PuTTY window closes after a few minutes – Fixed

If you find that the PuTTY window closes itself after a few minutes of inactivity, there is a setting change you can make to prevent this.

When PuTTY starts, click on ‘Connection’ on the left hand side, then change the entry ‘Seconds between keepalives (0 to turn off) from the default of 0 to 5.

*Caveat: You will now require to manually close the connection, as PuTTY will send null packets to the host to prevent it timing PuTTY out due to inactivity – unless the host has a maximum session timeout or a loss of network connection.

To keep the setting on future PuTTY sessions, click ‘Session’ on the left, click ‘Default Settings’ then press the ‘Save’ button.



Experiences of and preparing for a large public speaking event

Ian Noble @ Oil Gas ICT

How I got the gig

In November 2015, a call for speakers went out to various people around Aberdeen, UK, for the largest IT event on the calendar, Oil & Gas ICT, hosted by Scot-Tech.

I had a story to tell, and lessons to share, and it was right in my own backyard, but I had never done a public speaking event before, sure plenty of presentations and whiteboard sessions within the confines of company walls, but not subject to public scrutiny by your peers.

So I fired back the mail, with a rough guide of what I’d talk about, to Pete from Scot-Tech, and he’d seen similar talks before so he was keen to include it.

Over the next few months, I got more information through. It wouldn’t just be a public talk, it would be a panel Q&A session too, again another first, would someone ask a question I couldn’t answer and look foolish?

Preparation – an Introduction

I know plenty of others who have given public talks and attended many events on different subjects, seeing many different speakers and their different styles. I knew what was engaging & what wasn’t.

I had also seen various different advice, from rehearsing only the first 5 minutes, to the whole talk, to taking part in stand up comedy boot camps.

VMware run their own events around the world often and have a preference for community speakers over just vendor talks. One of their community evangelists, Mike Laverick, started up a FeedForward program, where if you agree to talk, he will support and guide you, giving critical feedback on your talk and slide decks etc.

I knew a few people who had benefited from Mike’s support, Nick Furnell, and Alan Norrie (a fellow Aberdonian, who gave a talk at the Scottish VMUG, and also talked at this same event).

However, as this wasn’t a VMware event, and Mike was no longer at VMware, this isn’t something I could take up, but it helped highlight the benefits of getting others to review & assist.

However, a friend was also giving a talk at the same event, his first one also, so we worked through the prep together, this didn’t just help for practical preparation, but moral support also!

Practical Preparation

Over the years, I’d collected several books on public speaking, I’d like to say I went back and read those cover to cover, but I felt I needed something more interactive, so I signed up to several video-based courses on Udemy (although I did end up buying one extra book in 2016).

I could watch the Udemy videos at home and listen to the audio on the commute into work, and they helped me both in the creation of the slide decks, the structure of the talk, and how to avoid any fears or nervousness. I learned many valuable lessons.

One of the most valuable was just one chapter (1) from one audio book, The Power of Now. I didn’t listen to the rest of the book, as it was pretty intense and I didn’t want to mess with my mind too much, but there was one tip in there that would keep the nerves in check leading up to the talk itself:

If you find yourself questioning yourself, being negative, that’s just your brain. It’s not you. Tell it to be quiet.

Given I was going to stand in front of 100+ peers, I also invested in looking smart, new shoes etc, and got contact lenses for the first time in my life.


Crush Your Fear of Public Speaking With Ease!

Presentation Bootcamp: Hands-On Content Creation Skills

Presentation Bootcamp: Hands-On Presentation Delivery Skills

Enhance Your Speaking Voice



How to be Brilliant at Public Speaking: Any Audience. Any Situation


 How to Deliver a TED Talk


Do You Talk Funny?


Get Your Public Speaking Mojo Back

Creating the Slide Deck

The main lesson learned from the Udemy courses was the talk is for the audience. It’s not for you. It’s not about you. It’s only about meeting the needs and wants of the specific audience in front of you. What will keep them engaged and not fall asleep or want to walk out of the room.

There is a specific course just on slide deck creation (Presentation Bootcamp: Hands-On Content Creation Skills) and would recommend following that if you need to put one together for the first time.

I would like to say my deck covered all the lesson’s, but it didn’t, pressures of time and other priorities saw to that, but it’s better than it could have been.

The day before

The conference was a 2 day event, so in the morning I attended and talked to several people, to get a feel for what people were talking about, what they wanted to know, how they were feeling, so that I could fine tune my talk.

The morning of the talk

In the morning, I printed out my final version of the slide deck and supportive notes, and did the first run through. Some nerves and heart palpations quickly got under control through some of the techniques above, including taking deep breaths. I took a break, put in my contacts, ran through it a few more times, focusing especially on the start, then walked out the door to drive to the conference.

