bytemark-healthcheck (2.6.19ubuntu1~12465.gbp4bf7ba) UNRELEASED; urgency=medium

  ** SNAPSHOT build @4bf7ba90210e59d1cc7a70e4e0dd2eec7a0a48f5 **

  [ Steve Kemp ]
  * Added tests for NTP-handling.
  * Rubocop fixes
  * Enable the tests for the `bad_ipv6` check.
  * Updated the code-comment.
  * rubycop fixup for the check.
  * Abort if we don't find a suitable `get-ntp-offset`.
  * Handle integer conversion more sanely.
  * Updated to resolve the syntax-error.
  * We expect a float to be output, not an integer
  * Updated alert title/detail.
  * Updated comment regarding expected output.
  * Re-enable tests for bad_ipv6 flags.
  * Ensure that we can parse our output into a float.
  * Added a test that we get an exception when non-float output is made.
  * Ensure that we use absolute time for checking.

  [ root ]

 -- Steve Kemp <steve@steve.org.uk>  Wed, 26 Jul 2017 07:36:23 +0000

bytemark-healthcheck (2.6.19) stable; urgency=high

  * Updated to ensure that NTP-checking doesn't have an error with round

 -- Steve Kemp <steve@bytemark.co.uk>  Tue, 26 Jul 2017 09:00:09 +0200

bytemark-healthcheck (2.6.18) stable; urgency=low

  * NTP-monitoring is IPv4-only, catch exceptions on bigvint.

 -- Steve Kemp <steve@bytemark.co.uk>  Tue, 25 Jul 2017 12:44:21 +0200

bytemark-healthcheck (2.6.17) stable; urgency=low

  * Added NTP-monitoring.

 -- Steve Kemp <steve@bytemark.co.uk>  Mon, 24 Jul 2017 12:55:21 +0200

bytemark-healthcheck (2.6.16) stable; urgency=low

  * Ignore aufs-filesystems when checking for free disk-space.

 -- Steve Kemp <steve@bytemark.co.uk>  Wed, 28 Jun 2017 08:55:08 +0200

bytemark-healthcheck (2.6.15) stable; urgency=medium

  * ANy output produced will now go to syslog, rather than via email.

 -- Steve Kemp <steve@bytemark.co.uk>  Tue, 2 May 2017 16:05:02 +0200

bytemark-healthcheck (2.6.14) stable; urgency=high

  * Improve error-detection around job-execution.
  * Improve ability to read or ignore bad state files.

 -- Steve Kemp <steve@bytemark.co.uk>  Tue, 2 May 2017 16:05:02 +0200

bytemark-healthcheck (2.6.13) stable; urgency=high

  * Drop the SSD-testing.

 -- Steve Kemp <steve@bytemark.co.uk>  Fri, 17 May 2017 20:55:02 +0000

bytemark-healthcheck (2.6.12) stable; urgency=medium

  * Added SSD-monitoring and explicit dependency upon smartmontools.

 -- Steve Kemp <steve@bytemark.co.uk>  Fri, 17 May 2017 09:22:09 +0000

bytemark-healthcheck (2.6.11) stable; urgency=low

  * Implemented new command-line flag `--clear-all`.

 -- Steve Kemp <steve@bytemark.co.uk>  Thu, 15 Dec 2016 15:22:51 +0000

bytemark-healthcheck (2.6.10) stable; urgency=low

  * New release with a new test for IPv6 duplicate-addresses.

 -- Steve Kemp <steve@bytemark.co.uk>  Mon, 25 Nov 2016 14:00:41 +0000

bytemark-healthcheck (2.6.9) stable; urgency=low

  * Don't alert on upgradable packages which are "held".

 -- Steve Kemp <steve@bytemark.co.uk>  Mon, 14 Nov 2016 14:44:41 +0000

bytemark-healthcheck (2.6.8) stable; urgency=low

  * Handle unattended-upgrades being missing.

 -- Steve Kemp <steve@bytemark.co.uk>  Fri, 10 Nov 2016 10:22:01 +0000

bytemark-healthcheck (2.6.7) stable; urgency=low

  * MegaRAID: Don't alert for GHS (Global-hot-spare) drives.

 -- Steve Kemp <steve@bytemark.co.uk>  Fri, 16 Sep 2016 13:00:31 +0000

bytemark-healthcheck (2.6.6) stable; urgency=low

  * New release to include the MegaRAID spares-checking.

