Serio Blog

Tuesday, 19 May 2009

Serio IT Service View comes with a Plugin for monitoring Linux servers, allowing you to monitor free disk space, free RAM, processor utilisation, and running processes.

This article describes how to set up monitoring of a Linux Server with Serio IT Service View. The specific distribution discussed is Ubuntu, but hopefully you will find it straightforward to adapt the procedure to your own version of Linux.

Part One: Installing snmpd (Part of the Net-SNMP suite)

Serio IT Service View monitors Linux servers through the SNMP Agent, snmpd, which is part of the Net-SNMP suite. snmpd must be installed and running on the Linux server you intend to monitor.

For users of all varieties of Linux, you can find advice on downloading, installing, and configuring Net-SNMP at the following website: www.net-snmp.org.

For Ubuntu users, there is no need to download as snmpd is included as standard in the Ubuntu distribution. To ensure that snmpd is installed on Ubuntu,

1. Open the Synaptic Package Manager (accessed from the 'System' > 'Administration' menu).

2. Select 'All' on the left.

3. In the list on the right locate the snmp and snmpd packages. If these are not yet installed, then mark them for installation and click on 'Apply'.

Part Two: Configuring snmpd

1. The snmpd.conf File

The configuration settings for snmpd are stored in a file called snmpd.conf. On Ubuntu, this file is located in the directory /etc/snmp.

A standard snmpd.conf file is supplied with Serio IT Service View. You'll find it in the 'Helpers' folder, under the folder were Serio IT Service View is installed (usually \Program Files\ITServiceView).

Back up your current snmpd.conf file on your Linux server, and then replace it with the standard one supplied with Serio IT Service View.

You can use the snmpd.conf file supplied with Serio IT Service View as-is, wihout changes. However, if necessary, you can make changes to the snmpd.conf file using a text editor. Use the comments within the file to guide you. There are two cases in which you may need to make changes to snmpd.conf:

a. to set the Community String

For security reasons, you may want to change the Community String within the file (set by default to 'public').

b. to specify the list of processes you would like to monitor

If you want use Serio IT Service View to check if certain processes are running on your Linux server, you need to list these processes in the snmpd.conf file. Add the processes you are interested in to snmpd.conf, using the example and comments within the file to guide you.

2. Command Line Options

You can run snmpd with various command line options (see http://www.net-snmp.org/docs/man/snmpd.html).

For the Ubuntu distribution, the default command line options are specified in the file etc/default/snmp, which you can open with a text editor. The default options are specified in the line beginning 'SNMPDOPTS':

SNMPDOPTS='-Lsd -Lf /dev/null -u snmp -I -smux -p /var/run/snmpd.pid 127.0.0.1'

One change is required here: you need to remove the final '127.0.0.1'. It specifies that snmpd should listen only for requests only on the loopback interface, which means that it will not detect requests from Serio IT Service View. With this change made the line should look like this:

SNMPDOPTS='-Lsd -Lf /dev/null -u snmp -I -smux -p /var/run/snmpd.pid'

Save the file after making your changes.

3. Restart the snmpd Service

After making changes to configuration files, you need to restart the snmpd service. On Ubuntu, you can issue the command:

sudo /etc/init.d/snmpd restart

Part Three: Testing snmp on your Linux Server

It's not necessary to do this step, but it will help you confirm that the SNMP Agent, snmpd, is running smoothly on your Linux server. First check that it is running, by entering the following command at the terminal on your Linux server:

ps -A | grep snmpd

Then try the following command, also from terminal on the Linux server:

snmpwalk  -c public -v 2c localhost dskTable

(Note: If you changed the Community String from 'public' to something else in step 1 above, you need to substitute your new Community String in the command above.)

You should see a long list of results about the disks on your Linux server.

If you don't get any results, or you get a time out or other error message, there's clearly something wrong with the way snmpd is running and you need to resolve that before continuing. In that case, review the steps above to see if you can spot the problem.

Part Four: Adding the Linux Server as a Device in Serio IT Service View

The final step is to add the Linux server as a Device to Serio IT Service View. Do this in the usual way using the New Device wizard.

When using the wizard, you need to specify some SNMP settings. Make sure you specify 2c as the version for SNMP. Also, if you changed the Community String (see Part Two, section 1 above), you will need to specify the Community String you have used.

That's it! You're ready to go online.

Thursday, 30 Apr 2009

Serio IT Service View uses WMI (Windows Management Instrumentation) to monitor Windows 2003/2008 Servers through its standard Plugins.

