We have started work on a long list of  new features and enhancements to our governance software suite: BPC RiskManager, and, assuming the world doesn’t come to an end this year, we think you are going to like what is coming – but more on that in future posts. 

Our four main service lines -  Software Sales,  Custom Software,  Survey Hosting and Consulting have opened the year with the strongest bookings in a number of years so 2012 looks like it will be an interesting and busy year.  The Custom Software and Consulting teams are now fully committed for the first quarter (obviously sales and hosting don’t have capacity limits) , which is a good way to start the new year.  

2011 was a great year at BPC, and we thank you, our clients and friends for your on going dedication and support – without you we don’t exist.  To our clients, while we don’t get to see many of you personally, our regular telephone conversations and email communications make us feel we know each of you as friends.  We couldn’t ask for a better bunch of clients – loyal, considerate, helpful cooperative, inspirational and fun.   A simple “thank you” just doesn’t seem enough.

 For BPC RiskManager V6 users there has recently been a major update released with lots of new features and a couple of progressive updates since then, so for those with current maintenance subscriptions, make sure you contact us and get the update, and those of you without, make sure you get yourself current and we’ll get you the update.

The problem with the Linux based Acer Aspire Easystore is that while the unit will happily allow drives to be replaced, if you lose the motherboard or the Ethernet port you are in serious trouble.   It surprisingly easy to suffer a catastrophic problem.  For example, if you upgrade the firmware from 2.0 to 2.5 your unit can stop responding.   Further you can lose the motherboard  for a variety of  other reasons – for example, they seem especially sensitive to minor DC power spikes.  If the dreaded red light indicating “unrecoverable software” lights up or the unit simply stops talking to the network, you will rapidly discover you  are essentially screwed. 

Acer’s maintenance support, seems excellent when you contact them.  They will happily replace the unit for you (in fact that is all they will do!) and even send out a technician to swap your old still working drives into the new unit.  Unfortunately that is when you discover you have a serious problem.

The replacement unit will not read the RAID array from the old drives – even if your drives are otherwise OK.  If you are lucky, it will recognise the user accounts – but it will not give you back access to your data.  Acer’s helpful response is that you must rebuild the array.  Do not under any circumstances do this or you will lose your data!

This is what you must do…

In our case we always use RAID 1 on low cost RAID systems.  This essentially gives 2 copies of each disk without co-dependence across multiple disks for data recovery.

Recovering your data from the Acer becomes quite simple in this situation.

What you will need:

A spare computer (a Windows desktop is fine) with:

•  at least one spare SATA port,
• a spare SATA cable, (or just unplug an existing drive from the desktop computer and use its cable),
• a spare HD drive  in FAT32 or Linux disk format so it can be accessed easily under Linux (or other NAS on the network) with enough space to store the files you recover from your Acer Easystore drive(s)
• a CD ROM. 

Either the on the same  machine, or another, you need a writable CD drive, a spare blank CD and an internet connection.

Now follow these steps:

Step 1  Assuming you do not have a way to write ISO formats onto a CD ROM go here and download the ISO writer:

http://isorecorder.alexfeinman.com/isorecorder.htm

Just double click to install it and once installed you merely right-click on an ISO image and choose “Copy image to CD”.

Step 2: Get a copy of the latest version of Knoppix Linux from here (or other Knoppix mirror).   I used Knoppix 6.4.4

http://www.mirrorservice.org/sites/ftp.uni-kl.de/pub/linux/knoppix/

You do want the ISO image and unless you are visually impaired you do NOT need the ADRIANE version.

Step 3.  Put your blank CD into your writable CD drive and right click on the ISO image you just downloaded.  Choose “copy image to CD”.  This will give you a bootable version of Knoppix Linux on a CD ROM.

Step 4.  Turn off and open up your spare desktop computer and plug the first RAID1 drive from your dead Acer Aspire Easystore into the SATA data cable and the SATA power cable and plug the other end of the SATA data cable into a spare SATA port on the motherboard (or SATA sub-board).  Note the number written on the motherboard this will help you find the drive later.  You only need to plug in one drive from each RAID1 pair.  The drives are arranged in pairs making one volume each in the easystore – so the top two are one volume and the bottom two are another volume.  You want one drive from each volume.  We will just do one volume in this sequence, and you should repeat the following steps for the second volume after completing the first.

Step 5.  Turn on the computer and if it does not already boot from a CD ROM, trigger the BIOS setup on startup and change the boot order so that it boots from a CD ROM first.  Save and exit and allow the computer boot up.

Step 6.  Once Linux has started (you may have some minor config steps to work through), click on the little filing cabinet in the bottom left hand corner of your screen.  This will show you the mounted devices.  If your machine also has NTFS drives in it or even an existing RAID array in it there will be some drives marked OS which you will not be able to access.  Ignore these.  Your FAT32 drives and your Acer EasyStore Raid1 drive should be visible.  The mounted Acer drive will show as something like “sdc4″, where the “sdc” part might be sda, sdb, sdc, etc.  The number refers to the partition on the drive that could be read.  This will not have your data in it – so don;t get excited just yet.

