thruk-agent unit stuck in error state look, failing update-status

Bug #1838548 reported by Zachary Zehring
16
This bug affects 2 people
Affects Status Importance Assigned to Milestone
Thruk Agent Charm
Fix Released
Critical
Alvaro Uria

Bug Description

Have a single thruk-agent that looks to be stuck in an error state loop where it fails hook: update-status. This causes aggregate thruk to lose connection to this instance as it is, I'm assuming, stopping and starting the thruk-agent every 5 mins or so (when hook is triggered). Log is posted below. Output is filled with the same loop.

2019-07-31 17:12:02 DEBUG update-status Hit:1 http://ppa.launchpad.net/canonical-bootstack/thruk/ubuntu bionic InRelease
2019-07-31 17:12:02 DEBUG update-status Ign:2 https://artifacts.elastic.co/packages/5.x/apt stable InRelease
2019-07-31 17:12:02 DEBUG update-status Hit:3 https://artifacts.elastic.co/packages/5.x/apt stable Release
2019-07-31 17:12:02 DEBUG update-status Hit:4 http://archive.ubuntu.com/ubuntu bionic InRelease
2019-07-31 17:12:02 DEBUG update-status Hit:5 http://ppa.launchpad.net/landscape/19.01/ubuntu bionic InRelease
2019-07-31 17:12:02 DEBUG update-status Hit:6 http://archive.ubuntu.com/ubuntu bionic-updates InRelease
2019-07-31 17:12:02 DEBUG update-status Hit:7 http://archive.ubuntu.com/ubuntu bionic-security InRelease
2019-07-31 17:12:02 DEBUG update-status Hit:9 http://ppa.launchpad.net/telegraf-devs/ppa/ubuntu bionic InRelease
2019-07-31 17:12:02 DEBUG update-status Hit:10 http://archive.ubuntu.com/ubuntu bionic-backports InRelease
2019-07-31 17:12:06 DEBUG update-status Reading package lists...
2019-07-31 17:12:07 DEBUG juju-log PGP key found (looks like Radix64 format)
2019-07-31 17:12:07 DEBUG juju-log Importing PGP key from keyserver
2019-07-31 17:12:07 DEBUG update-status Warning: apt-key output should not be parsed (stdout is not a terminal)
2019-07-31 17:12:07 DEBUG update-status Executing: /tmp/apt-key-gpghome.kQWEMfW3Dz/gpg.1.sh --keyserver hkp://keyserver.ubuntu.com:80 --recv-keys 7A207542
2019-07-31 17:12:07 DEBUG update-status gpg: key 4B9A81747A207542: "Launchpad PPA for Canonical Bootstack" not changed
2019-07-31 17:12:07 DEBUG update-status gpg: Total number processed: 1
2019-07-31 17:12:07 DEBUG update-status gpg: unchanged: 1
2019-07-31 17:12:08 DEBUG update-status Ign:1 https://artifacts.elastic.co/packages/5.x/apt stable InRelease
2019-07-31 17:12:08 DEBUG update-status Hit:2 https://artifacts.elastic.co/packages/5.x/apt stable Release
2019-07-31 17:12:08 DEBUG update-status Hit:3 http://archive.ubuntu.com/ubuntu bionic InRelease
2019-07-31 17:12:08 DEBUG update-status Hit:5 http://ppa.launchpad.net/canonical-bootstack/thruk/ubuntu bionic InRelease
2019-07-31 17:12:08 DEBUG update-status Get:6 http://archive.ubuntu.com/ubuntu bionic-updates InRelease [88.7 kB]
2019-07-31 17:12:08 DEBUG update-status Hit:7 http://ppa.launchpad.net/landscape/19.01/ubuntu bionic InRelease
2019-07-31 17:12:08 DEBUG update-status Hit:8 http://ppa.launchpad.net/telegraf-devs/ppa/ubuntu bionic InRelease
2019-07-31 17:12:08 DEBUG update-status Get:9 http://archive.ubuntu.com/ubuntu bionic-security InRelease [88.7 kB]
2019-07-31 17:12:09 DEBUG update-status Get:10 http://archive.ubuntu.com/ubuntu bionic-backports InRelease [74.6 kB]
2019-07-31 17:12:09 DEBUG update-status Fetched 252 kB in 2s (162 kB/s)
2019-07-31 17:12:10 DEBUG update-status Reading package lists...
2019-07-31 17:12:10 INFO juju-log Installing ['thruk', 'pwgen', 'apache2-utils'] with options: ['--option=Dpkg::Options::=--force-confold']
2019-07-31 17:12:11 DEBUG update-status Reading package lists...
2019-07-31 17:12:11 DEBUG update-status Building dependency tree...
2019-07-31 17:12:11 DEBUG update-status Reading state information...
2019-07-31 17:12:11 DEBUG update-status pwgen is already the newest version (2.08-1).
2019-07-31 17:12:11 DEBUG update-status apache2-utils is already the newest version (2.4.29-1ubuntu4.8).
2019-07-31 17:12:11 DEBUG update-status thruk is already the newest version (2.24-2ubuntu6).
2019-07-31 17:12:11 DEBUG update-status 0 upgraded, 0 newly installed, 0 to remove and 91 not upgraded.
2019-07-31 17:12:11 DEBUG update-status active
2019-07-31 17:12:13 DEBUG juju-log Incomplete relation: NEMRelation
2019-07-31 17:12:13 DEBUG update-status inactive
2019-07-31 17:12:13 DEBUG juju-log Incomplete relation: NEMRelation
2019-07-31 17:12:43 INFO juju-log Writing file /etc/thruk/thruk_local.conf root:root 444
2019-07-31 17:12:43 INFO juju-log thruk-agent starting

Related branches

Revision history for this message
Wouter van Bommel (woutervb) wrote :

As an addition. The problem is that every time update-status is called by juju (every 5 minutes) all kind of interaction with the repository is performed.

Due to bug LP#1311056 this will result in an additional commented out deb source file. Causing the update-status call to become slower and slower every time, and it will eventually bring the machine down and/or nagios will stop functioning for that cloud.

It seems most prominent on xenial deployments, but it feels as a waste of cpu cycles to check every 5 minutes of the sources files are okay, to run an 'apt update' and to make sure that the deb packages are installed. These should a 1 time install only imho.

Changed in thruk-agent-charm:
importance: Undecided → Critical
Revision history for this message
Wouter van Bommel (woutervb) wrote :

# added field high

Revision history for this message
Giuseppe Petralia (peppepetra) wrote :
Xav Paice (xavpaice)
Changed in charm-thruk-agent:
status: New → In Progress
assignee: nobody → Xav Paice (xavpaice)
Xav Paice (xavpaice)
Changed in charm-thruk-agent:
assignee: Xav Paice (xavpaice) → nobody
Xav Paice (xavpaice)
Changed in charm-thruk-agent:
status: In Progress → Triaged
Revision history for this message
Alvaro Uria (aluria) wrote :

The existing MP has been tested by multiple people and is waiting on final +1 to merge it. It will land on next stable release of the charm.

Changed in charm-thruk-agent:
assignee: nobody → Alvaro Uria (aluria)
status: Triaged → In Progress
milestone: none → 20.05
Alvaro Uria (aluria)
Changed in charm-thruk-agent:
status: In Progress → Fix Committed
Alvaro Uria (aluria)
Changed in charm-thruk-agent:
status: Fix Committed → Fix Released
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.