Latest FOG 0.33b
-
r1251 and r1252 released.
r1251 fixes an issue with the Snapin Deployments of jobs other than specified snapins.
r1252 adds, hopefully, performance tweaks on the nfs stuff. Tried implementing to the server side in r1252, but it doesn’t like it so they’ve all been moved to client side in the init.gz.
Thank you,
-
Hopefully the NFS updates work for everyone
If you have before and after numbers for speed that would be brilliant.
More ideas incoming…
-
[quote=“Tom Elliott, post: 21338, member: 7271”]MAC’s are supported as I’ve added the binaries to the init.gz, though I think this will have to be manual for the time being. I’ll have to add the MAC os to the OS Listing and have it translated to allow for hfs imaging.[/quote]
Tom, just out of curiosity, how far out do you think Mac imaging ability is in the priority list? We’d love to be able to ditch DeployStudio and move entirely to FOG. I know there are other priorities, I’m just curious
-
Would be great if that could be added. Tom has been on IRC for the past few days. Do you have an equivalent for the client to put in the hostname etc for MAC OS?
-
Personally, I’d be happy to just have the basic PXE/iPXE booting and imaging capability for now. For our school’s needs, we could simply use a script for the hostname, etc. The client would still be nice to have though. I tried booting a Mac after the recent move to iPXE, but no luck
-
[quote=“ArchFan, post: 23419, member: 19266”]Tom, just out of curiosity, how far out do you think Mac imaging ability is in the priority list? We’d love to be able to ditch DeployStudio and move entirely to FOG. I know there are other priorities, I’m just curious[/quote]
As I don’t have any Mac’s in my possession, it will likely be a little while, unless you can give what the output of the command: [code]blkid -po udev <partitionnumber>[/code] for each of the partitions on the system (especially a freshly installed system) it’ll probably be a little bit.
While I have made sure to have the partclone.hfsp binary available, and also included the hfs file system support in the kernel, I don’t know what to check for to get the MPS/MPA imaging types to use the proper binary.
-
r1253 released.
Redhat doesn’t like the php-json information so this had to be removed. Sorry about the issue before.
-
Once again Tom, thanks for all your help.
I was testing the snapin issues I was having on the latest version (1252) this morning thought I would share my results.
-
When I deploy an image on a computer it creates the snapin task as well. But when I tried to image the computer and see if everything worked I got the message in iPXE: fog/kernel/bzimage… No such file or directory ([url]http://ipxe.org/2d0c613b[/url])
-
When I tried to register a new computer the menu came up, selected full registration. Then it went to blank screen with blinking cursor.
-
I no longer get an error message when selecting a group to deploy all snapins but they only show in Active task and don’t actually deploy. Screenshots attached.
-
When using group to deploy single snapin I didn’t have a choice on which snapin. It also does not show anything in the snapin task.
-
When deploying all snapins on a single computer it would show in both Active task and Snapin task but I waited about a hour and it didn’t deploy.
-
When deploying all snapins to a second computer it would remove the Snapin task from the first computer. Screenshots attached.
-
When deploying a single snapin to a computer it shows in both Active and Snapin tasks. One deployed right away then tried another it sat there for over a hour with out deploying.
-
When deploying a single snapin to a second computer it states that the Snapin is already setup for tasking. Screenshot included.
Also when trying to created a group from the group management page I got the following errors. It did create the group.
Warning: Invalid argument supplied for foreach() in /var/www/fog/lib/pages/GroupManagementPage.class.php on line 239
Fatal error: Call to a member function get() on a non-object in /var/www/fog/lib/pages/GroupManagementPage.class.php on line 252[url=“/_imported_xf_attachments/0/558_snapins.zip?:”]snapins.zip[/url]
-
-
My guess is some of the problems you’re reporting (not specific to snapins) are related to the fog/images/init.gz fog/kernel/bzImage you’re reporting.
1.) The no such file or directory issue: fog/kernel/bzImage fog/images/init.gz etc… are because I modified the kernel and image stuff to obtain from the GUI which auto sets based on the values in config.php when you FIRST install fog. I fixed the “fresh” install so all should work properly.
