Jaqui's AIX Performance and Security Blog

Thu, 25 Jul 2013


July 25, 2013 How to move a standalone server to be HMC managed Just recently I had cause to move a number of standalone systems (power6 and 7) that had never been attached to an HMC to under the control of HMCs. Interestingly enough I could find clear documentation on how to add a completely fresh HMC and on how to deal with older servers but nothing current, so thought I would share what the process ended up being for me. First we connect the server HMC1 port to the HMC or to the private switch that the HMC is attached to Of course we make sure that whatever private IP range the HMC is using is not the same as the public IP address for the server Getting HMC to recognize the server After a few minutes the server should pop up under servers on the HMC At this time you can rename the server if you like as the default machinetype-model-serial The status should show "pending authentication" Click on this to set the passwords or you can go to operations, change password You should be prompted to set the password for HMC Access, ASMI General and ASMI Admin. Write these down somewhere in case you forget them later. Once it grants access you are ready to move on to the next step which is fixing issues around serial ports and access to the console. At this point under the server you should see an automatically built LPAR and profile. The servers I cut over only had one LPAR on them so the HMC built an LPAR and profile that owned all resources. You can modify these later if you want. Process after providing hmc access to server On HMC go to the server and bring up properties Under general ensure that "Manufacturing default configuration" is set to false If it is true then go to http://www-01.ibm.com/support/docview.wss?uid=isg3T1010949 and follow the procedure there. If it is false then we need to clear the vty definitions on the lpar as follows: We have to do this because the serial ports are usable whent his is a standalone server but they are not if it is HMC managed. Until you take this step you can control the LPAr from the HMC but you won''t get any information on the console. Login to the LPAR as root odmdelete -q name=vty0 -o CuDv odmdelete -q name=vty1 -o CuDv odmdelete -q name=vty2 -o CuDv odmdelete -q name=vsa0 -o CuDv odmdelete -q name=vsa1 -o CuDv odmdelete -q name=vsa2 -o CuDv odmdelete -q attribute=syscons -o CuAt bosboot -ad /dev/ipldevice sync; sync; sync savebase shutdown -r now


posted at: 18:05 | path: | permanent link to this entry

Thu, 18 Jul 2013


July 18, 2013 Thought I would provide some useful links that people have been asking me for Forsythe Talks 1. Forsythe - www.forsythe.com 2. Our Forsythe Talks series of Technical Talks 3. Replays of Our Forsythe Talks series of Technical Talks IBM Systems Magazine Articles 1. Charlie Cler 2. Andy Goade 3. Jaqui Lynch Other User Groups 1. PowerVM UK User Group Presentations 2. US AIX User Group Presentations Jay Kruemke Jay Kruemke Twitter – chromeaix 1. Jay's Twitter feed 2. Jay's blog for technical advocates Nigel Griffiths Nigel's Twitter feed Nigel's aixpert Blog Nigel's Youtube Channel Connecting You can connect to me on Linkedin, facebook or via email I have a twitter account but rarely if ever use it 1. Facebook 2. Email: aixperformance@facebook.com or performance@circle4.com Useful Tools 1. HMC Scanner 2. Workload Estimator 3. AIX Performance Home 4. AIX Performance Tools 5. VIOS Advisor 6. IBM Systems Energy Estimator 7. IBM SPT (Systems Planning Tool) nmon links 1. nmon wiki 2. nmon analyser wiki 3. nmon consolidator wiki


posted at: 13:38 | path: | permanent link to this entry

Wed, 20 Oct 2010


Oct 20, 2010 Well it has been a while so thought I would post a couple of useful links here I now work for Forsythe - www.forsythe.com I will post later the links to my presentation at Power Technical University 2010 in Las Vegas. Also posting the link to the latest article on AIX Performance: AIX Performance And all my eserver articles eserver And Charlie Cler's PowerVM Disk Virtualization article: PowerVM Disk Virtualization On Facebook search for performance@circle4.com