When you add a new Windows Server device to Serio, you'll be asked for Windows Login details for WMI, and you'll be able to test that WMI can connect to your target server.

If this test fails, you'll receive an error message, which may help you diagnose the problem. If you're still stuck, try working through the following checklist.

1. Starting with the basics, is the Windows Server you're trying to connect to switched on? Can you ping it?

2. Check your credentials. The Windows username and password you supply to connect to WMI must have Administrator rights on the target server. If you're using a domain administrator account, make sure prefix it with the domain name, as in:

mydomain\administrator

3. Is the Remote Procedure Call (RPC) service running on the target server? WMI uses RPC to connect to remote servers. Check that the 'Remote Procedure Call' service is running by looking on the Services panel within Administrative Tools on the Control Panel of the target server.

4. Is the target server Windows 2008? If so, there are a few changes you may need to make to the firewall, DCOM launch and activation privileges, and User Account Control (UAC) settings. It's all explained in this handy Microsoft article:

http://msdn.microsoft.com/en-us/library/aa822854(VS.85).aspx

5. Still stuck? Another option is to bypass WMI altogether and use SNMP to monitor the server with the Serio Inventory Agent.

Friday, 24 Apr 2009

A study from the US IT trade body CompTIA reveals that a minority of companies (and in the current economic turmoil a decreasing number) have given IT security training to their non IT staff.

Whilst any study from a trade body should be treated with a little caution, this one stands up. Here in the UK it seems to be pretty rare amongst medium to large companies to see any user-community focussed IT security training - which is interesting, as this end-users are probably the portal through which most viruses, trojans and losses of password cases arise.

The situation has been made worse in the past few years by the advent of cheap laptops, wireless access at home (often badly set-up), a profusion of mobile devices, public wireless networks, USB menory devices and more. Most users simply trust they are OK. I wonder how many corporate laptops get used at home on a personal network connection, and on public Wi-Fi networks?

The study shows the the biggest hit from security breaches is to the end-user themselves (33% of security breaches) - i.e., loss of their laptop or other service, and then the loss of productivity that comes from that. Further down the scale (19% of security breaches), but still with significant effect, are effects to corporate services and networks.

A little advice about safe surfing and safe use may pay more dividends than an emphasis on technology and control for preventing security breaches.

Banks are the Same (Mostly)

UK Banks are an example of this when it comes to on-line banking - spending vast sums on server and corporate-side security whilst the personal and small business customer has to sort the other end of the connection, the home or business computer, themselves with wildly varying success. Read the terms and conditions (the smallprint) of your on-line banking provider - you'll find the responsibility is pushed firmly onto you as a consumer, with little in the way of support.

However, one bank does deserve an honourable mention - RBS Group. RBS introduced a system over a year ago that extended the ridiculous and overused password system used by most banks with a requirement to enter a password but then to use something that you possess - a device and a card. Without this, you can't transfer funds and drain an account - meaning even if you suffer a password loss through spyware, the damage that can be inflicted on you is limited.

RBS have gone one better now with the launch of something called Raport. This product is basically there to step in and make the connection between your machine and the bank less vulnerable to malware - and it's offered for free.

I expect other banks will follows the RBS line in the coming year. I also think there is every possibility that they'll start to take more interest in customer computer security in the near future.

Thursday, 15 Jan 2009

A Happy New Year to all Serio users and blog readers!

Apologies for the recent lack of posts. We've all been working hard on the new release of Serio Helpdesk and Serio Service Desk, scheduled for the middle of this year (as well as enjoying the festive break  wink

In this post, I wanted to let you know about some of the new features we've been working on.

First of all the Service Diary/Resource Planner in SerioClient continues to improve, with the ability to colour code appointments in the next release. You will also see an audit action added to history of an Incident when you book or amend an appointment on it. And will be able to view the Service Diary in month as well as week view.

Mobile Agents will be able to access their appointments in the Service Diary through PocketSerio-i.

Extended Data are extra custom fields (which you define) that you can add to Incidents, Problems, Changes, or Service Requests in SerioClient. The new version will include some enhancements to make Extended Data more useful and accessible, especially in the realm of Change Management.

For example, if have added Extended Data memo fields to your Change Requests to capture, say, the backout plan or impact assessment, you will be able to view this critical data directly in the Service Explorer by expanding the Change Request. You will also be able also search on Extended Data attributes, add them to Issue Displays, and creating custom reports that include Extended Data will be much simpler.