Step 7.  In the bottom left hand corner of your screen, beside the filing cabinet icon is the Knoppix logo.  This is the equivalent of the windows “start menu”.  Click on this and from “preferences” choose “GParted”.  Let this do its stuff and when it has finished opening select the GParted menu, and from that the “Devices” submenu.  From this look for and select your Acer drive.  In my case it was on /dev/sdc -  yours may be different – but will be the one that was showing the “4″ in its name when you looked in the file system display.

Step 8.  After GParted has scanned it, you will notice this has 4 partitions.  The Linux swap and the system – both of which will be mounted and then two others – which will not be mounted.  The big one has your data, and the little one is just some spare space.  The big one will probably have a “2″ in its name.  In my case it was “sdc2″.  It will also be showing the file system type – probably “ext2″.  Note both the name (eg. sdc2) and the file system (eg “ext2″) – you will need them in a second.

Step 9.  Select this line by left clicking on on it and then right click to bring up the context menu.  From the context menu that displays choose “Manage Flags”.

Step 10.  In the “manage flags”  window that opens tick the “Raid” flag.  Then open the “Manage Flags” menu again and untick the “Raid” flag.  Close GParted.

Step11. From the “start menu” in the lower left hand corner of your screen choose “Accessories / Root Terminal”. 

Step 12.  When the console window opens type the following:

mount   /dev/sdc2   /media/sdc2 -oro -text2

NOTE:  change the “sdc2″ to the device name (and partition number) you saw in GParted earlier and the  “ext2″ file system type to the one you saw there as well.  It will most likely be ext2, but it might be ext3.   You have just mounted your lost data as “/media/sdc2″  – or whatever sd you entered above.

Step 13.  Click on the filing cabinet icon in the bottom left of the screen and in the address bar enter (adjusted appropriately for your device name):

/media/sdc2/

Your lost files should now be visible. 

Step 14:  Copy your lost files to your spare drive or other network store.   The Acer drive was opened in read-only mode by the mount command so you will can only copy or read them.

If you have two RAID1 volumes in your Acer EasyStore you should repeat all the steps  from step 4 for the second volume as well.

Good luck!  Feel free to drop me an email (or a comment here) if you have a question.  You can also rebuild a RAID5 array in a similar fashion – but it is a little more complex than the process for RAID1 recovery.

We received an excellent specification from one of our clients – MFS – for a ‘Topic Based” view of compliance to augment the risk of non-compliance concept supported in the earlier BETA’s and we have integrated that concept into this release of Beta7.   

There are some new charts, report views and export capabilities to be brought into the system yet to complement the assertion views of risk so expect BETA7  to be rapidly replaced by BETA8 

In this release:

1. Main Features Added

• Document Management
• Topic View and Topic Solver
• Publishing Authorities
• Current and Residual Bubble Charts and Expanded MS Word export

2. Detail

• Vastly improved flow of control in Compliance Unit.
• New custom screen architecture to allow inclusion of user customised data entry and access views and other optional screens.  The MFS Topic Solver is the first such example.
• New configurable MFS Risk/Compliance Topic Solver Sub-system.
• New Document management with sections module.  A Document+Section  can be linked to each risk, and compliance questions/assertion tested against the documents.  The document management centre and finder support views of risks by document list, document, section list and section, with word export and drill through to risk details from the document centre.
• New Current and residual risk bubble charts with controls ratings views
• Revised chart screen presentation
• Revised and improved MS Word export from Chart screen.
• Associated new Administration and Setup-Profiles configuration screens.
• SurveyManager – New data entry modes and markup tags added
• SurveyManager – New anonymous and on-demand survey publishing support
• SurveyManager – Potentially annoying bug due to bug tracking system dependencies removed.
• Bug Tracking – The “Phone Home” behaviour of the RM bug tracking system has been changed to write a local log instead.  The phone home system was good where client networks allowed the message through (as some of you discovered when we phone you to tell you there was something wrong with your server, but too unreliable for general use due to the few clients allowing the transmissions.  We have therefore replaced it with a log file, in the temporary files directory so we can at least get the log data if needed. It is only written in the event of a critical system issue.

In this BETA our latest thinking on how a whole of organisation/government governance system should work starts to become clear.   The unique twist is that we are using the risk record to store risks, compliance obligations and compliance obligation events.  So every risk record contains all the information necessary for it to be a compliance record, and vice versa.  Thus every risk has a compliance view and every compliance issue has a risk view – and you can enter data into either view at the same time.  With the addition of documents and sections (for legislation, policy & procedure references, etc) each of which has at least one person responsible for it, you can also see all the compliance obligations + the calculated risk of the obligation breaches by document/section.  Combining this with the already existing other of ways of structuring risks (folders, master-child trees, documents, people, complex searches, etc) you can navigate and summarise the risk/compliance landscape from different angles and using differing ratings and tie it all back to the corporate plan.

