Skip to content

Common Errors

Jump to bottom
SPO edited this page Mar 19, 2019 · 24 revisions

Packaging

Upload fails when trying to create a package

First make sure PHP has been configured to allow upload of large enough files.

In Apache running on a systemd-enabled server, temporary files are not created in /tmp. Make the following changes:

sed -i 's/PrivateTmp=true/PrivateTmp=false/' /lib/systemd/system/apache2.service
systemctl daemon-reload
systemctl restart apache2

Check in /var/log/apache2/error.log for any errors.

An error occurred while talking with GLPI

Inventories are not forwarded to GLPI and the pulse2-inventory-server.log shows:

GlpiProxy> An error occurred while talking with GLPI (details follow) 2018-01-08 09:48:16,980 DEBUG Raw error was: 2018-01-08 09:48:16,980 DEBUG With exception: no element found: line 1, column 0

The response expected is in xml format, but the actual response is not. This happens when for example Apache replies with an error message.

Make sure that Apache is properly configured and that the URLs configured in inventory-server.ini.local correspond to the installation. (eg. https instead of http)

Error -3 while decompressing: incorrect header check

Inventories are not forwarded to GLPI and the pulse2-inventory-server.log shows:

2018-05-02 11:51:22,190 ERROR Failed while decompressing the request from 127.0.0.1. 2018-05-02 11:51:22,195 ERROR Error -3 while decompressing: incorrect header check

The most likely cause is that the setup of GLPI or FusionInventory is not complete. Log on to GLPI and complete the setup.

To check the real error, run

wget http://<pulse_server>/glpi//plugins/fusioninventory/front/plugin_fusioninventory.communication.php

Traceback while deleting a machine: local variable 'sessionwebservice' referenced before assignment

Pulse cannot access GLPI's REST API. To find out what exactly the problem is:

  1. Generate the base64 string containing username:password that is used to access GLPI as defined in /etc/mmc/plugins/glpi.ini.local parameters [webservices]/glpi_username and [webservices]/glpi_password:

echo -n "<glpi_username>:<glpi_password>" |base64

  1. Try to connect to the API using the base64 string generated above and the API URL as defined in /etc/mmc/plugins/glpi.ini.local parameters [webservices]/glpi_base_url:

curl -X GET -H 'Content-Type: application/json' -H "Authorization: Basic <base64_string_generated>" '<glpi_rest_api_url>/initSession'

  1. Check the error received and configure GLPI accordingly

XMPP incorrect password error

Machine agent does not configure itself and an error of the following type occurs in xmpp-agent.log

<presence from="configmaster@conference.pulse/w10-1-rah_9c" to="conf0iaujoggrr@pulse/conf6t0axve0fo" type="error" xml:lang="en"> <x xmlns="http://jabber.org/protocol/muc"> <password>Hv1DMmBFBhhS</password> <history maxchars="0"/> </x> <error code="401" type="auth"> <not-authorized xmlns="urn:ietf:params:xml:ns:xmpp-stanzas"/> <text xmlns="urn:ietf:params:xml:ns:xmpp-stanzas">Incorrect password</text> </error> <x xmlns="http://jabber.org/protocol/muc#user"> <item/> </x> <nick xmlns="http://jabber.org/protocol/nick"/> </presence>

This usually happens when a server has been migrated and the initial configuration does not have the same password. Make sure the password in ejabberd.yml in the following section is correct: mod_muc / default_room_options / password To resolve it, stop mmc-agent and destroy the room named configmaster

systemctl stop mmc-agent

ejabberdctl destroy_room configmaster conference.pulse

systemctl start mmc-agent

ERROR Executing plugin guacamoleconf No module named plugin_guacamoleconf

mmc-agent.log shows the above log. This means that the relay server is not running. To check this, run the following SQL query on xmppmaster:

SELECT * FROM xmppmaster.machines WHERE agenttype='relayserver';

When a relay server is present, the query should list it. Check the xmpp-agent.log on the relay server to determine the cause.

Clone this wiki locally