SerioClient Express, the web based client, will catch up with the new tabbed interface that we added to SerioClient in version 5, which judging by the comments we've received, has greatly enhanced navigation between chapters.

SerioClient Express will also now give you access to the Serio Inbox, giving you easy access your support emails and attachments from wherever you are.

The SerioClient Express browser window will allow resizing so you can take better advantage of your screen space. And of course we'll be rolling up the performance enhancements and other improvements to SerioClient Express from the Service Pack for version 5 into the new release.

I'll be posting further articles about the new release later, including news of a new Serio product in the pipeline, and further enhancements to SerioClient and SerioAdmin.

Wednesday, 26 Nov 2008

For Serio Release 5 users, we've issued an updated Service Pack.

The main change is some improvements to SerioClient Express - it's now much faster in almost all aspects of operation. We've also addressed bugs and other things that have come up in Release 5.

To download the latest pack, simply log into the Serio support website (www.serio-support.com) and follow the links on the welcome page. Installation is pretty easy - just update the files that need updating (instructions are available in the Service Pack).

Thursday, 13 Nov 2008

This is a follow up to Are we all going to Disappear in a Cloud?

Some of the factors that will move applications used by SMEs into the Cloud.

+ Cost. For me, the most important factor is cost - if companies can see they'll save money, then more and more services will be Cloud-based... ...

but, I think the issue is complicated by the fact that a large proportion of companies don't actually know how much individual services cost. They can probably obtain a few invoices - for instance, for their Exchange license in the case of an eMail server - but my guess is many won't know the other costs (staff costs, hardware depreciation costs, costs of running a machine in their server room, costs of backups and so on).

So in order for cost to be a motivating factor then the savings must be both obvious and compelling - for instance, less than their annual license cost.

+ Availability. Another factor is the promise of better availability (though with caveats - see this post). High availability is, in theory, something which can be provided much more easily and cheaply by big, consolidated providers - with the understanding you'll need reliable and redundant Internet connections.

+ Skills. Where skills to run a certain service are difficult to obtain or are expensive, or where IT Directors feel they are dependent on a few key employees, this will provide a considerable push to remove that dependency by 'Clouding' the service.

+ ITIL V3.Having gone through a recent refresh, ITIL has reorganised itself to focus on services as we've blogged about here. It would be surprising if this focus, and the resulting demand for improved availability and capacity, did not act as a catalyst for change.

+ Users. Users of services will start to realise they can buy services they need directly from vendors, bypassing the IT department. It presents a huge challenge to the traditional control exercised centrally by the IT function (something we've mentioned before). I'll leave you to decide if this is a Good Thing or not.

But there are some pretty significant factors that will slow down (but not stop) this move as well.

- Trust. You have to

  • trust that your provider isn't going to go bust and is paying their bills
  • trust that they have redundant infrastructure
  • trust that they take regular backups and know how to restore those backups
  • trust that they will look after your data, because they hold it - you don't
  • trust that their security is pretty good

That's a lot of things to take on trust, especially as these are the things that are junked when companies run low on funds. However, I don't see this as something that will stop Cloud-based services - it's more of a cultural thing. Once companies start using these kinds of services and start using them, this will become less and less of an issue as it gets less and less attention.

- Reliance on the Internet. How many SMEs have truly redundant Internet access? Probably not many, but if you move mission-critical applications this is something you're going to have to look at.

- Integration. If you have System A which is integrated in some way with System B, it can be something with will no longer work when you move either or both to the Cloud (the area of integration is something not handled well in my view by Cloud end-user applications, in my view).

- Culture. Possibly the culture of local provision is well established and difficult to change.

So having asked the question 'are we all going to disappear in a cloud' I'll offer my own observations.

I don't think we (IT service and support staff) are going to disappear BUT I think that the role of the Helpdesk/Service Desk is going to change from (primarily) a provider of services to a manager of services - and that this will, at some point, have an impact both on staff headcount and staff skills.

Out: Detailed operating system and configuration skills
Out: Provision of service skills, and a provision of service focus
 
In: Contingency planning
In: Vendor relationship management (time to read up on those old Supplier Management posts).
In: Network provision and performance, as this is currently how Cloud services are delivered.
In: Monitoring and reporting of end-user experience and availability

Friday, 07 Nov 2008

I've spoken to a couple of people today who are planning to take the ITIL V3 Manager Bridge Exam (I blogged about exams back in 2007).

