Sentinel Monitoring Product Blog
We have set up a new checking node in New Zealand and should help make confirmation checks even more accurate of those of you in New Zealand and Australia.
We have more server monitoring nodes being rolled out across the globe in the coming months so stay tuned.
{ 0 comments }
What does this mean? It means that, as well as checking that your mail server is up and running and responding in a timely fashion, we can also check that it is actually authenticating mail accounts correctly.
We also support encryption for the check via SSL or TLS .. or none.
You can add the authentication checking either when you add a new POP3 or IMAP service … or simply by editing an exitsing POP3 or IMAP service.
Just scroll down the advanced options and enter a username and password that Sentinel should use to run the check with. Click ‘save’ and you are done.
Also, it goes without saying, but we recommend that you don’t use your personal email account for this … just for security reasons.
{ 0 comments }
We’ve just released an update to Sentinel which allows you to Acknowledge an alert. This means that if you are aware of a problem and you don’t want Sentinel to keep sending you alerts (who would?) then just log into the Sentinel Dashboard, locate the server/service with the problem and click ‘Acknowledge’.
Sentinel will then supress any further alerts from that service until the service returns to an OK state.
Once it returns to an OK state, the acknowledgement is deleted and alerting continues as normal.
At the moment, you have to log in to the dashboard acknowledge the alert but we hope to be adding the ability to reply via SMS or Email to fire the acknowledgement also.
{ 0 comments }
Tim has been busy over the last couple of weeks and managed to push through some updates to Sentinel including releasing a brand new checking node to our network in the UK.
We’re getting much improved check times from that node and that will certainly help to remove any false alarms due to network congestion from our other nodes.
As well as the new node, Tim has also added;
{ 0 comments }
We’ve had a Sentinel Agent available for Windows & Linux based servers for a while now and we’re really happy to announce that we now have an Agent for Mac OS X 10.6 based servers.
The Mac Agent will allow you to monitor DISK SPACE, LOAD, PROCESSES and NTP and set custom thresholds for each.
Check out the FAQ article here for a bit more information and, if you want to add this functionality, just have a read through the help article inside the Sentinel Dashboard.
{ 0 comments }
We’ve just released an update to the way that we use additional Sentinel nodes to aid in monitoring your servers and websites.
This is the way our checks now work;
1 – Sentinel detects that a problem has occurred with your server.
2 – Sentinel runs a confirmation check from our primary node to confirm the problem
3 – Sentinel then initiates all of our checking nodes and they each run the confirmation check against your server.
4 – The checking nodes report back to the primary Sentinel node and then any alerts are issued.
So this provides a much wider coverage area and helps avoid any potential network problems between our primary node and your server.
It will also allow you to set a threshold. Eg. ‘only alert me if every single node reports a failure’ (once I develop make some changes to the UI).
Another great advantage that this system has is that it allows us to deploy checking nodes very rapidly and locate them near clusters of our customers … which is something that we’ll be doing very shortly.
It wont cost you any more to use these additional checking nodes either and you’ll benefit from them immediately.
If you have any questions please feel free to contact us here.
{ 0 comments }
Sentinel is a brilliant server and monitoring website application … and we’ve just made it even better.
When we launched Sentinel, our aim was (and still is) to keep things really simple. So our dashboard page simply listed all of the servers on your account vertically down the page.
Now, this works really well until you start managing a lot of servers. When that happens you need a couple of things;
Firstly, you need a way to see more on screen and secondly, you need a way to be able to manage these servers and arrange them however you see fit.
Enter the new Grid View!
Note the new icon top right
This is the default Grid View
You can create new groups ...
Change titles & drag and drop servers into groups
Arrange groups however you want
{ 0 comments }
One of our longtime customers uses Zendesk as his help desk product. It’s a great product, in fact we actually use it ourselves.
The problem our customer has is that every time Sentinel sends an alert to his Zendesk email address, Zendesk creates a ticket. So .. When he gets told that a service is having problems a ticket gets created .. and then, when we alert him that the service comes right, another ticket gets created. Read more
{ 1 comment }
A great new feature for Sentinel Server and Website monitoring is our super simple alert API (Application Programming Interface).
Through this API you’ll be able to have Sentinel post information about alerts to third party apps (like your CRM system or your Helpdesk system) or your own systems so you can then run scripts etc. Lots of possibilities. Read more about the Sentinel API
{ 0 comments }
Flap Detection is a process that detects if your server or website is rapidly changing between OK and non-OK states.
If we detect that the service is flapping then we WONT send you an alert for each change of state. If we did, then that could mean that you get swamped with alerts and that is no use to anyone.
Instead. We send you one alert when we detect that flapping has started and another when we detect that it has stopped.
Flap Detection is enabled by default for most services that you add and you can make adjustments to it in the ‘advanced’ properties of the service.
So. How does Flap Detection work?
Flap Detection looks at how your service has performed historically by measuring the last 21 checks of the service.
It then compares the result of its current check with what has happened in the past.
Flap Detection takes two parameters. A high threshold and a low threshold. The high threshold is used to decide when a service has started flapping and the low threshold is used to decide when a service has stopped flapping.
The default values are 20% for the high threshold and 5% for the low threshold. We find that these seem to work out ok for the most part but if you have a particularly finicky service then you can adjust these values accordingly. If your service is prone to going into a flapping state then you can adjust the high threshold up accordingly.
Core upgrade
There is no fancy UI upgrade with this update, just a solid update to our core functionality. Tim and I are passionate about making Sentinel a thoroughly robust server monitoring and website monitoring service and the support for flap detection helps cement that ideal.
As with all Sentinel Server and Website Monitoring upgrades this wont cost you any extra and, by the time you read this, it will automatically be added to your account.
All the best.
Mark Shingleton & Tim Philips
{ 0 comments }