Posts made by Wayne Workman

In Fedora, there are some systemctl commands to check the status of various services. There’s probably something equivalent in Ubuntu.

If you look at this:

[url]http://fogproject.org/wiki/index.php/FOG_(r2922)Configuration_on_Fedora_21_Workstation_inside_Windows_Server_2012_Hyper-V_using_ProxyDHCP#FOG_Server_Troubleshooting_Help.28Fedora_21.2C_r2922_.26_up.29[/url]

It’ll give examples of those commands in Fedora.
It’s basically to check and see what’s running, what’s not, what has errors…

It’s a place to start.

In that particular revision, it’s possible to change the multicast address & port used. Did you do that, by chance? (just a stab at helping)

[quote=“Wolfbane8653, post: 42382, member: 3362”]It is also possible to install fog on a [URL=‘http://www.raspberrypi.org/’]Raspberry Pi[/URL]. This could be a cheap solution for you. As of now we have only tested and checked that it has installed correctly on a Pi. But the investment is around $50 a Pi and could possibly be a solution for you. The Pi can sit right next to the router and plugin to the same power strip.

Of course we would be willing to help you at every turn.[/quote]

Now that’s a solution!

I vote for this one.

Clearly, you should just buy 1 at first, and we can get that working and go from there.

[quote=“Junkhacker, post: 42375, member: 21583”]there are ways you could make fog work for your situation, but it would take some work. I’ve done something along these lines as an experiment on a feature in development, but you honestly might want to look at different solutions unless you’re convinced fog is the way to go.[/quote]

He hit on the “Cost” factor pretty hard…

[quote=“Alex Elkins, post: 42373, member: 28621”]Okay, I understand your suggestion. Wouldn’t moving the images from the master to the storage node still use up the same amount of bandwidth, but only one time (or however many times you push a new image to the storage node)?

E: Site-to-site VPN or what?[/quote]

I was figuring that since he must go to each site anyways to set up the storage nodes, might as well take a copy of the needed images with him on a flash drive, and copy them where they should be, and make the necessary changes for it to work.

Also, if a transfer from master to storage node is necessary, this can be scheduled during off-peak hours… like over the weekend, overnight on Sunday night… whatever works best. This still has tremendous benefit.

And,

No matter what route you choose, I think you’re going to have to get some sort of VPN going (as [URL=‘http://fogproject.org/forum/members/uncle-frank.28116/’][SIZE=11px][COLOR=#969696]Uncle Frank[/COLOR][/SIZE][/URL] asked about) just for the systems to communicate properly.

Unless, all the machines you put in at the 70 locations are completely independent. This would mean having some sort of remote-desktop software running on them all the time, that works across the internet. TeamViewer wouldn’t be ideal as you need an access code from the target machine for each instance.

[quote=“Alex Elkins, post: 42365, member: 28621”]The only equipment we have laying around (mostly) are computers that we use for backup in bad situations. I will speak with my manager about purchasing smaller stations to use for this.

I guess my biggest misunderstanding is how will I get those computers to connect to my FOG master to get an image? How will everything be connected?

Here’s what I’m thinking:

Master connects to storage nodes
Workstations talk with storage nodes
Master sends reimage request to storage node
Storage node reimages workstations

So, in essence, the storage node handles the PXE/TFTP portion on their end. The master and storage nodes just connect somehow to transfer their information, and I think that is where it gets muddy in my head.

Do the storage nodes handle the PXE/TFTP, or would that need pushed somewhere else? Also, the DHCP server at each location has no knowledge of the FOG master’s network…is that an issue?[/quote]

I updated my first post to point out some other issues.

Others would have to help you with the storage node questions, as I only use one FOG machine at one location currently.
But, I still think your best bet is putting in older machines at each location.

I’m going to take a stab at this question. I’ve been using FOG for about 2 months now, it’s working great.

So, you don’t want a storage node at each location because of cost.
My solution to that is to just put a storage node at each location.

It doesn’t need to be anything fancy. An old dual-core with a gigabit interface will get the job done just fine. You could even go with an old P4 with Hyper Threading. It’d be slower, but still.

Also,
You should probably point out to your superior how much fuel savings AND bandwidth savings you’d have by putting in a storage node at each location. You could even justify it by replacing 70 select machines with new ones, and re-appropriating the old ones as storage nodes.

Imagine the slowness of transferring 10 - 40 GB over the internet? How long would that tie up the various location’s internet service for? A day at least? How much productivity time would be lost due to that? What if the connection hiccups? What if two machines (or more) go down at once? How much bandwidth are you going to use up at your central location?

Others may reply with other ideas, but I think this would be better IMHO.

Image sizes are not reporting correctly for SVN 2920 on Fedora 21.

vsftpd should have all the permissions it needs… probably…

Where should I start?

Oh, well that’s awesome.

I think it’d probably be a good thing if I didn’t have to recompile the FOG Service’s Hostname change module in order to change the encryption passkey for storing the AD password.