I met up with some friends and colleagues from work and chatted with them and toured around the vendor stalls and talked to them also, quite a few I already knew. At the Simplivity stand, I bumped into one of the people I’d talked to on twitter, Hani El-Qasem who was a veteran public speaker doing 15 a year, who gave me last minute moral encouragement, then into the hall itself.

1 hour before the talk

I saw a photo of the main hall a few weeks prior so wasn’t completely surprised, but when I walked in and saw all the chairs, I didn’t expect the display projection to be so big, it was huge! Luckily, my slide deck didn’t look bad. Phew!

The next 30 minutes was getting introduced to Ray Bugg, and my co-panel presenters and the chair, where we swapped what we would be talking about, our backgrounds etc. I’m glad there was water on the table because I drank a LOT (because it was dry, not because drinking water keeps your voice in good shape, or Hani saying that Tony Blair used to drink lots before a talk so he could wrap it up quick!)

The Chair also asked what my talk was about in one sentence so he could give his introduction, and that was that. Time for us to give our talks.

Chair, Mark Stephen, BBC Broadcaster

Panelist, Doug Webster, Executive IT Consultant and former Global ERP Manager at Shell

Panelist, Morgan Eldred, Research Director, Gartner

5 minutes before my talk

Another panelist was doing his talk before me, and with the introduction that Mark Stephen, the chair, would give when it came to my turn, it quickly became apparent that the very start of my talk was irrelevant. A quick scoring out and re-structuring of my notes, more water, and I was ready.

The talk

What I had learned kicked in and it was pretty much automatic. Make eye contact with some of the audience. Look around at different sections of the audience. I felt it was going ok, apart from getting jumbled at one section, it did actually go ok.

The Q&A Panel

I need not had worried about the Q&A Panel so much, I knew my content and knew industry specifics.

After the talk

My co-panelists, who were experienced public speakers said I did well, shook hands and that was it. It was over.

I was glad that I did it, the biggest feeling was anti-climatic, having spent so long preparing for and building up for this, and now it was over.

I learned a lot throughout the experience that will be beneficial not just for public speaking.

There is still a lot to learn, and I’ll keep learning for the next time, but I’ll know next time I’ll be able to do better, and that’s what it’s all about.

I shared my story, what I knew, I know I could have done better, but I didn’t crash and burn, and I’ll be much better the next time for having done this.

Lessons Learned

If I were to do it again, I’d also do the following:

Set myself personal deadlines. There’s nothing like a deadline to force you to get stuff done.

Practice more. Practice so much that you don’t need any notes. So you can focus 100% on delivery and audience reaction.


Thanks go to Simon Wardley who gave me the ok to include his content.

More Pictures

Ian Noble @ Oil Gas ICT

Ian Noble @ Oil Gas ICT

My day at the Scottish VMUG, 26 November 2016

The Scottish VMware User Group (VMUG) was held this November at Dynamic Earth in Edinburgh.

It was a great location, with floor to ceiling windows giving lots of natural light and just 15 minutes walk from the central train station.

Dynamic Earth, Edinburgh

(some) of what I learned

unmap – to free up space after storage vmotions and deletions, can be run on some storage arrays with no performance impact (check with the manufacturer how it behaves with your storage – I went back to the office and freed up 1TB of space).

There is less talk about building an internal private cloud to compete with Amazon and Azure (have a read over at Simon Wardley’s blog post ‘What’s wrong with my private cloud …‘).

There isn’t enough talk (yet) on utilising the public cloud.

– VMware VSAN is getting better and better with every release (Overview of 6.1), and has a compelling use case for ROBO deployments, and in harsh environments where solid state storage is preferable to spinning disk.

Even with RAM caching, a PernixData deployment can be made even faster by swapping out the internal SSD layer with faster storage.

IGEL has the best thin client management console I’ve seen to date.

A reminder that Latency & Throughput is what ultimately matters when it comes to storage Input / Ouput performance.

An experienced (ex) user of Puppet in the audience at one session raved about how Ansible was better (summary of differences).

The next Scottish VMUG will be in April 2016.

WLC Control Path Down on a Cisco WIFI Mobility Group


The guest WIFI cannot be connected to.

When you log onto the Cisco Wireless Lan (WLC) controller, under the mobility group status it shows a “Control Path Down”.

No other related outages are currently active, configuration changes have been not made on the WLC or to firewall rules (review if they have), and the guest WIFI has been operational previously.

WLC Control Path Down

Possible Solutions:

1) Log onto the firewall between the internal and DMZ WLC, locate the firewall active session table, search for sessions with source = the internal WLC IP Address & destination the anchor WLC IP address.

If any are active, close those specific sessions.

How to achieve this varies from firewall to firewall.

