definitely gone |
| | * 2.12 (07-Jul-2008) |
| | -* fixed an issue during load data, check the error section above |
| | * 2.11 (04-Jul-2008) |
| | -* web front-end was re-designed, it is much nicer now |
| | * 2.10 (13-Jun-2008) |
| | -* using time from HMC to figure out how much records should be read from utilization DB on the HMC, there could appeared holes if time was even slightly different on the HMC and LPAR2RRD hosted server. |
| | -* last update time in graphs is again with ":" separator |
| | -* added last data value into daily graphs into the comment line |
| | -* changed menu order in web |
| | * 2.05 (06-Jun-2008) |
| | -* fixed an issue with not showing anything in daily/weekly/monthly charts after 2.x upgrade and only for systems with 1h sample rate |
| | If you already use 2.X with 1h sample rates of some managed systems then remove those managed systems from LPAR2RRD as part of the upgrade procedure |
| | {noformat} |
| | $ rm -r /home/lpar2rrd/lpar2rrd/data/__managed_name_with_1h_sample_rate__ |
| | {noformat} |
| | Thanks to Darrell Hoffman LPAR2RRD can now work with latest RRDTool (before just with up to version 1.2.18) |
| | -* Darrell also fixed a small issue in sample_rate.sh |
| | * 2.04 (05-Jun-2008) |
| | -* after change sample rate from 1hour to 1minute, LPAR2RRD graphs still show old not being updated hourly data shared pools appeared twice in all views |
| | * 2.03 (05-Jun-2008) |
| | -* fixed an issue with showing shared CPU pools when Max CPU units was always 0 |
| | -* fixed html install script which did not put lpar names on the web if there was mixed environment of managed systems with 60s and 1hour sample rates |
| | * 2.02 (04-Jun-2008) |
| | -* support of more CPU shared pools |
| | * 2.01 (28-May-2008) |
| | -* support for sample rates 30s, 60s, 300s, 1800s and 3600s |
| | * 1.88 (26-May-2008) |
| | -* entitled utilization was by a mistake replaced by number of processors in 1.86. Pls update to this version if you use 1.86-1.87 |
| | * 1.87 (26-May-2008) |
| | -* fixed a problem in not updating daily/weekly/monthly/yearly pages since version 1.84 |
| | * 1.86 (23-May-2008) |
| | -* fixed a problem with displaying dedicated LPARS after HMC upgrade to 7.3.3 when the tool was failing to load utilization data |
| | * 1.85 (16-May-2008) |
| | -* fixed a problem with LANG variable on HMC set to other than en_US with date format different from MM/DD/YYYY |
| | * 1.84 (15-May-2008) |
| | -* already deleted LPARs disappear from the WEB menu 2 days after they are deleted, they had never disappeared before |
| | * 1.83 (10-Dec-2007) |
| | -* fixed problem during adding more than 1 new systems into monitoring on 1 HMC with freshly enabled utilization data collection then only first system appeared, the 2nd appeared after 1 day when daily stats were available |
| | * 1.82 (11-Nov-2007) |
| | -* fixed a bug when a dot appeared in lpar name, then the tool does not process it correctly. |
| | -* Fixed a problem with creating lpar named "*" when no any lpar is configured yet on the managed system |
| | * 1.81 (03-Sep-2007) |
| | -* script which creates html front-end now runs always even if lpar2rrd.pl fails |
| | * 1.80 (13-Jun-2007) |
| | -* fixed a bug in wrong daily-weekly-monthly resolution as it always was 1 day instead 1hour. It happened only if there were more than one managed system controlled over 1 hmc, first was always ok, the others were affected. |
| | -* There were not created charts which are always overwritten for daily and hourly stats, it just speeds up processing |
| | * 1.70 (11-May-2007) |
| | -* enhanced load.sh about MANAGED_SYSTEMS_EXCLUDE (put systems with old firmware which do not support utilization data collection to avoid them) |
| | * 1.60 (23-Apr-2007) |
| | -* fixed a problem when there is a gap within the input data |
| | * 1.50 (02-Feb-2007) |
| | -* *Ive made the project public on IBM Wiki* |
| | -* fixed a bug when it did not work with lpars which had a space in their name |
| | * 1.40 (24-Jan-2007) |
| | -* fixed wrongly counted CPU Utilization for Shared processor pool graphs |
| | * 1.00 (16-Jan-2007) |
| | -* *initial release* |
| | * project start (Dec-2006) |
| | -* *I was asked to graph lslparutil data from iSeries servers by one of our customers* |
| | |
| | h1. Disclaimer |
| | |
| | This tool is not officially supported by IBM. No warranty is given or implied, and you cannot ask for help IBM. |
| | It is a personal project of the author. It is being written only in his free time, *it is not an IBM sponsored project!* |
| | |
| | h1. Nice examples |
| | * Aggregated graph fully utilized p595 (64CPUs, 94 LPARs) |
| | | * Aggregated graph fully utilized p595 (64CPUs, 94 LPARs, on the bottom are CPU dedicated partitions) |
| | !aggregated-p595.png! |
| | |
| | | * Another p595 in yearly graph |
| | !aggregated-p595-1.png! |
| | * anyone has another nice one? Then post me it. |
