If you just want the S812LC pictures, you should scroll down a bit ...     For reasons I do not understand there are very few picture of this new machine from IBM.  So I want to fix that in this blog entry and as a picture says 1000 words this will hopefully explain a lot about the machine - the equivalent of 37,000 words!   First, thing you need to know is that it is not in the regular IBM Power Systems series of machines. It is designed and developed by OpenPOWER Foundation early member Tyan use the POWER8... [More]

Etiquetas:  hat openpower l4 ubuntu tyan spark controller mike opal power8 s812lc pearson habanero red hadoop powerkvm cache

I recently got asked about this - here is my response (the language has been toned down a bit from the original). Internal disks on a E870/880? - you having a laugh!   Are you mad? Are you a Luddite? Is the date 1995 in your country? Three reasons: VERY slow (200 IOPS) compared to a caching SAN disk unit with dozens of disks or even a Flash SAN unit at hundreds of times faster.  Why would you want your VIOS to be that slow at disk I/O?   MUCH lower RAS than a SAN disks (single brown spinning disks are single points of... [More]

This is an update to an older AIXpert Blog entry here: Shared Storage Pools 4 - The search for a better lu -list command   With Shared Storage Pools phase 5 coming soon,  we have the lu -list command that output the Logical Unit (virtual disks) in Tier then Alphabetical order - hurray! But it still has challenges and one command output can't please every one!   Trouble is everyone has there own ideas of  “perfect layout” It is a "no... [More]

    We have just had our Advanced Technology Support Early Ship Program (ESP) proto-type E850 machine refurbished at the Power Development Lab in the USA to bring it up to the General Available (GA) hardware level.  A large number of parts were replaced (these part seemed to be working fine during our test but need to get to the official level).  This is perfectly normal ESP process of removing early and test parts of a machine that is now in normal service.   What changed apart from internal electronics which... [More]

Update Version 14: Just release version 14 (17th March 2016) Changed format a little in light of experience of uploading to the Work Load Estimator Changed to official Column header line Note: it is mandatory to include the header line generated but nmon2WLE -c The online IBM WLE tool needs a specific string to identify the System = the web tool pops-up the list for you to select from. Unfortunately, these strings and user friendly server names ( like Power Systems E850) are not available within AIX so nmon reports just the System Model:... [More]

  Warning: If you get the nslim or dd commands wrong then you  will destroy your LU virtual disks & virtual machine content.   Please tell me, if you do - I enjoy a good laugh :-) Please, test the commands & your understanding on a test SSP!   Shared Storage Pool four Ksh script commands and a program that I hope you will find useful:   ncluster - status of all VIOS   nlu - improved lu replacement   npool - storage pool use   nmap - finds if a lu is online (mapped) on any VIOS   nslim - copy... [More]

I have been asked a couple of times for something like this for Shared Storage Pools (SSP) and I don't know of an official Best Practice and FAQ so here goes. This will always be a work in progress and covers new SSP releases as new features arrive - last updated March 2017 -  NEW  I have had loads of input/ from the VIOS SSP developers too, so its not just my thoughts and jokes!!   If you have a question or a good tip or suggestion PLEASE: make a comment and I will include it. Shared Storage Pool - Best Practice and FAQ... [More]

A POWER user asked ATS: "How to replace an old HMC (7042-CR4) for a new model HMC (7042-CR8) with the latest software?" Out of the two of us: my colleague Gareth Coates is by far the HMC expert and so I deferred to him and here is the response.   The HMC documentation and Redbooks don't seem to cover a HMC hardware replacement but it is simple to do - if you just treat it as a software upgrade only to a different machine. Fortunately, if you unplug your old CR4 after the backup then you have a very simple recovery... [More]

I often get asked:  "I (or my customer) thinks that the AIX system is going a bit slow!" and can I send you some nmon data. First, while nmon has performance data it is not aimed at problem diagnostics. Second, until you have a PMR (the IBM Problem Management Record) - you don't actually have a problem by definition here at IBM. Here is my boiler plate answer - I regard this as a "work in progress" as I might add further thoughts ( GREEN is a pre PMR sanity check , BLUE is the PMR preparation , RED is the... [More]

  Finally, I get round to trying a PowerVC AIX image deploy using the Cloud-Init activation engine instead of the older Activation Engine originally used with System Director.   First, let me say I am sold on the Cloud-Init approach and will be using it from now on but it can be a bit tricky to get started - in this blog I hope to smooth your way so its quick and simple.   There are two hiccups you need to over come: The documentation is not not helpful. I do not like manual pages that try to cover software install onto four... [More]

OpenStack is a fast moving open source project and rapidly adding new features. It is actually very impressive how they have so many interrelated projects (one or more for every sub-system lets call it 60) and still manage to coordinate regular testing and releases every six months.  This is of course, primarily, designed for running virtual machines in "cattle mode" = lots of them, they are all the same and if one gets sick then shot it in the head and make another one. PowerVC then takes the official newly released OpenStack... [More]

General Listing of the LUs The lu -list commands outputs the Logical Units (LU = virtual disk from the pool of LUNs) in a random order at present (this may get fixed). Also the look could be improved greatly I dislike the intermixed SNAPSHOT header lines and snapshot names - that makes it very hard to scan with the eye The LU names are on the left - if you use PowerVC is uses 64 character long LU names than means all the stats are a complex jumble on the right - that makes it impossible to scan 99.999% of the time I don't what those 16... [More]

Assuming you have grown your Shared Storage Pool in the past by adding more LUNs and you have say 32 LUNs (twice if mirroring with a failgrp) then you can grow you SSP the easy way and by magic.  Just get the SAN team (that could be yourself, of course) to expand the size of the Pool LUNs.  Our minimum size LUN policy is 128 GB and our Repository LUNs are 1 or 2 GB.  I tried this experiment based on one line in a presentation on the next SSP release by the developers, where they said "The Share Storage Pool for a few version... [More]

Here are my notes of the Shared Storage Pools commands that you need to remember.  I have said it before but SSP is big on concepts and system admin time saving but very simple to operate.   Creating the pool (once only) by example: Assumption: 1 GB Repository disk = hdisk15 and Pool data disks = hdisk7 to hdisk14, cluster and pool both called "alpha" Assumption: IBM V7000 A has LUNs: hdisk7 hdisk8 hisk9 hdisk10 Assumption: IBM V7000 B has LUNs: hdisk11 hdisk12 hisk13 hdisk14 Examples cluster -create -clustername alpha ... [More]