Posts made by Ben Warfield

Hi everyone,
I’m getting back into making more images in preparation for our summer re-imaging and want to revisit something I wasn’t successful with last year.

I have an answer file that does 99% of what I want, but one thing I’ve never been able to get to work correctly is the computer name. At this point, imaging is a hands-off process except for one step - after imaging is complete, I have to click Next at the screen that says please enter a name for your computer. I leave the default name of PC filled in because the fog client changes it later, but still, it’s one step I would love to eliminate.

I attached the unattend file I’m currently using, any help would be appreciated, thanks!

[url=“/_imported_xf_attachments/1/1654_unattend.txt?:”]unattend.txt[/url]

Okay, thank you.
Do you mind if I work with your code for a few days to see if I can get AD authentication working? I have another PHP application that works this way and it looks like it may be fairly easy to re-use a function I wrote there.

Hi Fernando,
I am also interested in the LDAP plugin. I have activated the plugin and configured, but now when I try to log in I just get a blank white page. Another question I had was groups. Is there any way for the ldap plugin to look to see if a user is part of an active directory group?

I’m also willing to help test, as this is something that would be useful for us.
Thanks!

It was earlier today, I think I remember seeing 2968. Although it was on several versions before that too.

But I did come across something, it might have something to do with phpmyadmin. Phpmyadmin wouldn’t load after I did the updates. So I re-installed it, and just happened to notice something about it removing mysqldn and then the apache log error file started filling up with fog errors. Then I ran the fog install script, got that working, and phpmyadmin was broken again. I’ll have to look into that more tomorrow when I’m back at work.

I have 2 servers I’ve been testing the fog SVNs on, and with the more recent updates the install always gets to installing php5-mysqlnd and just stops. I’ve let it sit for about 15 minutes, but every time I have to ctrl-c to end the script and then run it again and it completes okay. Anyone else noticing this?

I’ve never used the location plugin, so I can’t speak for that.

I had the failed to connect problem as well and narrowed it down to a MySQL config change. In my.cnf, I had to comment out the bind-address line for 127.0.0.1 and restart mysql. I also had to make sure the server config was correct in Config.class.php. Maybe check those 2 things?

I have the same concern, with someone or even myself deleting things accidentally. One thing I put in place is a cron job that backs up the entire database every night. I have a folder full of .sql files able to be restored if ever needed. User permissions would be wonderful, but an administrator can accidentally be just as dangerous as anyone else

That fixed it. Thanks!

Update: Another task has finished, but this time the page does not come back. I’m seeing a slightly different error in the log:

PHP Fatal error: Call to undefined function get() in /var/www/fog/lib/pages/TaskManagementPage.class.php on line 435, referer: [url]http://10.162.1.212/fog/management/index.php?node=service[/url]

Hi,
I just updated from 1.1.2 to SVN 2928 to see if it fixes multicast on our network. But something else isn’t working quite right… when I have an active task, the task page will not load. The error I see in the apache error.log is this:

PHP Fatal error: Call to undefined function get() in /var/www/fog/lib/pages/TaskManagementPage.class.php on line 435, referer: [url]http://10.162.1.212/fog/management/index.php?node=group&sub=deploy&id=71&type=8[/url]

The page loads fine when the task is complete. Thanks for any help you can give.
Ben

I’m curious, is there any sort of schedule for planned releases? I’m very happy with the updates/fixes in the svn and just wondering when it might officially become version 1.3. Thanks!

I had this same problem a few months ago. The solution was to set the service to delayed start. See this thread: [url]http://fogproject.org/forum/threads/hostname-changer.10622/#post-29029[/url]

I’ll chime in from a networking point of view… yes the broadcast needs to be 192.168.63.255 as that is the highest address in the subnet. With a subnet mask of 255.255.240.0, 192.168.48.255 falls within the valid IP range and will not work as a broadcast. By sending a WOL packet to 48.255, you are only targeting one IP.

I’m running FOG 1.1.2 and just went through this a few days ago with Windows 8.1 Enterprise and it worked for me. In my past experience it seems most of my issues come from the client side, not the FOG server or Active Directory side.

Is the fog client installed and have you verified the settings are correct? Ex: fog server ip/name is correct, hostname changer is enabled…

The fog client creates a log called fog.log in the root of the C drive. Does it give any valuable information?

