Latest FOG 0.33b

[fabritrento]

[quote=“Tom Elliott, post: 21678, member: 7271”]What is blkid of the swap partition?[/quote]

sample on my server: /dev/mapper/fog–vg-swap_1: UUID=“82810d84-f19c-40ee-a3e8-fa51d6f6e84f” TYPE=“swap”

(only TYPE is to be used obviously)

[rado]

Hi guys, I don’t wanna disturb you, but just small question how to update between revisions? Just download from svn and then bin/installfog.sh? I would try it myself, but don’t wanna mess my current installation. Thanks

[Tom Elliott]

If you downloaded the svn to do the install, from the trunk directory (or whatever you named it) perform:
[code]svn update[/code]

Then re-run the installer:
[code]cd bin
./installfog.sh[/code]

[Troye Johnson]

Morning Tom,

Question for you is there any way to import and export snapins?

[Tom Elliott]

Unfortunately, not at the moment. I haven’t the time to figure out how to zip all the snapins so you can download, reupload yet.

You can, however, save the entire database under FOG Configuration->Configuration Save, which would, at least, save the snapins created.

[Troye Johnson]

ok yea I dont think that will work because I am trying to go from one server to another in a diffrent location. Just wanted to make sure there was no easier way before I did it manually.

[Tom Elliott]

r1125 Released.

Should fix typo on multicast task creation.

Recodes elements of service scripts. Getting less and less ClientServiceManager.php needs.

[rado]

[quote=“Tom Elliott, post: 21684, member: 7271”]If you downloaded the svn to do the install, from the trunk directory (or whatever you named it) perform:
[code]svn update[/code]

Then re-run the installer:
[code]cd bin
./installfog.sh[/code][/quote]

thanks, worked.

But is multicast working? I tried to create multicast task in latest (1124) revision. Got the same blank screen as previously (see post #504), so changed [I]$MulticasetSessionAssoc->save();[/I] and blank screen was solved. But there was no multicast task in Active Multicast Tasks (only individual tasks in Active Tasks) - this happend even in 1118. And creating task took too long (few minutes)[I] - [/I]I found out, that mysql daemon is causing high load. Then I looked at [I]multicastSessions[/I] table and found circa 60000 tasks with random base ports and -1 msState. So I deleted them manually and now I’m getting another blank screen and in apache error log there is:

[I][CODE]PHP Fatal error: Allowed memory size of 134217728 bytes exhausted (tried to allocate 16 bytes) in /var/www/fog/lib/fog/FOGManagerController.class.php on line 124, referer: http://localhost:8081/fog/management/index.php?node=tasks&sub=listgroups[/CODE][/I]

So, perhaps I deleted too much or what. I tried to figure out what’s wrong.

Rev. 1125 didn’t help.[I][/I]

[Tom Elliott]

restart apache.

FOR REDHAT
[code]service httpd restart[/code]

FOR UBUNTU
[code]sudo service apache2 restart[/code]

[rado]

if the restart was to me - it didn’t help.
but thanks

[rado]

Do you have any other tip (if the restart was for me)? If not, I’m going to reinstall it.

[Tom Elliott]

I’m not seeing the same issues as you.

At all.

I can create the task, watch the task get created. Start my system, and watch it image. But I’ll try some more details.

[Tom Elliott]

I think i narrowed down the issue.

I was imaging a winxp system with a resizable harddrive.

This means I have a file (FILE not DIRECTORY) that is called:

/images/winxpimage

It finds and images that perfectly fine.

The issue, right now, is the task is trying to gunzip the directory for a windows 7 image, rather than the files within the directory. I’m a big idiot and I’m sorry about that. Working on a fix now.

[rado]

OK, I was going to reinstall FOG and thought I should backup db first. So I did and found in sql dump file a table (multicastSessionAssoc) with lot of entries, so now I understand why it didn’t work as I deleted only entries in multicastSession and not here

Now everything seems to work, even multicast task starts (not only individual tasks), but 2 issues:
[LIST=1]
[]one host gets proper pxe config file with good porbase (say 58720) but second host gets 224 (it seems lik 1. octet of multicast address), so I have to change it manualy to start multicasting
[]but then it complains it’s not partclone image (I created it yesterday with 0.33 fog, so it should be partclone image); Is this what you are talking above? This is how it looks like in both versions:
[LIST=1]
[]0.32 is gunzip -c “/images/test/d1p1.img” | /usr/local/sbin/udp-sender … (I have a screenshot)
[]0.33 is gunzip -c “/images/cpu/”|/usr/local/sbin/udp-sender …
[/LIST]
[/LIST]

[Tom Elliott]

I fixed it. r1126 released to address the multicast task creation.

You should be good. I don’t know what was generating all the multicastSessions you are talking about, but it shouldn’t do that anymore.

Thank you,

[Tom Elliott]

Also,

Instead of reinstalling the DB.

Just clear out the sessions and Assocs:

truncate table multicastSessions;
truncate table multicastSessionsAssoc;

[rado]

You are welcome, Tom. Now multicast seems to work, but there is stil the issue with different ports assigned to hosts. I’m going to look at it.

[Tom Elliott]

I guess I don’t understand? Are the tasks going to different hosts on different ports from within the same tasking?

[rado]

I’m not sure if I understand you now but, when you create multicast tasks, it should create pxe file per client (MAC) and 1st my client got now multicast base port 45084 and 2nd got now 9964, which is, as I in the meantime found out that it’s some default port (I see it in globalSettings db table). But it was not every time this one, it looked like this client (and it seems that every time it’s the same client) gets port from previous multicast session (and perhaps now, as I started this multicast task after truncating those session tables, it gets some default value).

I guess, creating pxe config file is based on hostID, which correlates, this is how my hosts table for those hosts looks like:

[CODE]mysql> select * from hosts;
±-------±---------±------------------------------------------------±-------±----------±-------------±--------------------±-------------±------------------±----------±-------------±---------±-----------±-----------±-----------------±---------------±-----------±-----------+
| hostID | hostName | hostDesc | hostIP | hostImage | hostBuilding | hostCreateDate | hostCreateBy | hostMAC | hostUseAD | hostADDomain | hostADOU | hostADUser | hostADPass | hostPrinterLevel | hostKernelArgs | hostKernel | hostDevice |
±-------±---------±------------------------------------------------±-------±----------±-------------±--------------------±-------------±------------------±----------±-------------±---------±-----------±-----------±-----------------±---------------±-----------±-----------+
| 1 | 772 | Created by FOG Reg on January 20, 2014, 3:13 pm | | 6 | 0 | 2014-01-20 15:13:34 | FOGREG | d4:3d:7e:0c:8c:6f | | | | | | | | | |
| 2 | 768 | Created by FOG Reg on January 20, 2014, 3:15 pm | | 6 | 0 | 2014-01-20 15:15:48 | FOGREG | d4:3d:7e:0c:90:ed | | | | | | | | | |
[/CODE]

edit: and host 772 gets wrong port

[Tom Elliott]

I want to make sure I get this right!

Are you creating a multicast task for the individual hosts? (One task per host)

Or are you creating a multicast task from the group? (One task to multiple hosts)

I haven’t implemented a code base to kill the original task/job/command and regenerate if the two individual tasks have the same image id.

Does this make sense?