Skip to content

Releases: assimilation/assimilation-official

fake release to test attaching binary

05 Sep 23:50
Compare
Choose a tag to compare
Pre-release

test

fake release to test attaching binary

17 Aug 07:04
Compare
Choose a tag to compare
Pre-release

test

the "From The Heart" release

12 Jun 20:44
v1.1.7
Compare
Choose a tag to compare

New Features

  • Added two subgraph queries to assimcli
  • Changed drawwithdot to use subgraph queries - much faster results
  • Added support for Docker containers
  • Added support for Vagrant VMs
  • Updated to Neo4j 3.0.1
  • Updated to Py2neo 2.0.8

Bug Fixes

Fixed an obscure bug in the fileattrs discovery script

Caveats

  • Neo4j 3.x requires Java 1.8 - this eliminates some platforms
  • Not compatible with database formats before 1.1.0
  • No alerting, or interface to existing alerting beyond a sample email script. (hooks to build your own interface are included)
  • high availability option for the CMA is roll-your-own using Pacemaker or similar
  • queries could benefit from more indexes for larger installations.
  • The CMA will suffer performance problems when discovering IP addresses when large numbers of nanoprobes are on a subnet.
  • no GUI
  • Our current process only allows us to create 64-bit binaries. Feel free to build 32-bit binaries yourself. They still work for Ubuntu, and Debian, SuSE and 7.0 and later versions of CentOS.

the DevOpsDaysRox (Rockies) release

20 Apr 15:36
v1.1.6
Compare
Choose a tag to compare

New Features

Bug Fixes

Caveats

  • a few options were shuffled for assimcli query score reporting queries.
  • Not compatible with database formats before 1.1.0
  • No alerting, or interface to existing alerting beyond a sample email script. (hooks to build your own interface are included)
  • high availability option for the CMA is roll-your-own using Pacemaker or similar
  • queries could benefit from more indexes for larger installations.
  • The CMA will suffer performance problems when discovering IP addresses when large numbers of nanoprobes are on a subnet.
  • no GUI
  • Our current process only allows us to create 64-bit binaries. Feel free to build 32-bit binaries yourself. They still work for Ubuntu, and Debian, SuSE and 7.0 and later versions of CentOS.

version 1.1.4 - the April Fools Release (not a joke)

01 Apr 22:59
Compare
Choose a tag to compare

So the joke is that it's not a joke ;-)

New Features

  • Recognize and automatically monitor Oracle
  • Added new argmatch() function for returning a portion of a string matching a () regex
  • Added support for IPv4-only systems (whether disabled by either known method)
  • Added support for systems which create pidfile directories for us
  • README updates
  • Updated documentation to reflect use of GitHub instead of Mercurial (hurray!)
  • Updated basic coding standards documentation
  • Added yet-another-RedHat clone to the installer (and made that process easier)

Bug Fixes

  • corrected the URL to the IT Best Practices project in syslog messages

Caveats

  • Not compatible with database formats before 1.1.0
  • No alerting, or interface to existing alerting beyond a sample email script.(hooks to build your own interface are included)
  • high availability option for the CMA is roll-your-own using Pacemaker or similar
  • queries could benefit from more indexes for larger installations.
  • The CMA will suffer performance problems when discovering IP addresses when large numbers of nanoprobes are on a subnet.
  • no GUI
  • Our current process only allows us to create 64-bit binaries. Feel free to build 32-bit binaries yourself. They still work for Ubuntu, and probably Debian and 7.0 and later versions of CentOS.

1.1.3 - the Leap Day Release

27 Feb 14:03
Compare
Choose a tag to compare

New Features

  • Added the drawwithdot command - draw pictures of subsets of the graph data
  • We now work with (and enable) Neo4j authentication
  • Improved LLDP data capture (including LLDP-MED)
  • Added test code for LLDP and CDP packet handling
  • Added best practice scoring system
  • Added three new canned best practice score reports (queries)
  • Added discovery of the contents of /etc/auditd.conf
  • Added auditd.conf best practice rules
  • Added new assimcli subcommand for printing scores summarized by discovery type
  • Added new assimcli subcommand for printing scores summarized by discovery type and hostname
  • Added new assimcli subcommand for printing scores summarized by discovery type and ruleid
    The new subcommands and queries are great for planning security/compliance triage

Bug Fixes

  • Fixed a bug in conversion of JSON floating point numbers
  • Made CDP data capture work
  • Made discovery code obey timeout and warntime directives
  • Fixed checksum program default configuration

Caveats

  • Not compatible with database formats before 1.1.0
  • Documentation has not been updated to reflect move to github. No doubt other shortcomings exist as well. Sorry! Please fix and generate a pull request.
  • No alerting, or interface to existing alerting beyond a sample email script.(hooks to build your own interface are included)
  • high availability option for the CMA is roll-your-own using Pacemaker or similar
  • queries could benefit from more indexes for larger installations.
  • The CMA will suffer performance problems when discovering IP addresses when large numbers of nanoprobes are on a subnet.
  • no GUI
  • use with recent versions of Neo4j requires disabling authentication on Neo4j
  • Our current process only allows us to create 64-bit binaries. Feel free to build 32-bit binaries yourself. They still work for Ubuntu, and probably Debian and 7.0 and later versions of CentOS.