If you already had it installed and performed an update it wouldn’t have changed your database values. So to fix, go to FOG Configuration Page->FOG Settings and change the fog/memtest/memtest, fog/kernel/bzImage, and fog/images/init.gz to memtest, bzImage, init.gz respectively and all should work properly for you.
2.) Probably related to Item number one.
3.) Will look into that, I only fixed the Snapin issues on the Host side. While the Host side still creates the “image” package, I probably missed something and will look into that.
4.) Haven’t added that component yet. Just trying to get the “individual” stuff working absolutely before making the rest of the changes. The other issue I foresee with it is that detecting which snapins are equally associated across the group can be painful. So have to try to figure out the stats on that and how to get them working across the board.
5.) Snapins are a fickle little thing. You actually have to log in to the system for the snapin to truly deploy as it requires user rights to install.
6.) Understood, I’ll take a look at that today when I get home.
7.) What do you mean one deployed right away, while the other just sat there? You mean you setup two systems to deploy a single snapin and one deployed right away, while the other just sat there? Was the one that deployed “right away” logged in and the other no user logged in?
8.) Probably the method of checking isn’t verifying the host. Will look into when I get home.
The last item’s I’ll take a look into as well.
Warning: Invalid argument supplied for foreach() in /var/www/fog/lib/pages/GroupManagementPage.class.php on line 239
Fatal error: Call to a member function get() on a non-object in /var/www/fog/lib/pages/GroupManagementPage.class.php on line 252 -
as further info, I’m also getting the same 2 error messages when trying to just about anything to a newly created (empty) group, on r1249.
PHP Warning: Invalid argument supplied for foreach() in /var/www/fog/lib/pages/GroupManagementPage.class.php on line 239, referer: [url]http://localhost/fog/management/index.php?node=group&sub=delete&groupid=1[/url]
PHP Fatal error: Call to a member function get() on a non-object in /var/www/fog/lib/pages/GroupManagementPage.class.php on line 252, referer: [url]http://localhost/fog/management/index.php?node=group&sub=delete&groupid=1[/url]Rgds
-
[quote=“Tom Elliott, post: 23476, member: 7271”]
7.) What do you mean one deployed right away, while the other just sat there? You mean you setup two systems to deploy a single snapin and one deployed right away, while the other just sat there? Was the one that deployed “right away” logged in and the other no user logged in?
[/quote]I see what you are saying about a user being logged in. And yes two separate computers, same plugin, one deployed right away the other didn’t. I believe that there was a user logged in at the time that I tried to deploy the snapin. There was at the beginning of my testing. When I check the log on the computer itself it shows that a user was logged on when I initially set it up. That user logged out about 20 min after that then another user logged in for about a hour.
-
Fixed the deploy issue where it was overwriting another set tasking.
-
Fixed the group, it never had the associations. Haven’t fixed it to find the matching associations on the hosts, but working on it.
-
r1254 should fix many of the issues reported.
Specifically:
3, 6 and the error thrown. it’s based on Location. I forgot to add checking if the plugin was enabled or not. If it is, it actually matches the location if all hosts in the group match. -
Items 1 and 2 weren’t necessarily an issue, just an oversight that’s relatively simple to fix.
Haven’t fixed #4 yet is it’s more complex logic, so I just added all snapins in the system. If they aren’t associated with the host, they won’t deploy to the host.
Item number 5 is more of an issue due to the login, nothing that I can fix right now. Item number 7 once again just the same as item number 5.
Haven’t worked on 8 yet, will work now.
-
r1255 released.
Fixes item #8, please let me know if there’s any more issues. Thank you for the detailed report.
-
r1256 released.
Fixes an issue in the TaskManagementPage.class.php where The taskType wasn’t returning properly.
Fixes the Wakeup icon to show up properly. Still needs edit’s though.
-
r1257 released.
Fixes the TaskManagementPage scheduled tasks stuff. Can delete scheduled tasks now, and they display more appropriate.
-
Tom, have updated to 1257 and now I am unable to deploy or register new hosts
-
Can you provide error logs by chance?
I just tested registration and deployment and all works well on my end.