 -- Steve Kemp <steve@bytemark.co.uk>  Wed, 14 Sep 2016 14:22:41 +0000

bytemark-healthcheck (2.6.5) stable; urgency=low

  * Added error-detection for spare drives in software RAID.
  * Raise the threshold for memory-scrubbing errors. (1 -> 5).

 -- Steve Kemp <steve@bytemark.co.uk>  Thu, 1 Sep 2016 15:22:51 +0000

bytemark-healthcheck (2.6.4) stable; urgency=low

  * Our process-checker will correctly handle malformed UTF-8.

 -- Steve Kemp <steve@bytemark.co.uk>  Fri, 05 Aug 2016 10:00:01 +0000

bytemark-healthcheck (2.6.3) stable; urgency=low

  * Only alert if we see THREE "MCE Memory errors".

 -- Steve Kemp <steve@bytemark.co.uk>  Tue, 05 Jul 2016 11:33:11 +0000

bytemark-healthcheck (2.6.2) stable; urgency=low

  * Catch more errors relating to iterating over processes.

 -- Steve Kemp <steve@bytemark.co.uk>  Thu, 05 May 2016 16:17:18 +0000

bytemark-healthcheck (2.6.1) stable; urgency=low

  * Updated to remove the link to the old update-checker.

 -- Steve Kemp <steve@bytemark.co.uk>  Mon, 25 Apr 2016 12:33:21 +0000

bytemark-healthcheck (2.6) stable; urgency=low

  * New release:
    - Remove the `apt-get update` check; it fails on squeeze.
    - Added test to look for a running syslog process.

 -- Steve Kemp <steve@bytemark.co.uk>  Sat, 23 Apr 2016 09:09:09 +0000

bytemark-healthcheck (2.5) stable; urgency=low

  * New release:
    - Remove the clear-only obsolete heartbeat.
    - Change the raising of the heartbeat alert to account for the time
      taken running the tests.

 -- Steve Kemp <steve@bytemark.co.uk>  Tue, 16 Feb 2016 10:44:11 +0000

bytemark-healthcheck (2.4) stable; urgency=low

  * Simplify our post-install logic.
  * Install a temporary "clear heartbeat" script beneath /etc/cron.d

 -- Steve Kemp <steve@bytemark.co.uk>  Thu, 11 Feb 2016 06:33:06 +0000

bytemark-healthcheck (2.3) stable; urgency=low

  * Changed to raising a heartbeat after four minutes.
  * Remove obsolete /etc/cron.d/byteamrk-heartbeat file, if present.

 -- Steve Kemp <steve@bytemark.co.uk>  Wed, 10 Feb 2016 15:44:51 +0000

bytemark-healthcheck (2.2) stable; urgency=low

  * Changed to native-package.

 -- Steve Kemp <steve@bytemark.co.uk>  Tue, 19 Jan 2016 16:16:16 +0000

bytemark-healthcheck (2.1) stable; urgency=low

  * Bump a new release.

 -- Steve Kemp <steve@bytemark.co.uk>  Tue, 19 Jan 2016 16:00:16 +0000

bytemark-healthcheck (2.0-4) stable; urgency=low

  * A new release to include James Lawrie's postfix-mailq check-changes.

 -- Steve Kemp <steve@bytemark.co.uk>  Tue, 19 Jan 2016 15:33:51 +0000

bytemark-healthcheck (2.0-3) stable; urgency=low

  * The `apt-update_check` is now much simpler.

 -- Steve Kemp <steve@bytemark.co.uk>  Mon, 19 Oct 2015 10:44:01 +0000

bytemark-healthcheck (2.0-2) stable; urgency=low

  * The `apt-upgrade_check` only alerts if there are pending security
    upgrades, rather than all pending package upgrades.

 -- Steve Kemp <steve@bytemark.co.uk>  Wed, 14 Oct 2015 14:33:41 +0000

bytemark-healthcheck (2.0-1) stable; urgency=low

  * New release.
  * Only alert if >1 package upgrade is pending.

 -- Steve Kemp <steve@bytemark.co.uk>  Tue, 13 Oct 2015 14:33:41 +0000

bytemark-healthcheck (1.9-20) stable; urgency=low

  * Fixed the spelling on the pending-package-upgrades alert.