Happy 2016 Release

03 Jan 00:19
Compare
Choose a tag to compare

New Features

  • Now produce packages and installer works on openSUSE
  • best practice compliance code now issues warn and unwarn events
  • new code for debugging bad key id problems
  • added overview documentation of cma python files
  • you can now say "make tests" to run tests
  • changed libsodium RPM dependency to not be so particular about the version of libsodium available
  • added support for scientific and scientificfermi linux
  • enabled /etc/sudoers discovery by default

Bug Fixes

  • discovery of /proc/sys now ignores I/O errors (this happens on some newer kernels)
  • linux os discovery won't issue funky messages when lsb_release is not installed
  • assimcli now works with a non-empty database again
  • fixed sudoers command to support += operator

Caveats

  • Not compatible with database formats before 1.1.0
  • Documentation has not been updated to reflect move to github. No doubt other shortcomings exist as well. Sorry! Please fix and generate a pull request.
  • No alerting, or interface to existing alerting beyond a sample email script (hooks to build your own interface are included)
  • high availability option for the CMA is roll-your-own using Pacemaker or similar
  • queries could benefit from more indexes for larger installations.
  • The CMA will suffer performance problems when discovering IP addresses when large numbers of nanoprobes are on a subnet.
  • no GUI
  • use with recent versions of Neo4j requires disabling authentication on Neo4j
  • Our current process only allows us to create 64-bit binaries. Feel free to build 32-bit binaries yourself. They still work for Ubuntu, and probably Debian and 7.0 and later versions of CentOS.

Thanksgiving Release

26 Nov 16:12
Compare
Choose a tag to compare

version 1.1.1 - the Thanksgiving Release

New Features

Bug Fixes

Caveats

  • Not compatible with database formats before 1.1.0
  • Sudoers discovery is disabled for this release - will reappear later on
  • Documentation has not been updated to reflect move to github. No doubt other shortcomings exist as well. -Sorry! Please fix and generate a pull request.
  • No alerting, or interface to existing alerting - sample code to send emails is included
  • high availability option for the CMA is roll-your-own using Pacemaker or similar
  • queries could benefit from more indexes for larger installations.
  • The CMA will suffer performance problems when discovering IP addresses when large numbers of nanoprobes are on a subnet.
  • no GUI
  • use with recent versions of Neo4j requires disabling authentication on Neo4j
  • Best practices alerts currently only come out in syslog - not as events. Sorry!
  • Our current process only allows us to create 64-bit binaries. Feel free to build 32-bit binaries yourself. They still work for Ubuntu, and probably Debian and 7.0 and later versions of CentOS.
  • The magic installer can't install CMAs onto Fedora due to Neo4j dependency issues.

1.1.0 - the MCH release

04 Nov 21:43
Compare
Choose a tag to compare

Special Note This release cannot make use of older databases.

The workaround is to start the CMA one time with the –cleandb flag, and restart all your nanoprobes. Sorry about that.

This release is named after my Father-in-law - whose 94th birthday was 1 November 2015. Happy Birthday!

Bug Fixes

New Features

Caveats

  • Not compatible with previous database formats.
  • Sudoers discovery is disabled for this release - will reappear later on
  • Documentation has not been fully updated to reflect move to github. No doubt other shortcomings exist as well. Sorry! Please fix and generate a pull request.
  • No alerting, or interface to existing alerting (hooks to build your own interface are included)
    high availability option for the CMA is roll-your-own using Pacemaker or similar
    queries could benefit from more indexes for larger installations.
  • The CMA will suffer performance problems when discovering IP addresses when large numbers of nanoprobes are on a subnet.
  • no GUI
  • use with recent versions of Neo4j requires disabling authentication on Neo4j
  • Best practices alerts currently only come out in syslog - not as events. Sorry!
  • Our current process only allows us to distribute 64-bit binaries. Feel free to build 32-bit binaries yourself. They still work for Ubuntu, and probably Debian and 7.0 and later versions of CentOS.
  • The magic installer can't install CMAs onto Fedora.

Pre-Columbus-day release (bug fix only)

08 Oct 12:58
Compare
Choose a tag to compare

This is a bug-fix-only release of things discovered once we got more people to install it with the easy-installer in 1.0.1.

Bug Fixes

  • Added net-tools dependency for CentOS >= 7
  • Worked around brain-dead-bug in systemd
  • Fixed bug around iterable Drone objects which caused some discovery to be ignored.
  • Fixed broken links on the web site
  • Created directories for nanoprobe pid file
  • Fixed ldconfig typo in RPM packages
  • Added '.' character as permissible system name
  • Increased maximum system name length
  • Disabled sudoers discovery to avoid periodic errors

New Features

none

Caveats

  • Sudoers discovery is disabled for this release - will reappear later on
  • No alerting, or interface to existing alerting (hooks to build your own interface are included)
  • high availability option for the CMA is roll-your-own using Pacemaker or similar
  • queries could benefit from more indexes for larger installations.
  • The CMA will suffer performance problems when discovering IP addresses when large numbers of nanoprobes are on a subnet.
  • no GUI
  • use with recent versions of Neo4j requires disabling authentication on Neo4j
  • performance with Neo4j is poor. Strangely, it's not a scalability problem. Fixes will be in a future release.
  • Best practices alerts currently only come out in syslog - not as events. Sorry!
  • Our current process only allows us to distribute 64-bit binaries. Feel free to build 32-bit binaries yourself. They still work for Ubuntu, and probably Debian and 7.0 and later versions of CentOS.
  • The magic installer can't install CMAs onto Fedora.