Snapin server not found 503 error

[xburnerx00]

@Sebastian-Roth

Thanks for the reply

I checked the log files that were in your signature. httpd error log was showing this:
link text

I noticed at the 14:12:29 time stamp that the FQDN on the server could not be determined. I added servername to the httpd.conf file, and that error seems to have gone away. I then tried again running snapin at 14:50:00 and the pic posted shows the error at 14:51:51. The FOG log file on the host machine is the same error, that I posted in my original post, just different timestamps. My snapin is still not being downloaded to the host machine.

Also, I do not know if it’s relevant but I have a red circle with a white ‘!’ on the host management list in the web gui saying “No such device or address”. I looked in to other posts for a fix on this but the replies made it sound like the red circle is not important to the over all function of FOG.

[Sebastian Roth]

@xburnerx00 Thanks for posting the log file picture. The line about hostname can be ignored. But there are a couple of important lines like “Error dispatching request to”. This is a real issue. It’s strange you seem to be able to use the web UI! On first sight of this errors I’d have expected the PHP-FPM process to be dead. But it’s probably not.

What kind of Linux OS do you have? Please see if you have any kind of /var/log/php*-fpm.log and post contents or picture of that here.

Also, I do not know if it’s relevant but I have a red circle with a white ‘!’ on the host management list in the web gui saying “No such device or address”.

Not related and not an issue. Just means that FOG is not able to contact that client via port 445 (Windows stuff). This online check was added in FOG long time ago and is often failing because of Windows firewall or other means of port blocks.

[xburnerx00]

@Sebastian-Roth
These are the four error logs that were found in/var/log/php-fpm directory:
link text

Linux OS:
CentOS Linux release 7.6.1810 (Core)

I do not know if it matters, but this machine is on real hardware, it is not a virtual machine. Hardware is a standard HP Pro 3500 series.

[Sebastian Roth]

@xburnerx00 Hmm, the logs show some interesting things but from what I see I don’t thinks this will lead us to what is causing this.

First let’s try to tackle this from a different angle. Can you please create another snapin but use a simple one this time - not snapin pack. Just to see if that makes a difference. I kind of doubt it, but let’s see.

As well please try following the log files while the client is trying to pull the snapin, just to see if you get any messages in the logs while this is happening. For that have your client ready and use the command to watch the log files on your FOG server: tail -f /var/log/httpd/error_log /var/log/php-fpm/*.log (now hit ENTER a couple of times so you see when new lines of logging come in and schedule a snapin task for the client - check the client logs as well so you know when it happens)

[xburnerx00]

@Sebastian-Roth
I created a new snapin and ran the command to see the log files live. Only one update to the log came through and I was pressing Enter the whole time.
link text

The task seemed to take a bit longer than before, but eventually the host machine did pick up the .zip file, but did not download the contents.

[Sebastian Roth]

@xburnerx00 Just got an idea. What size is that ZIP file? Possibly it’s “too big” for the server to handle in a certain amount of time and so it quits the connection. Have you tried a normal snapin (not pack) with let’s ay a one mega byte small executable to deploy to a client via snapin? See if that works.

[xburnerx00]

@Sebastian-Roth I went to check the file size of the snapin and that opened a new set of issues. First, I checked on the FOG server itself at /opt/fog/snapins and to my surprise the zip file was at 0 mb size. Then I went to the web interface and checked the properties of the snapin under the ‘info’ tab and it also showed a filesize of 0 also. So, it seems the .zip file never uploaded correctly and I never noticed it.

So, I created a new snap in. In the pics (link text) it shows the program I want and the batch command to execute to install the program. Also included the snapin set up in the FOG web interface. After creating the snapin, I check that there is now a snapin on the server with 48.3 file size and on the web interface it has 46.08 filesize. I try pushing the snapin to the host again, and it still has the same problems. I checked error logs and it’s same errors I posted earlier.

I then did your idea of creating a smaller snapin. I created a ‘install internet shortcut on desktop’. But it still came back with the same error of the previous logs sent.

[Sebastian Roth]

@xburnerx00 Let’s see if increasing the timeout values will help you here. Edit /etc/httpd/conf.d/fog.conf and add the following three lines:

    <Proxy "fcgi://127.0.0.1:9000">
        ProxySet timeout=300
    </Proxy>

To make it look like this:

<VirtualHost *:80>
    <Proxy "fcgi://127.0.0.1:9000">
        ProxySet timeout=300
    </Proxy>
    <FilesMatch "\.php$">
        SetHandler "proxy:fcgi://127.0.0.1:9000/"
    </FilesMatch>
    KeepAlive Off
...

Then systemctl restart httpd to make the change go live.

[xburnerx00]

@Sebastian-Roth
This seems to have worked but the snapin did not execute correctly. The notification from the fog client popped up on the lower right corner of the host saying the task is happening and then it completed, but neither the application was installed and the internet shortcut on the desktop was not placed. I more than likely created the snapins incorrectly, since this is my first time creating them. Can you offer some guidance on that?

[Sebastian Roth]

@xburnerx00 Maybe try the 7-Zip example here: https://wiki.fogproject.org/wiki/index.php?title=Snapin_Examples

I know the examples are not really up to date with 1.5.5 but the basics should still be valid I hope.

[xburnerx00]

@Sebastian-Roth
Ok, Thank you for the help. I will have to research snapins some more. It seems the issue for this post has been resolved. So I will mark that post as the answer. If I have anymore problems I will make a new post.