 -- Steve Kemp <steve@bytemark.co.uk>  Tue, 13 Oct 2015 13:44:31 +0000

bytemark-healthcheck (1.9-19) stable; urgency=low

  * Add a check on `apt-get upgrade` to reveal pending package upgrades.

 -- Steve Kemp <steve@bytemark.co.uk>  Tue, 13 Oct 2015 13:33:31 +0000

bytemark-healthcheck (1.9-18) stable; urgency=low

  * Add a check on `apt-get update` age.

 -- Steve Kemp <steve@bytemark.co.uk>  Tue, 13 Oct 2015 09:44:09 +0000

bytemark-healthcheck (1.9-17) stable; urgency=low

  * A final update on raising alerts for MCE - show the first match
    in the summary.

 -- Steve Kemp <steve@bytemark.co.uk>  Tue, 18 Aug 2015 13:00:00 +0000

bytemark-healthcheck (1.9-16) stable; urgency=low

  * Updated MCE-error-handling, to show more detail in the detail-field.

 -- Steve Kemp <steve@bytemark.co.uk>  Tue, 18 Aug 2015 12:55:21 +0000

bytemark-healthcheck (1.9-15) stable; urgency=low

  * The MCE errors are only alerted in a human-readable sense, not
    with the lines that triggered the event - because that is too much.

 -- Steve Kemp <steve@bytemark.co.uk>  Tue, 18 Aug 2015 12:00:21 +0000

bytemark-healthcheck (1.9-14) stable; urgency=low

  * Avoid alerting on numerous MCE errors. (i.e cap at ten)

 -- Steve Kemp <steve@bytemark.co.uk>  Fri, 14 Aug 2015 12:00:02 +0000

bytemark-healthcheck (1.9-13) stable; urgency=low

  * Added new check for DIMM failures.

 -- Steve Kemp <steve@bytemark.co.uk>  Fri, 14 Aug 2015 11:55:11 +0000

bytemark-healthcheck (1.9-12) stable; urgency=low

  * Removed the NIC-checker

 -- Steve Kemp <steve@bytemark.co.uk>  Tue, 21 Jul 2015 16:16:16 +0000

bytemark-healthcheck (1.9-11) stable; urgency=low

  * Updated the NIC-checker to alert only on >=8192 errors.

 -- Steve Kemp <steve@bytemark.co.uk>  Tue, 21 Jul 2015 13:14:15 +0000

bytemark-healthcheck (1.9-10) stable; urgency=high

  * Updated the handling of battery-related errors for hosts running
   the HP-RAID cards.

 -- Steve Kemp <steve@bytemark.co.uk>  Tue, 21 Jul 2015 10:11:12 +0000

bytemark-healthcheck (1.9-9) stable; urgency=high

  * Check for errors in TX/RX for all known networking-devices.

 -- Steve Kemp <steve@bytemark.co.uk>  Wed, 15 Jul 2015 16:55:16 +0000

bytemark-healthcheck (1.9-8) stable; urgency=high

  * Load prior events correctly.

 -- Steve Kemp <steve@bytemark.co.uk>  Thu, 09 Jul 2015 13:55:55 +0000

bytemark-healthcheck (1.9-7) stable; urgency=low

  * Bump the heartbeat alert-time, and give the failures a unique ID.

 -- Steve Kemp <steve@bytemark.co.uk>  Thu, 09 Jul 2015 13:33:31 +0000

bytemark-healthcheck (1.9-6) stable; urgency=low

  * Fully-qualify the path to `hpacucli`.

 -- Steve Kemp <steve@bytemark.co.uk>  Thu, 09 Jul 2015 13:14:15 +0000

bytemark-healthcheck (1.9-5) stable; urgency=low

  * New release specifically to set a PATH for the hp_hardware_raid test.

 -- Steve Kemp <steve@bytemark.co.uk>  Thu, 09 Jul 2015 09:22:09 +0000

bytemark-healthcheck (1.9-4) stable; urgency=high

  * Job-errors will report STDERR if their execution fails.
  * Record failures persistently, such that they require effort to clear.
    This stops flapping.

 -- Steve Kemp <steve@bytemark.co.uk>  Wed, 08 Jul 2015 12:44:21 +0000

bytemark-healthcheck (1.9-3) stable; urgency=high

  * Correctly test all HP Hardware RAID controllers.

