Posts made by Tom Elliott

Restart the FOGMulticastManager service.

This functionality is now a part o FOG 0.33b as well as checks to ensure hostname is unique.

r1116 released.

Truncates hostname to 15 characters. Also added a service script that verifies that the hostname is valid/unused. This way less confusion between GUI and Fog registration scheme.

r1115 released,

Hostname_early appears to work properly for windows 7 systems. XP already worked, but 7 didn’t. Should be good now.

Still don’t know about the changer from FOGClient, but I’ll try to look into it tomorrow.

That double ='s is not a typo, it’s a comparison operator. Meaning, if the value of: “$bldhcp” is equal to “0”, then continue on. In shell scripting, you can use double == or = as comparison for if statements.

Don’t worry guys, I’m already aware that hostname_early is not working…yet.

I thought it was, I tested the scripts but I just got done testing a job. It seems that I have to export the win7sys variable otherwise the changeHostname function doesn’t know how to mount the registry hive. Why this worked for windows XP, I haven’t the slightest, but I imagine exporting that variable will do the trick as well. I’m testing it right now, will be about 20 minutes or so.

I have CentOS 6.5, Ubuntu 13.10, Ubuntu 12.04, and Debian 7. (Just for testing of course.)

r1114 released.

Had a mistake on checking. $_POST[‘snapin’] should always be null as it’s a file. The vars to check against are: $_POST[‘snapin’] and $_FILES[‘snapin’][‘name’].

r1113 released.

Snapins that already exist on the system are now able to be chosen. Better error management/visability on the coding realm as well.

r1112 released.
Hostname Early should now work properly. I don’t know if the same keys are used for Win 8, but they are working for Win 7. Task refresh rate updated to 5 seconds instead of 10. Limit progress information to active tasks screen. Hopefully you enjoy.

r1111 released. MPS imaging (Multi-Part Single Disk) should work now. MPA always did, but MPS seemed to have issues due to ntfsfix command. It’s unneeded during deploy anyway, only really needed on upload.

While still testing, it should upload/redeploy the EFI and MSR partitions Windows 8 natively seems to generate. Hopefully we’re that much closer to true Windows 8 Support.

if you want to try it download it from:
[url]https://mastacontrola.com/fogboot/init.gz[/url]

I found an issue with mps deploy in that it wouldn’t cycle thru the partitions. I’ve modified the code base so that, from my perspective, it seems to cycle through the partitions again, though I don’t know how well it actually works.

I’ve also added some checks/balances if using windows 8. I’m told the EFI Partition is fat32, so I added a base element to partclone.fat on the second partition if the os is windows 8. There’s a third partition that is blank, but seems to be needed, so I also added a base of code that adds partclone.dd on the third partition if the os is windows 8. Hopefully this will work, but I have no way of knowing as I don’t have windows 8.

I’ll look into that.

I agree it should work all the time, but haven’t figured out where it needs to be changed. The file for this is called service/hostname.php and I’ve added the code base so it should work. If you’re still on 1093, I might have made the change after that so I don’t know. Initially it was throwing an exception rather than just reporting back. If this is still an issue let me know and I’ll try to dedicate to getting that fixed.

Do your hosts have the snapins associated before trying to deploy it?

It seems your /var/www/fog/ OR /var/www/html/fog/ may be corrupted. Has anybody else been “playing” with your fog installation?

I’d also try updating. I’m now at r1110.

If you don’t want any of the tasks there, try this:

TRUNCATE TABLE snapinTasks;
TRUNCATE TABLE snapinJobs;

rado,
Unless there’s a particular reason, can you test out 0.33b code? I got multicast tasks working. I can create multiple tasks, though I haven’t played too much with interfaces. I imagine, however, with my new understanding, it’s relatively easy to setup. The only issue I foresee is the storage of the “2nd” interface would have to be either group and/or host related. This is because in 0.33, you can create individual multicast jobs to a host. I haven’t, yet, figured out how to kill a previously generated task and recreate a new one (while also changing the task and pxe files) to reflect the new change. The port would still work as the original task would have been used.

I believe the issue you were initially running into was because the ports were all the same. It wouldn’t matter, as far as I can tell, if two interfaces were using the same port, which is why you had to create different ports. The same port was causing your issue, not the interface.

I Imagine that you’d only need two interfaces, or more, if you’re trying to separate different segments of the network for testing or some real major need. However, it’d be easier (and you could use the same ports, though it may confuse systems trying to communicate on that/those ports) to setup two separate server rather than trying to make your server perform all the work.

To check which revision you’re at, I’m assuming you’re using SVN?

From the trunk folder perform:
[code]svn info | grep “Revision” | awk ‘{print $2}’[/code]

I think you need to delete all tasks with a stateID of 1,2,3

Something like
DELETE FROM tasks WHERE taskStateID IN (‘1’,‘2’,‘3’);

Then delete the 01- files relating to those tasks.