Every risk or compliance item can have an indefinite number of groups of assertions/questions linked to it so differing teams can apply multiple and differing checklists and evaluation criteria to the same issue.  In this mode, risk-causes become factors explaining a less than optimal question response adding a layer of structure to a risks usually flat causes list, and each cause/contributing factor can have its own list of strategies / controls to manage / respond to it.   Thus the risk’s flat strategy list becomes structured by the causes of the risk.  Given that all this can be linked to a document and a section within that document (and the document and section have their own activation and update control responsibilities and tracking) you now have an additional set of ways to slice the risk/compliance data.

There is quite a bit more coming in this area as we surface the new corporate planning, compliance obligations, financial elements, and performance management modules progressively over coming releases.  There are a number of other components in the wings.  You may noticed the FlowCharting module with its script engine already in your client that has been there for sometime.  Apart from allowing you to draw and execute scripted multi-user flowcharts, the module is also intended to connect to hooks sewn through your database.  It has been waiting for the some of the other parts of the system to catch up, and you should soon see new flowchart views hooking the various objects  like risks, obligations, strategies, etc together.

Compared to a commercial environment, open-source developers don’t have access to the same tools, nor the ability to hand pick and hone development teams over many years.  Debugging, optimisation and code tracing tools are simply not the same.  Contrast to a Delphi environment costing thousands of dollars per developer with full production grade debugging support and the ability to use commercially purchased components – the open source development team (if not necessarily the individual developer) must use a mixture of tools not necessarily purpose built to work together.  Teams self select, although inevitably filtering occurs over time, and project direction is committee based and therefore more difficult to keep focused than CEO based direction.  So the standard achieved by all of these components is remarkable.

In our case the instability seemed to be caused by insufficient memory being allocated to PHP.  We upped it to 256MB and the intermittent page failures disappeared.  I suspect we will need to go higher than that as the usage grows – or possibly lower, depending on whether the space is session based or shared. 

I am not sure how the PHP memory management interacts with the OS memory management, not whether the PHP memory space is shared between PHP applications, between sessions or specific to each session or application.  We, therefore, will adopt an approach of incremental growth/tuning until we have had a chance to read up on the PHP memory management system and had more time to understand the loads being placed on the servers.  I suspect that the PHP memory model differs between CGI and in-process implementations implementations anyway.   We have a number of large database and graphical processing systems on the dev cluster  so I don’t want to unnecessary rob memory from those systems, otherwise we would just allocate a few gigabytes to the PHP systems.  Based on the observed behaviour, I suspect it is shared between sessions and apps within the same inprocess server (which would make sense).  Whether is is assigned, or meerely available is another issue, and not clear at this point, nor whether unused memory is free to paged out of RAM or returnes to the OS when not required.

We are noting that we tend to have to specifically instruct the PHP apps to tell browsers not to store the pages in the caches (see the earlier post).  I think this is probably a good point for PHP programmers to note.  There are very few scenarios where a PHP web page should be cached – because by definition, they are generated – so it is a good idea to ensure that the header sent by the PHP page includes a ‘header(“no-store”);’ instruction by default – unless you specifically want it to be cached on the browser.

I must say: the more I look into the architecture and code of the WordPress system, the more impressed I am.  What is going on behind the scenes to display this blog to you is simply outstanding.  It is not so much the display method that is so clever (although that is clever enough), but the architecture and the administration/maintenance and update mechanisms are nothing short of brilliant.  There have been some very, very talented people working on this application.

The reality is that our email response system has been so good and the robustness of the applications so high that there simply is not the demand for a bulletin board style help system.  Even though the team.bishopphillips.site has been available for client access, clients have almost universally opted simply to email issues and questions and get a response in a few minutes to a couple of hours.  We sweep the emails every month and add common issues, or issues we think are of public consumption to the FAQ in the riskwiki.  The fact that a traditional layer of support is really just not needed with BPC RiskManager  is difficult to explain to new clients and so we have decided to launch a site.

It is monitored continuously by the BPC Dev team so we should see stuff arriving pretty quickly.  Please remember that it is new and the software is a little experimental so it may have short outages from time to time.  If it is down when you access, try again in a few minutes and it should automatically restart in about 2 minutes.

The RiskWiki (MediaWiki), the Blog (WordPress) and the Forum (MyBB) all use the same underlying technology - PHP and MySQL.  The introduction of the Forum to that set has introduced some instability which we are still working through so periodically one causes the other to fall over.  Everything is on auto restart, so if something is down when you first access it will come back  up in a few minutes.  I am sorry about this, but hopefully over the next week we will find the culprit of the problem and get back to our normal 99.9% system reliability.