posted at: 16:01 | path: | permanent link to this entry

Thu, 04 Sep 2008


Latest Tunable recommendations for AIX v6 and AIX v5.3 BOTH NETWORK no -p -o rfc1323=1 no -p -o sb_max=1310720 no -p -o tcp_sendspace=262144 no -p -o tcp_recvspace=262144 no -p -o udp_sendspace=65536 no -p -o udp_recvspace=655360 Also check the actual NIC interfaces and make sure they are set to at least these values For AIX v5.3 No need to set memory_affinity=0 after 5.3 tl05 MEMORY vmo -p -o minperm%=3 vmo -p -o maxperm%=90 vmo -p -o maxclient%=90 vmo -p -o minfree=960 vmo -p -o maxfree=1088 vmo -p -o lru_file_repage=0 vmo -p -o lru_poll_interval=10 The parameters below should be reviewed and changed (see vmstat –v and lvmo –a later) PBUFS ioo -p -o pv_min_pbuf=1024 JFS2 ioo -p -o j2_maxPageReadAhead=128 j2_dynamicBufferPreallocation=16 Default that may need tuning Replaces tuning j2_nBufferPerPagerDevice JFS ioo -p -o numfsbufs=1024 ioo -p -o maxpgahead=16 For AIX v6 Make the network changes above Memory defaults are already correctly set and should not be changed If you upgrade from a previous version of AIX using migration then you need to check the settings though The parameters below should be reviewed and changed (see vmstat –v and lvmo –a later) PBUFS Tune these using lvmo for the individual volume group pv_min_pbuf is now a restricted tunable JFS2 ioo -p -o j2_maxPageReadAhead=128 (default above may need to be changed for sequential) j2_dynamicBufferPreallocation=16 Default that may need tuning Replaces tuning j2_nBufferPerPagerDevice JFS ioo -p -o numfsbufs=1024 (now restricted) ioo -p -o maxpgahead=16 (now restricted) CHANGES TO CONSIDER TO ADAPTERS AND HDISKS Check to make sure your vendor supports these: fcs0 etc adapters init_link pt2pt INIT Link flags True max_xfer_size 0x200000 Maximum Transfer Size True num_cmd_elems 2048 Maximum number of COMMANDS to queue to the adapter True Be careful with init_link as it defaults to AL - this setting is switch and SVC dependent. Check queue_depth for hdisks - lsattr -El hdisk? Use iostat -aD to look for sqfull conditions that would indicate the need for an increase in queue_depth


posted at: 22:57 | path: | permanent link to this entry

Mon, 03 Dec 2007


Latest Set of General Purpose AIX v5.3 Tunables Please ensure you test these first before implementing in production as your mileage may vary Network no -p -o rfc1323=1 no -p -o sb_max=1310720 no -p -o tcp_sendspace=262144 no -p -o tcp_recvspace=262144 no -p -o udp_sendspace=65536 no -p -o udp_recvspace=655360 nfso -p -o rfc_1323=1 nfso -p -o nfs_socketsize=600000 nfso -p -o nfs_tcp_socketsize=600000 Memory Settings vmo -p -o minperm%=5 vmo -p -o maxperm%=90 vmo -p -o maxclient%=90 vmo -p -o minfree=1000 vmo -p -o maxfree=1200 vmo -p -o lru_file_repage=0 vmo -p -o lru_poll_interval=10 Leave strict_maxperm and strict_maxclient at their defaults IO Settings Leave minpgahead and J2_minPageReadAhead at their defaults of 2 ioo -p -o j2_maxPageReadAhead=128 ioo -p -o maxpgahead=16 ioo -p -o j2_nBufferPerPagerDevice=1024 ioo -p -o hd_pbuf_cnt=1024 ioo -p -o numfsbufs=2048 ioo -p -o pv_min_pbuf=1024 #### this is the 5.3 value - below is the one for 5.2 ### ioo -p -o pv_min_pbuf=1024 If doing lots of raw I/O you may want to change lvm_bufcnt Default is 9 ioo -p -o lvm_bufcnt=12 Others left to default that you may want to tweak include: ioo -p -o numclust=1 ioo -p -o j2_nRandomCluster=0 ioo -p -o j2_nPagesPerWriteBehindCluster=32 ioo -p -o j2_maxRandomWrite=32 ioo -p -o maxrandwrt=32 These are starting points only Other areas to look at tuning: Fiber cards Check out max_xfer_size - 0x100000 is 16mb dma, 0x200000 is 128mb dma For heavy I/O you may want to change this to 128mb hdisks Check out queue_depth. If seeing sqfull on iostat -D then increase queue_depth if your disk subsystem allows it


posted at: 09:48 | path: | permanent link to this entry

Thu, 08 Feb 2007


2/8/2007 AIX v5.3 Links to get perfscript and instructions Script to gather data http://www.circle4.com/jaqui/perfdata53-latest.sh.txt Instructions for script and nmon http://www.circle4.com/jaqui/perf-script-instructions.txt


posted at: 18:18 | path: | permanent link to this entry


2/8/2007 AIX v5.3 Tunables Update When making tunable changes please be sure to test them first. 1. Please be careful when setting lru_file_repage. If you are asked to change this value from the default (which you should) please ensure you first set maxperm and maxclient back to 80% 2. I do want to mention that a couple of the parameters may not work well with certain workloads, specifically maxrandwrt and maxpgahead. The situation where I saw a problem was very application specific (Topform and Providex) but try it both ways. I normally set them for JFS and JFS2 as follows: ioo -p -o j2_maxPageReadAhead=128 ioo -p -o maxpgahead=16 ioo -p -o j2_maxRandomWrite=32 ioo -p -o maxrandwrt=32 However, you may be better to leave them at the defaults: ioo -p -o j2_maxPageReadAhead=128 ioo -p -o maxpgahead=8 ioo -p -o j2_maxRandomWrite=0 ioo -p -o maxrandwrt=0