Vinod Agrasala has recently taken (and passed) the exam and has written a few tips for those taking the exam.

Is someone in their 40's old for the IT industry? Maybe. Whatever the answer to that is, whenever someone says

'[insert technology here] will revolutionize the way companies use technology and IT departments work'

my immediate response it to yawn and move on to the next article, or switch the TV off. I've seen it all before.

Does anyone remember the Network Computer, and if so have you seen one recently? No neither have I.

How about WAP? BT spent over £1M marketing this rubbish as the next wave of mobile internet.

I still think mobile (handheld) Internet is over-hyped as well. Sat in Central Park NYC a couple of weeks ago I tried to find out what exhibitions were on at the nearby Museum of Natural History and failed, and couldn't get the correct address for Bloomingdales for my wife either. The problem was none of the websites I wanted to visit worked with my Blackberry. They had a ton of Flash and Javascript and were completely inaccessible and unusable with my tiny device (on the other hand, my ASUS eee is a small but perfectly formed piece of usability).

But I didn't want this to turn into a rant.

What I do want to talk about is 'Cloud computing' and how it is going to affect the IT industry over the next few years. In case you've been living in a bunker for the past few years, it is claimed Cloud computing will revolutionize the way companies use technology and IT departments work.

Firstly I'll try to define what 'Cloud Computing' means for business and Helpdesks/Service Desks. Right now you maybe manage an Exchange Server - you provide the server platform, install the software, patch it and install it. With the Cloud paradigm you get another company to most of that for you, delivering just a service, not a server, to you the consumer. Patching, security, Availability Management and Capacity Management, backups, recovery all become the responsibility of the provider of the service. You do a bit of configuration, and that's it. The Cloud part simply refers to an abstraction of delivery - how the service delivered is opaque to us.

Take another example. Maybe you have an Intranet site - one you bought, or one you home baked. A Cloud solution might be to replace it with a Facebook group (I know a company that have done this and it's great).

Right now, provision of almost any type of system from word processing to sales order management to warehousing is available this way (though with varying results, have you tried to use the Google word processor recently? It's pretty poor).

So what does all this mean for the IT department? It's tempting to ask Can you sack your IT department? as Mary Branscombe asked last month, and whilst I don't think something so radical is on the cards my opinion is there is real change coming, it will affect us all, and that this is not another case of over-hype.

I'll follow this post with thoughts and comments next week.

Wednesday, 29 Oct 2008

Ever wondered how you can print colour on a Black & White printer, the dailywtf explains how...

Monday, 27 Oct 2008

I'm prompted to write by this post by Jeff Dray over at Techrepublic on the subject of getting out and about once in a while - working temporarily in other (end-user) departments and locations.

My own experience is it can be very worthwhile, particularly if relations between the Helpdesk/Service Desk are a bit stained at times.

My own experience of this was after I'd taken over an IT Manager type-role at a Call Centre back in the 1990's. The sales team (i.e., those people actually taking orders and enquiries from customers, remember there were no Internet online shops back then) hated the IT group of 8 staff with a passion, and were not slow in showing this to anyone and everyone in the company who would listen. The sales order processing system I inherited worked after a fashion, but had a tendency to lose orders - not many, but enough to cause a steady stream of calls from customers saying 'where's my stuff?'.

The root of the problem was that they felt that their problems and issues were not taken seriously enough, and dealt with quickly enough. On the IT side, the IT team was constantly struggling to keep-up with a constant stream of new campaigns.

It was the sales team that suggested that members of the Service Desk be seconded onto the phones, and to be honest their mood was such I thought it was better not to say no - regardless of how busy we were.

Over the next few weeks each of us took a 3-day stint as a sales representative working on different campaigns.

As it turned out, the sales team were right - we didn't take their issues seriously enough, and we were not dealing with them quickly enough. The problem was that customers, not surprisingly, are not prepared to wait, get agitated and angry - and these had to be handled with the sales team, increasing stress levels all round. In many cases attention tomorrow was no use - it had to be today, often 'right now'.

The interesting thing was that members of IT coming back to their regular jobs invariably said something like 'we've got to do something' or 'no wonder they are always on the warpath', and had a lot of useful suggestions as to how we could improve things until the sales order processing system was replaced at the end of the year.

The extra insight allowed changes in attitude and working that had a quite positive improvement, despite the high pressure nature of the environment.

Pages