I feel like the web GUI should allow you to change the passkey, but not show you the passkey. This way, that data isn’t transmitted any more than it absolutely needs. I.E., sending a new passkey, but never showing the current one, nor retrieving it.

I feel like a change like this (or similar) would improve the robustness of the FOG imaging solution.

I did check /images/dev
The image is not in there.

The image that disappeared is not a new one. It’s one that’s been on the FOG server for probably 3 weeks now. I’ve used that image many times already, including imaging an entire lab of 29. All went well.

I have a suspicion of what caused it… Won’t let off on it yet but, if I am correct, I’ll post here.

I’m going to copy my images from the /images directory to somewhere else from now on. Just in case this happens again; I can simply copy the image back to where it should be.

It’s going to use up space, but I don’t know what else to do to mitigate the problem in the future. I suppose this will use less space than snap shots.

Ok I just figured out the problem… it’s scary, like, really scary…

I went to look at the actual images folder on the FOG server, and this problematic image isn’t there… it’s just… gone…

It was there before… I used it two weeks ago on an entire lab! And it’s still showing up in the Web GUI ?!?

This is crazy. What happened to my image?

So, I made a snapshot of the FOG server as it currently is, so I can troubleshoot later if possible.

I’m going to just press forward and re-make the image. Good thing I had a backup copy in GHOST format.

I created a new storage group called “MB” and added all the images to that group. I then assigned MB to the default storage node, and I deleted the default storage group. Still no luck.

Also, it would seem that it’s not just my host, it’s that particular image. The other images create tasks just fine. It’s just this one particular image that is troubling. It’s weird though, because, it was working fine two weeks ago, and I haven’t changed anything (before it broke).

I don’t get it.

It would appear that all three of my images are assigned to the “default” storage group.

I clicked on “Images” at the top, clicked “List all images”, then clicked the problem image, and clicked “Storage Group” on the left.

It’s got a check mark for the default one, and that’s the only one listed for all three of the images.

So, when I tried to image my workstation today (in order to update my image), I get this error…

“Failed to create deployment tasking for the following hosts
You must first upload an image to create a download task”

But, I have about three images created already on the fog server. They show up under images. And I have the correct image assigned to my workstation.

I can also create tasks for other hosts… just not MY host lol.

I’ve not actually logged into the Fog machine in quite some time, and haven’t changed any settings.
Today, I did some unicast imaging on some laptops earlier and it worked fine.

I tried rebooting the Fog machine, twice.
I’ve tried deleting my host and re-creating it… thought it was just a web GUI issue, so I did the full-registration when booting to the network, and set it up with an image that way, and told it to image. It goes through the paces, and then says “Image not found” or something.

I ran some systemctl status commands, and found only one thing that looked weird.

systemctl status FOGImageReplicator.service -l

it gave about 10 errors, all identical.

PHP Warning: Invalid argument supplied for foreach() in /var/www/html/fog/lib/fog/ImageReplicator.class.php on line 48

I’ve checked the Log Viewer in the Web GUI, and nothing seems out of place.

Using SVN 2920 (code named the “Wayne” version)

Any help is appreciated.

I’d like to be able to right click on the fog client (in the system tray) and have the option to show all client details in a read-only format, and this info come from the FOG database.

I’d like to see the serial number for the machine, especially. I’d also like to see the info in the custom field be displayed too. Along with what image is assigned, what groups the client is a member of, all that good stuff.

Thanks,
Wayne

This issue has been resolved. Tom got me going, big thanks!

I’ll be getting my installation documentation up on the wiki soon.

[quote=“Tom Elliott, post: 40882, member: 7271”][url]http://fogproject.org/wiki/index.php/Upgrade_to_trunk[/url][/quote]

I was able to get the version in Fog_Trunk going.

I saw the new multicast settings, that’s great.

Did have a few problems with NFS service, but I got them sorted out. The installer changed my configuration in /etc/exports and added a line to a directory that didn’t exist. I got rid of that, and set the settings back the way I had them for ProxyDHCP, then restarted NFS and all was well.

After that, I double checked the other files that I made custom changes to and they were all fine. I input my multicast settings & port number. Then gave the FOG server a reboot for good measure.

I created a multicast task to one client (a group with just one machine in it), and tried… and it doesn’t work…

Got the same issue I had before. It gets to the partclone screen and just sits saying

"Starting to restore image (-) to device (/dev/sda1)

inside /opt/fog/log/multicast.log there aren’t any new errors… just old ones from the 13th saying “StorageNode Not found on this system.” I think that error was from when I was still figuring out how to get NFS, and FTP to work right for the imaging process…

Keeping in mind that unicast works for uploading & downloading… My images move from the /images/dev folder to the /images folder properly via FTP (that was a hard one to fix).

Where should I look next? OR, Tom Elliott, would you like to mess with the multicast feature more? I understand you made these changes YESTERDAY, and man are you quick! So there might be a glitch somewhere you overlooked… I’ll be patient, because you guys are awesome.