Update: This is resolved.

There were 2 issues:

• I had the wrong mysql username/password in Config.class.php on the nodes. During the node install, I had given it a generic root account but found out earlier today that I specifically needed the username/password for storage nodes, which I found under the storage node section in settings on the main server.

• Second: I realized the nodes couldn’t even connect to the main database, and I was able to narrow that down to the my.cnf file on the main server. It had the bind address set to 127.0.0.1. I commented that line out, restarted mysql, and then they were able to connect.

After I verified everything was talking correctly, I restarted the FOGImageReplicator service and everything instantly started copying. The fogreplicator.log file on Node1 shows Group #2 as I expected it might.

This also fixed an issue that I never really cared about enough to fix… On the main fog dashboard page under disk information, It always told me “Failed to connect to” when I selected Node1 or Node2. that’s fixed now and I can see the disk usage for them.

Thanks for the help!

Sure, here it is. Some more information too since the log will show some other details: All 3 servers were in the same group at one point and the main server successfully replicated to the other 2 nodes. On October 8th, the 2 nodes ran out of space because they have smaller hard drives than the master. So I decided to take split them out into their own group and have Node1 be the master and have it replicated to Node2 for redundancy.

You’ll see in the log on 10-08-14 11:43:28am is when the drives ran out of space. I removed the nodes from the group at 10-08-14 2:57:35pm, which is when the main server started saying there’s only one member and it doesn’t need to copy anything. While the 2 nodes weren’t part of any group, I manually deleted the images from the /images directory. Then I created a new group and added the nodes to it. I uploaded an image without issue and I see it in the /images directory on node1 but it has never been copied to node2. Side note: When I try to deploy this image, I get an error on the client saying “Unable to locate image store”… Any chance these 2 problems are related?

Tom pointed out above that servers only handle replication within their group, so it makes sense that the log is not showing attempts for group number 2. However, there is no fogreplicator.log file on Node1 to view. When I restart FogImageReplicator, nothing happens.

[url=“/_imported_xf_attachments/1/1421_fogreplicator.txt?:”]fogreplicator.txt[/url]

Yep - here you go:

[url=“/_imported_xf_attachments/1/1418_log-screenshot.png?:”]log-screenshot.png[/url]

oh duh. Okay it appears that service is running now

• Stopping FOG Computer Imaging Solution: FOGImageReplicator
  start-stop-daemon: warning: failed to kill 12557: No such process [ OK ]

• Starting FOG Computer Imaging Solution: FOGImageReplicator [ OK ]

However, still no logs and still no sign of replication happening.

P.S. Thanks for the quick responses!

Ahh, good to know. I was under the impression that the main server handled replication for all groups.

So… knowing that, my next issue is that the FOGImageReplicator will not start on Node A. There are no logs in /opt/fog/log. When I run /etc/init.d/FOGImageReplicator Start, I get this:

• Starting FOG Computer Imaging Solution: FOGImageReplicator [fail]
  I even tried restarting and get this:
• Restarting FOG Computer Imaging Solution: FOGImageReplicator
  start-stop-daemon: warning: failed to kill 12522: No such process
  What might be causing the service to fail?

Hi,
I have a question regarding replication for additional storage groups.

My fog setup consists of 1 main server and 2 storage nodes (let’s call them nodes A and B.) The main server is in a storage group by itself and I don’t care about replication there because it’s a VM that backs up with snapshots. Node A is in a second storage group with the “Is Master Node” enabled. Node B is also in the second storage group.

My problem is that images are not being replicated from Node A to Node B. Am I thinking correctly that that’s the way this should work? I see this in the replication log every 10 minutes:

[COLOR=#555555][10-09-14 1:04:42 pm] * Starting Image Replication.
[10-09-14 1:04:42 pm] * We are group ID: #1
[10-09-14 1:04:42 pm] * We have node ID: #1
[10-09-14 1:04:42 pm] * I am the only member, no need to copy anything!.[/COLOR]
[COLOR=#555555] [/COLOR]
It’s almost like the replication service doesn’t know about the second storage group. Thoughts?

Running fog 1.1.2 (recently downgraded from 1.2.0 because of problems with imaging XP)

Thanks!