In AppBeat 1.4.5 we introduced support for running SSH commands where you can trigger monitoring alert by checking result of:
- command exit status code (for example you can simply trigger error if exit status code is not 0)
- checking specific text in standard output (stdout) by running regex expression
- checking specific text in standard error (stderr) by running regex expression
You can add new SSH monitor by clicking "Add new periodic monitor", enter monitor name and from "Agent type" select "Run Secure Shell (SSH) command":
On next step enter SSH hostname or server IP address and type command you would like to run (for example script name or cat /proc/uptime, ...) and specify server authentication. You can use standard username/password authentication or key based authentication (with private key in OpenSSH format).
After you finished "New periodic check" wizard we will immediately start monitoring your SSH server by running your command and checking for expected result.
p.s. if you don't have AppBeat account yet, you can create one for free!
In last few days we upgraded our core scheduling algorithm which assigns agent resources (CPU, network) to your monitoring tasks.
New version now handles agent resources much better and distributes monitoring tasks more equally. This results in much lower average resource usage on our entire pool of monitoring agents.
Here is screenshot of CPU usage on one of our agents where this effect is clearly visible:
We are currently finishing command line tool for basic AppBeat management. Soon you will be able to get your system status, pause / resume / delete services and checks, create new checks.
For example, this is minimum configuration for creating new check:
"Description": "Check created with AppBeat command line tool / API",
AppBeat command line tool will be available for Windows, Linux and Mac.
On September 8th we deployed AppBeat 1.4.3 which brings you more detailed logging for your web monitors. If error occurs we will save server response (HTTP body with header) and send you unique link where you or your support team can view it. Previously you would receive just HTTP status code.
Detailed logs can now help you to diagnose root cause of problem more quickly. They also allow you to keep historic track of incidents (please see note below about how long we keep your detailed logs).
List of all changes are:
- improved logging for all web checks created after August 9, 2016 - when check changes state from good to non-good we now save detailed error message with response body and header (if available). Older web checks will be migrated to new model gradually.
Note: we will keep your detailed server response for limited time, based on your subscription level:
- Free: 1 day
- Basic: 7 days
- Starter: 1 month
- Standard: 3 months
- Advanced: 6 months
- Enterprise: 1 year
- adding or editing checks: you can now select service by entering service name (especially useful if you have large number of services)
- downtime calculation returned error if user previously edited check and changed service
- other minor improvements
We are continuing with our rapid development and last week new version was published (1.4.2).
In this version we introduced new powerful report - Logs & Performance, and published final version of downtime report. Here are some screenshots of new features.
Performance statistics & hourly averaged response time chart for your periodic check:
Filtering monitoring logs:
As you may noticed, we now also allow you to filter logs and calculate downtime reports by "service group". Service groups are optional groups useful especially for larger organizations. It allows you to group any combination of your services into its own group. Those groups can then be used to assign email/SMS contacts for error notifications (contacts can be members of service groups), or they can be used for filtering data in reports.
List of all changes in version 1.4.2:
- improved "Downtime overview" report (added downtime tooltips to timeline, added aggregated check report in service details). It is now out of beta.
- new added "Reports / Logs & Performance" module for more powerful log analysis (you can drill down your data to lowest check level)
- removed removed "Reports / Statistics & Logs" (replaced by new, more powerful reports)
- update simplified "Live status" tiles & optimizations
- update other minor improvements
We are already working on some new useful features for next version! Stay tuned...
Yesterday we published new AppBeat version with improved downtime report. We also added service timeline where you can visually track details about downtimes:
List of all changes for version 1.4.1:
- improved "Downtime overview" report (added more filters and options, added uptime/downtime timeline, more accurate service downtime calculation - if multiple checks on same service fail at same time, we now automatically exclude overlapping intervals)
- update "Service groups" tab moved to "Services & Checks"
- removed removed "Uptime" component from "Statistics & Logs" (replaced by more powerful "Downtime overview" report)
bug fix context menu could not be entirely visible if item was clicked in bottom or right part of browser window
- bug fix some dialogs did not display vertical scrollbar if there were too many items
- update other minor improvements
In next version we plan to release final version of downtime report and then we will continue with new "Log and Performance" report, which will be much more improved version of current log report.
It is true! It is so powerful and useful feature that we use it as much as possible.
Here is recent code snippet which uses "Linq magic", resulting in much smaller and more readable code:
Thank you LINQ team!
Yesterday we rolled out new version of AppBeat which brings fully customizable website monitoring (as promised in previous post):
List of all changes:
- more friendly wizard for creating new checks
- update web monitor is now fully customizable (you can post data, change protocol version, IP version, send custom headers, ...)
- update DNS monitor can now accept IPv4 or IPv6 address of DNS server used for querying
- removed "Automated Web Testing" early preview is now removed from UI because we received enough initial feedback. Thanks to all who participated in this preview!
- update other minor improvements
If you find any issues or have questions, please let us know!
We are currently actively testing advanced web monitor which will allow you to:
- select HTTP method (GET, POST, PUT, DELETE, HEAD, TRACE)
- select HTTP protocol version (1.0, 1.1 or newest HTTP 2)
- select IP version which should be used when sending request to server (IPv4 or IPv6)
- specify custom headers (you will be able to set custom user agent, ...)
- define custom payload (for example for JSON POST, ...)
We are making our free monitoring plan even more powerful - now with 50 free checks! Sign up now for free!