posted at: 18:07 | path: | permanent link to this entry


2/8/2007 General Purpose AIX v5.2 at >ML4 Tunables Current recommended AIX v5.2 ML4 Performance Parameters: Please ensure you test these first before implementing in production as your mileage may vary Network no -p -o rfc1323=1 no -p -o sb_max=1310720 no -p -o tcp_sendspace=262144 no -p -o tcp_recvspace=262144 no -p -o udp_sendspace=65536 no -p -o udp_recvspace=655360 nfso -p -o rfc_1323=1 nfso -p -o nfs_socketsize=600000 nfso -p -o nfs_tcp_socketsize=600000 Memory Settings vmo -p -o minperm%=5 Leave maxperm and maxclient at default of 80 Leave strict_maxperm and strict_maxclient at their defaults vmo -p -o minfree=960 vmo -p -o maxfree=1088 vmo -p -o lru_file_repage=0 vmo -p -o lru_poll_interval=10 IO Settings Leave minpgahead and J2_minPageReadAhead at their defaults of 2 ioo -p -o j2_maxPageReadAhead=128 ioo -p -o maxpgahead=16 ioo -p -o j2_maxRandomWrite=32 ioo -p -o maxrandwrt=32 ioo -p -o j2_nBufferPerPagerDevice=1024 ioo -p -o hd_pbuf_cnt=1024 #### this is the 5.3 value - above is the one for 5.2 ### ioo -p -o pv_min_pbuf=1024 ioo -p -o numfsbufs=2048 If doing lots of raw I/O you may want to change lvm_bufcnt Default is 9 ioo -p -o lvm_bufcnt=12 Others left to default that you may want to tweak include: ioo -p -o numclust=1 ioo -p -o j2_nRandomCluster=0 ioo -p -o j2_nPagesPerWriteBehindCluster=32 These are starting points only


posted at: 18:06 | path: | permanent link to this entry


2/8/2007 General Purpose AIX v5.3 Tunables Current recommended AIX v5.3 Performance Parameters: Please ensure you test these first before implementing in production as your mileage may vary Network no -p -o rfc1323=1 no -p -o sb_max=1310720 no -p -o tcp_sendspace=262144 no -p -o tcp_recvspace=262144 no -p -o udp_sendspace=65536 no -p -o udp_recvspace=655360 nfso -p -o rfc_1323=1 nfso -p -o nfs_socketsize=600000 nfso -p -o nfs_tcp_socketsize=600000 Memory Settings vmo -p -o minperm%=5 Leave maxperm and maxclient at default of 80 Leave strict_maxperm and strict_maxclient at their defaults vmo -p -o minfree=960 vmo -p -o maxfree=1088 vmo -p -o lru_file_repage=0 vmo -p -o lru_poll_interval=10 IO Settings Leave minpgahead and J2_minPageReadAhead at their defaults of 2 ioo -p -o j2_maxPageReadAhead=128 ioo -p -o maxpgahead=16 ioo -p -o j2_maxRandomWrite=32 ioo -p -o maxrandwrt=32 ioo -p -o j2_nBufferPerPagerDevice=1024 ioo -p -o pv_min_pbuf=1024 ioo -p -o numfsbufs=2048 If doing lots of raw I/O you may want to change lvm_bufcnt Default is 9 ioo -p -o lvm_bufcnt=12 Others left to default that you may want to tweak include: ioo -p -o numclust=1 ioo -p -o j2_nRandomCluster=0 ioo -p -o j2_nPagesPerWriteBehindCluster=32 These are starting points only


posted at: 17:35 | path: | permanent link to this entry

Mon, 29 May 2006


Using and Obtaining nmon Nmon is available for AIX and Linux (x86 and Power) Download the latest nmon for AIX v5.3 from http://www-941.ibm.com/collaboration/wiki/display/WikiPtype/nmon You will need to gunzip it and untar it I untar it to /usr/local/perf Then run nmon_aix53 as follows: cd /usr/local/perf nmon_aix53 -fT -c 30 -s 240 This will cause nmon to run for about 2 hours gathering data - the output will be put in /usr/local/perf Once it is finished (check using ps -ef | grep nmon) then use nmon analyzer and MS Excel to analyze the output nmon analyser can be found at: http://www-128.ibm.com/developerworks/eserver/articles/nmon_analyser/index.html or http://www-941.haw.ibm.com/collaboration/wiki/display/WikiPtype/nmonanalyser You can find out more about nmon at Nigel Griffiths' online workshop at: http://www-941.ibm.com/collaboration/wiki/display/WikiPtype/nmon+Introduction+Workshop


posted at: 22:07 | path: | permanent link to this entry

Wed, 19 Apr 2006


Welcome to my Blog. As I have time I will put updates on various technical issues here for you to check out. If you have questions, feel free to email me and I will either answer them here or I will answer them via my column in IBM's Systems Magazine for pSeries (used to be eServer for UNIX),

posted at: 13:43 | path: | permanent link to this entry