forum.opsi.org

Hi again,

I'm going to test out the latest preloginloader - can you describe the changes, though?

The opsiclientd is using wmi to trigger some actions.
The wmi queries will wail if the WMI-Service is not yet running.
We changed the retry behaviour in this cases.

I tried preloginloader 3.4-31 - which doesn't work at all on Vista, as far as I can tell. I get errors in my clientconnect log when using it:
[3] [Oct 15 06:32:26] [opsiclientd] Refusing to set empty value for config value 'block_notifier_command' of section 'opsiclientd_notifier' (opsiclientd.pyo|2294)
[0] [Oct 15 06:32:26] [opsiclientd] Opsiclientd version: 0.5.9 (opsiclientd.pyo|2700)
[0] [Oct 15 06:32:26] [opsiclientd] Commandline: C:\Program Files\opsi.org\preloginloader\opsiclientd.exe (opsiclientd.pyo|2701)
[0] [Oct 15 06:32:26] [opsiclientd] Working directory: C:\Windows\system32 (opsiclientd.pyo|2702)
[2] [Oct 15 06:32:37] [event gui_startup] Ignoring wql param because event type is 'gui startup' (opsiclientd.pyo|214)
[2] [Oct 15 06:32:37] [event wait_for_gui] Ignoring wql param because event type is 'gui startup' (opsiclientd.pyo|214)
[2] [Oct 15 06:33:03] [action_processor_starter.exe] Cannot mount: (1312, 'WNetAddConnection2', 'A specified logon session does not exist. It may already have been terminated.') (Windows.pyo|258)
[1] [Oct 15 06:33:03] [action_processor_starter.exe] Traceback: (Logger.pyo|646)
[1] [Oct 15 06:33:03] [action_processor_starter.exe] line 86 in '<module>' in file 'action_processor_starter.py' (Logger.pyo|646)
[1] [Oct 15 06:33:03] [action_processor_starter.exe] line 259 in 'mount' in file 'OPSI\System\Windows.pyo' (Logger.pyo|646)
[1] [Oct 15 06:33:03] [action_processor_starter.exe] ==>>> Cannot mount: (1312, 'WNetAddConnection2', 'A specified logon session does not exist. It may already have been terminated.') (action_processor_starter.py|99)
[3] [Oct 15 06:33:03] [action_processor_starter.exe] Request to 'https://localhost:4441/opsiclientd' failed: (10053, 'Software caused connection abort'), trying to reconnect (JSONRPC.pyo|319)
[2] [Oct 15 06:33:04] [action_processor_starter.exe] Failed to process action requests: Cannot mount: (1312, 'WNetAddConnection2', 'A specified logon session does not exist. It may already have been terminated.') (action_processor_starter.py|103)


preloginloader 3.4-30 seems to work OK. I'll roll it out and see if that fixes the issue we're hitting.

Unfortunately, we've still got Vista slaves hanging after installing the version 3.4-30 of the preloginloader. Any other suggestions?

Please provide the log file c:\tmp\opsiclientd.log
If the service does not start at all, please have a look at the windows event log.

I'll grab those as soon as I can.

opsiclientd.log looks like this:

[3] [Oct 15 16:33:06] [opsiclientd] Refusing to set empty value for config value 'block_notifier_command' of section 'opsiclientd_notifier' (opsiclientd.pyo|2276)
[0] [Oct 15 16:33:06] [opsiclientd] Opsiclientd version: 0.5.8.1 (opsiclientd.pyo|2671)
[0] [Oct 15 16:33:06] [opsiclientd] Commandline: C:\Program Files\opsi.org\preloginloader\opsiclientd.exe (opsiclientd.pyo|2672)
[0] [Oct 15 16:33:06] [opsiclientd] Working directory: C:\Windows\system32 (opsiclientd.pyo|2673)
[2] [Oct 15 16:33:16] [event gui_startup] Ignoring wql param because event type is 'gui startup' (opsiclientd.pyo|218)
[2] [Oct 15 16:33:16] [event wait_for_gui] Ignoring wql param because event type is 'gui startup' (opsiclientd.pyo|218)

The only thing in the system log that is relevant is an event that says "The opsiclientd service terminated unexpectedly. It has done this 1 time(s)." - it provides no details on this. The system log also has many (at least 50) entries labeled "opsi login blocker". The details looks like this:

Log Name: Application
Source: opsi login blocker
Date: 10/15/2009 4:34:23 PM
Event ID: 0
Task Category: None
Level: Information
Keywords: Classic
User: N/A
Computer: talos-rev2-vs10
Description:
The description for Event ID 0 from source opsi login blocker cannot be found. Either the component that raises this event is not installed on your local computer or the installation is corrupted. You can install or repair the component on the local computer.

If the event originated on another computer, the display information had to be saved with the event.

The following information was included with the event:

C

I'm still trying to dig up more information, but any ideas in the interim would be appreciated.

I was finally able to get a verbose log from a stuck slave. Here it is: http://people.mozilla.org/~bhearsum/mis ... entd.log.2

This machine required a hard reboot before it was usable again.