datadog dpkg returning error, breaking apt-get

[ssbarnea]

root@info:~# apt-get install ldap-utils
Reading package lists... Done
Building dependency tree
Reading state information... Done
The following NEW packages will be installed
ldap-utils
0 to upgrade, 1 to newly install, 0 to remove and 12 not to upgrade.
1 not fully installed or removed.
Need to get 283 kB of archives.
After this operation, 698 kB of additional disk space will be used.
Get:1 http://gb.archive.ubuntu.com/ubuntu/ precise-updates/main ldap-utils amd64 2.4.28-1.1ubuntu4.4 [283 kB]
Fetched 283 kB in 0s (813 kB/s)
Selecting previously unselected package ldap-utils.
(Reading database ... 198034 files and directories currently installed.)
Unpacking ldap-utils (from .../ldap-utils_2.4.28-1.1ubuntu4.4_amd64.deb) ...
Processing triggers for man-db ...
Setting up datadog-agent (1:5.0.0-496) ...
System start/stop links for /etc/init.d/datadog-agent already exist.
usermod: user dd-agent is currently logged in
dpkg: error processing datadog-agent (--configure):
subprocess installed post-installation script returned error exit status 8
Setting up ldap-utils (2.4.28-1.1ubuntu4.4) ...
E: Sub-process /usr/bin/dpkg returned an error code (1)

[ssbarnea]

Others seem to have same problem https://gist.github.com/moodh/af122aa53cd5d8579bcb

[remh]

Hi @ssbarnea . Thanks for the report.

Did you have a previous version of the agent running ?
Can you look at the running processes:
ps aux | grep dd-agent

And kill them ?

Then try to install the agent:
sudo apt-get install datadog-agent

It should fix your issue.

[ssbarnea]

thanks, this solved the problem. The jmxfetch was still running.

[ssbarnea]

Please reopen this, it seems that there is serious bug around:

root@info:/var/log/datadog# ps aux | grep dd-agent
dd-agent  6218  0.1  0.2 16000124 137244 ?     Sl   Jun06 324:38 java -jar /usr/share/datadog/agent/checks/libs/jmxfetch-0.3.0-jar-with-dependencies.jar --check tomcat.yaml --check_period 15000 --conf_directory /etc/dd-agent/conf.d --log_level INFO --log_location /var/log/datadog/jmxfetch.log --reporter statsd:8125 --status_location /tmp/jmx_status.yaml collect
dd-agent 10363  0.2  0.2 16066628 144428 ?     Sl   May09 432:37 java -jar /usr/share/datadog/agent/checks/libs/jmxfetch-0.3.0-jar-with-dependencies.jar --check tomcat.yaml --check_period 15000 --conf_directory /etc/dd-agent/conf.d --log_level INFO --log_location /var/log/datadog/jmxfetch.log --reporter statsd:8125 --status_location /tmp/jmx_status.yaml collect
root     13592  0.0  0.0   9392   936 pts/74   S+   13:26   0:00 grep --color=auto dd-agent
dd-agent 16101  0.1  0.2 16000124 146296 ?     Sl   Jun30 250:26 java -jar /usr/share/datadog/agent/checks/libs/jmxfetch-0.3.0-jar-with-dependencies.jar --check tomcat.yaml --check_period 15000 --conf_directory /etc/dd-agent/conf.d --log_level INFO --log_location /var/log/datadog/jmxfetch.log --reporter statsd:8125 --status_location /tmp/jmx_status.yaml collect
dd-agent 20020  0.1  0.3 16000108 180360 ?     Sl   Aug27  94:38 java -jar /usr/share/datadog/agent/checks/libs/jmxfetch-0.3.0-jar-with-dependencies.jar --check tomcat.yaml --check_period 15000 --conf_directory /etc/dd-agent/conf.d --log_level INFO --log_location /var/log/datadog/jmxfetch.log --reporter statsd:8125 --status_location /tmp/jmx_status.yaml collect
dd-agent 29460  0.1  0.2 16000108 143096 ?     Sl   Jul29 175:05 java -jar /usr/share/datadog/agent/checks/libs/jmxfetch-0.3.0-jar-with-dependencies.jar --check tomcat.yaml --check_period 15000 --conf_directory /etc/dd-agent/conf.d --log_level INFO --log_location /var/log/datadog/jmxfetch.log --reporter statsd:8125 --status_location /tmp/jmx_status.yaml collect
dd-agent 31411  0.1  0.3 15933544 180172 ?     Sl   Sep10  53:18 java -jar /usr/share/datadog/agent/checks/libs/jmxfetch-0.3.0-jar-with-dependencies.jar --check tomcat.yaml --check_period 15000 --conf_directory /etc/dd-agent/conf.d --log_level INFO --log_location /var/log/datadog/jmxfetch.log --reporter statsd:8125 --status_location /tmp/jmx_status.yaml collect
root     31834  0.1  0.2 16069528 126796 ?     Sl   Sep24  10:33 java -jar /opt/datadog-agent/agent/checks/libs/jmxfetch-0.3.0-jar-with-dependencies.jar --check tomcat.yaml --check_period 15000 --conf_directory /etc/dd-agent/conf.d --log_level INFO --log_location /var/log/datadog/jmxfetch.log --reporter statsd:8125 --status_location /tmp/jmx_status.yaml collect
dd-agent 38975  0.1  0.3 16000124 151332 ?     Sl   Jul31 169:30 java -jar /usr/share/datadog/agent/checks/libs/jmxfetch-0.3.0-jar-with-dependencies.jar --check tomcat.yaml --check_period 15000 --conf_directory /etc/dd-agent/conf.d --log_level INFO --log_location /var/log/datadog/jmxfetch.log --reporter statsd:8125 --status_location /tmp/jmx_status.yaml collect
dd-agent 39873  0.1  0.3 16000108 197292 ?     Sl   Aug17 121:14 java -jar /usr/share/datadog/agent/checks/libs/jmxfetch-0.3.0-jar-with-dependencies.jar --check tomcat.yaml --check_period 15000 --conf_directory /etc/dd-agent/conf.d --log_level INFO --log_location /var/log/datadog/jmxfetch.log --reporter statsd:8125 --status_location /tmp/jmx_status.yaml collect
root     42884  0.5  0.0 288524 24788 ?        S    Sep16 105:10 /opt/datadog-agent/embedded/bin/python /usr/bin/dd-agent restart
``
Clearly something like this should not happen!

[ssbarnea]

I have to kill all these processes just in order to be able to run apt-get autoremove again, that's not right. I have to do this again today after I had a similar problem 3 days ago.

[remh]

Thanks @ssbarnea . Do you mind sending your logs (from /var/log/datadog) by email to support at datadoghq.com ?

Also it looks like that some of the instances of jmxfetch have been running for a while now, which is not related to the latest version of the agent (most jmxfetch instances use the old path /usr/share/...). There was a known bug with a specific version of the agent that had been fixed with Agent 4.2.0.

Can you kill all these instances and see if the issue is still happening with Agent 5.0.0

[remh]

Closing the issue as we didn't receive any clear report of it happening.

I suspect this was caused by the usage of dd-agent restart to restart the agent which is not the proper way to restart the agent. (#1180 )