 -- Steve Kemp <steve@bytemark.co.uk>  Wed, 08 Jul 2015 10:33:11 +0000

bytemark-healthcheck (1.9-2) stable; urgency=high

  * Correctly handle symlinks in our numbered job-directories.

 -- Steve Kemp <steve@bytemark.co.uk>  Wed, 08 Jul 2015 10:33:01 +0000

bytemark-healthcheck (1.9-1) stable; urgency=high

  * New release, with significant input from Patrick to improve
    reliability and testability.

 -- Steve Kemp <steve@bytemark.co.uk>  Wed, 08 Jul 2015 10:11:01 +0000

bytemark-healthcheck (1.8-4) stable; urgency=high

  * Avoid masking errors with job-running.

 -- Steve Kemp <steve@bytemark.co.uk>  Mon, 06 Jul 2015 09:44:09 +0000

bytemark-healthcheck (1.8-3) stable; urgency=high

  * Correctly cope with SPARE drives for HP-RAID

 -- Steve Kemp <steve@bytemark.co.uk>  Thu, 02 Jul 2015 11:33:11 +0000

bytemark-healthcheck (1.8-2) stable; urgency=high

  * Run jobs more frequently than previously.

 -- Steve Kemp <steve@bytemark.co.uk>  Thu, 02 Jul 2015 11:22:11 +0000

bytemark-healthcheck (1.8-1) stable; urgency=high

  * Updated to use /dev/shm for storing state.

 -- Steve Kemp <steve@bytemark.co.uk>  Tue, 23 Jun 2015 13:00:31 +0000

bytemark-healthcheck (1.7-2) stable; urgency=high

  * Clear more alerts, post-install.

 -- Steve Kemp <steve@bytemark.co.uk>  Tue, 09 Jun 2015 12:00:21 +0000

bytemark-healthcheck (1.7-1) stable; urgency=high

  * Install the MegaRAID check by default.

 -- Steve Kemp <steve@bytemark.co.uk>  Mon, 08 Jun 2015 09:06:03 +0000

bytemark-healthcheck (1.6-1) stable; urgency=low

  * Correctly expire previous alerts - their TTL was being treated
    as seconds, not hours.

 -- Steve Kemp <steve@bytemark.co.uk>  Wed, 03 Jun 2015 09:25:09 +0000

bytemark-healthcheck (1.5-2) stable; urgency=low

  * Check the state of HP RAID batteries.

 -- Steve Kemp <steve@bytemark.co.uk>  Wed, 03 Jun 2015 09:08:07 +0000

bytemark-healthcheck (1.5-1) stable; urgency=low

  * Bugfix release: Attempting to rescue errors in parsing JSON caused
    a bug.

 -- Steve Kemp <steve@bytemark.co.uk>  Tue, 02 Jun 2015 11:44:11 +0000

bytemark-healthcheck (1.4-1) stable; urgency=low

  * Clear the alert that would have been raised by the script we're
    replacing at postinstall-time.

 -- Steve Kemp <steve@bytemark.co.uk>  Mon, 18 May 2015 10:00:01 +0000

bytemark-healthcheck (1.3-1) stable; urgency=low

  * Assuming a script terminates cleanly then all previous alerts it will have
    raised will be auto-cleared.
  * If the alerts are still geneuine/present it is assumed they'll re-raise
    because they'll be re-output.

 -- Steve Kemp <steve@bytemark.co.uk>  Thu, 14 May 2015 10:00:01 +0000

bytemark-healthcheck (1.2-1) stable; urgency=low

  * Updated such that each job has a TTL.

 -- Steve Kemp <steve@bytemark.co.uk>  Wed, 13 May 2015 11:11:11 +0000

bytemark-healthcheck (1.1-1) stable; urgency=low

  * New release with fixed bond-checker.

 -- Steve Kemp <steve@bytemark.co.uk>  Mon, 11 May 2015 15:33:51 +0000

bytemark-healthcheck (1.0-1) stable; urgency=low

  * First public release.

 -- Steve Kemp <steve@bytemark.co.uk>  Fri, 01 May 2015 10:00:01 +0000

bytemark-healthcheck (0.9-1) stable; urgency=low

  * Initial package.

 -- Steve Kemp <steve@bytemark.co.uk>  Wed, 29 Apr 2015 09:09:09 +0000