If this does not resolve, move onto the next possible solution:

2) Reboot the guest WIFI WLC, if this does not change the mobility anchor back online, arrange for and then also reboot the internal WLC (the latter will result in the WIFI network going offline if unless you have high availability solution).

The #VMunderground panel sessions at #VMworld are available to watch live

#VMunderground – “A series of panel sessions ‘by the community, for the community.'”

An impressive list of panelists discussing Storage, Automation, Data Protection, Infrastructure, Networking & Careers.

Even if you aren’t registered for VMworld, you can still watch them live here:


A quick way to convert CPU Ready summation values to % in VMware vSphere

Although I have vRealize Operations manager to tell me if my VMware cluster is getting stressed or not, I wanted to quickly check what the values in the vSphere client were telling me.

In the client, you can select to view different time intervals, realtime, past day, week, month & year, but as the summation calculation is different, to convert into a more useful % CPU ready value requires a formula. However, some genius has created a website that takes all that work away.

Match the chart type to the same as the one you are looking at, type in the value, and it will tell you the figure instantly (it will even tell you if the value is an issue or not – though note that VDI workloads have a higher CPU ready tolerance than used on the tool).

Website: www.vmcalc.com

Thank you to whoever created it!

A story of how Nimble Storage detected two non storage related app issues that other monitoring missed

I have long been a believer of monitoring systems, even back when computer systems were less complex than they are today.

My journey began with NetIQ AppManager, a product designed to monitor Microsoft Operating Systems, and more, by providing a framework into which you could create custom monitoring rules. It was the bees knees of monitoring at the time. I could create a rule to monitor anything and even tied it into a paging system with a bank of modems (that shows how long ago this was!).

Microsoft bought the product, whilst NetIQ also kept their product available. I was faced with do I continue with NetIQ, or do I take up the special promotion from Microsoft to tempt people away to them? Fortunately, that’s what I chose. 

This led to the 2nd generation of monitoring solutions, Microsoft would create all the rules for you, test them, you just had to download, tweak slightly, and suddenly you had thousands of best practice rules monitoring your environment. Gone were the days of manually creating them. Hallelujah!

Then the age of VMware arrived. They had their own monitoring product, Operations Manager, now vRealize Operations. This was another level again, a 3rd generation of monitoring perhaps.

It didn’t just send out point in time email alerts, it crunched all the numbers, it learned my environment, when it was busy, when it was quiet, and in the latest version, it actually uses plain English to tell me there might be an issue. 

I also implemented dedicated networking monitoring solutions (which I will say are 100% beneficial due to the nature of what they monitor).

Along the journey I learned about Riverbed’s SteelCentral product, that offered full end to end visibility & monitoring, from individual mouse clicks linked through to database calls, AppDynamics and many others such as LiquidwareLabs Stratusphere, LoginVSI, VMware’s LogInsight & more, however all these were too expensive when proving a ROI would be difficult.

So, 15 years of utilising monitoring systems. Where on earth does Nimble Storage fit into all this? It’s a storage product right?

Well they advertise this infosight product, which gives various information, in addition to what you can access on the box itself. They added per VM visibility recently, and they demoed heat maps of datastore latency, noisy neighbours etc. 

As a lover of monitoring solutions, having information available, this sounded great, as it is built into their product, I just had to click two buttons and it was working (far easier than my previous experience with monitoring solutions).

What happened next I was unprepared for.  

They have this view (above), it shows top IO users per VM. One day I saw our CRM system was 2nd top. What could a CRM system be possibly using all that IO for? I asked the admin of that product, and unknown to him there was a permissions problem causing the sync engine to fail for some users, and kept retrying. First thought, wow! But I have Microsoft Operations Manager, and VMware Operations Manager, dedicated monitoring products, and neither box highlighted this.

That was a one off I thought. Wrong.

A month later, again an unexpected VM was in the top VM list, this time a print server. Someone must be printing something right? I looked on the Nimble array, at the datastore, and it showed high usage once every 5 minutes, but all quiet on the printer spools datastore. A quick file search for recently modified files showed the Microsoft BITS download files were constantly modifying, a restart of that service and the Configuration manager service, and the datastore activity was back to normal. 

Two application issues detected by Nimble Storage, that would either not have came to light, or not until later when something else failed to work.

In the space of two months, my Nimble Storage product has detected and allowed me to resolve two non storage related issues that my other monitoring solutions did not. Perhaps there is something to learn here. It is not the collection of data that it is important, it is the surfacing of that information and optimising the path through which the message can be received that is.

I can only congratulate Gim Mahasintunan & his team of data scientists for achieving this. You can watch him present his pre-release (now available) demo of storage performance correlation analysis at storage field day below:


Further information:

The Business Value of Per-VM Monitoring in InfoSight http://www.nimblestorage.com/blog/technology/the-business-value-of-per-vm-monitoring-in-infosight/