Latest Development FOG
-
SVN 2642 released.
With this, while seemingly a large jump, comes multicast session joining to non-registered hosts.
It also fixes issues with the SnapinReplicator.
It fixes compatibility of “old” snapins as well to allow non-grouped snapins to still deploy to hosts.
Fixes consistent pinging of hosts on non-host pages.
Hopefully you all enjoy.
-
SVN 2651 released.
This release adds/changes the capabilities of Quick image.
Quick Image will now present a list of the images from your fog server. You can make a selection of any image of your choice and image your system with that image.
If the host is registered and has a valid image assigned to it, it will auto select the host’s image. If the host is registered, but does not have an image assigned, it will always default to the first item in the list.
The Host registered items, regardless of the image used, will be logged, so your dashboard will appropriately display the imaging task.
Now, the big guy I think most will be excited about (or at least I’m hoping so due to the commonality of this type of request).
Non-registered hosts can also be imaged using the quick image option. The option is been moved back into the non-registered and registered items. (BYPASS REGISTRATION ON ADVANCED MENU IS NO LONGER NEEDED YAY!)
What does this mean?
It means quick image is just that. It is a means to quicking image any host (regardless of the registration status). If the host is not registered, it will not be logged, but it will image just fine.Also with this, you can now specify which menu items you “Do not list” on any of the menu entries. Before there was either all, non-registered, or registered hosts. Now you can specify a simple do not list to not show a specific menu item.
Hopefully you all find this useful.
-
Yay team! Thx muchly for Quick Image. It works wonderfully!
-
[quote=“Tom Elliott, post: 39087, member: 7271”]SVN 2651 released.
This release adds/changes the capabilities of Quick image.
Quick Image will now present a list of the images from your fog server. You can make a selection of any image of your choice and image your system with that image.
If the host is registered and has a valid image assigned to it, it will auto select the host’s image. If the host is registered, but does not have an image assigned, it will always default to the first item in the list.
The Host registered items, regardless of the image used, will be logged, so your dashboard will appropriately display the imaging task.
Now, the big guy I think most will be excited about (or at least I’m hoping so due to the commonality of this type of request).
Non-registered hosts can also be imaged using the quick image option. The option is been moved back into the non-registered and registered items. (BYPASS REGISTRATION ON ADVANCED MENU IS NO LONGER NEEDED YAY!)
What does this mean?
It means quick image is just that. It is a means to quicking image any host (regardless of the registration status). If the host is not registered, it will not be logged, but it will image just fine.Also with this, you can now specify which menu items you “Do not list” on any of the menu entries. Before there was either all, non-registered, or registered hosts. Now you can specify a simple do not list to not show a specific menu item.
Hopefully you all find this useful.[/quote]
The Quick Image make the same job as Capone Deploy ?
Regards,
Ch3i. -
I don’t know what you mean?
Sort of yes, it falsifies the task entry information so it can run without being registered.
-
[quote=“Tom Elliott, post: 39153, member: 7271”]
Sort of yes, [B]it falsifies the task entry information so it can run without being registered[/B].[/quote]Ok, noted !
Regards,
Ch3i. -
SVN 2658 released.
With this comes hopefully the starting components with the new client to allow for a more secure channel of communication between the two. Legacy clients should remain as normal, but these steps will make working with the new client that much simpler.
-
SVN 2682 released.
With this release comes many improvements.
First major improvement was working with the Dashboard page. You can now select different time settings. While I know the time itself isn’t 100% accurate, it’s based on the delay’s between return of data from the php that’s called and the delay that’s actually in place. However, the data points do adjust proper for their required time setting. Also, there’s no more sleeping in the php script that is called. This was causing a delay between each node called which ultimately had to be processed before anything could happen such as switching out a link. Sometimes, the largest delay occurred between home page and another link simply because the delay that was set by the PHP plus the delay in timing of when it would re-request data, happened too quickly. This posed issues in Browser Memory leaking as well and could break the page. This is now fixed.
Basically, bandwidth.php now just does it’s action and sends the data back to jquery. Jquery then delay’s and if the previous setting is set, does all the math required to get the proper Mbps. Response is much faster.
Second, the mac-list page. While used to be fast, then made really slow, is now fast again. How fast you ask? Well, before it was collecting the file data from the proper place, and looping the data properly as well. After it got the data, it would insert into the database.
Now it creates the query to insert the data in one line. This ultimately means instead of 45 seconds or more to update, it might take about 3 - 10 seconds. Most of that time is only due to it downloading the latest file. With this, also means, the file is not downloaded into the web directory. This means I can prune svn/git of all the oui.txt files and save quite a bit of space as the file size is now around 3.3MB.
Third, updated the way kernel’s are loaded on the screen. All of my kernels from my website are now on SVN and there’s a new script to generate the file listing as necessary. This means we no longer have to hand peck the kernel update file to insert new records. Because of this, the selector now has three entries. The default being the new mechanism of getting files, which currently just has all the kernel’s I’ve built. This new script also passes the arch type back to the form where you enter the name of the file. It will suggest the name of the file (bzImage/bzImage32) based on the arch of the file returned ultimately meaning you can just start without having to remember which file you clicked on. As always, the existing files are backed up.
With that also comes the download to allow for proxy authentication as well seeing as some proxies require auth to operate. This was lacking before which caused issues for those behind authentication based proxy servers.
The new selections are: Published Kernels, Old Published Kernels, and Unofficial kernels.
Those who know my work, will notice that many of the files in Published Kernels are also the same as on Unofficial kernels. This is intended as I just, quite literally, copied all my kernel files to SVN to get the data else where.
-
Tom,
On latest svn I cant export my reports to excel anymore or pdf.
-
SVN 2694 should fix the export issue.
-
Running 2697 tom and still doing same thing .
-
What’s the error you’re getting?
-
[ATTACH=full]1519[/ATTACH]
[url=“/_imported_xf_attachments/1/1519_screen shot.png?:”]screen shot.png[/url]
-
That looks like a specific problem with IE/Firefox?
-
Tom,
Just tried it on another pc and I received the same error.
-
Beyond another PC have you tried a Nother browser
-
Yes same in IE and Mozilla.
-
What version of IE? 9, 10, & 11 are not all the same.
-
11
-
SVN 2707 released.
With this comes improvements in building the Reports, especially in dealing with large table sets as you may have seen in another forum.
Also comes some simpler mechanisms in jquery/javascript. I may work towards a complete rewrite down the road, but for now I’m just cleaning them up and adding functionality where I think it’s kind of been needed.
For example, on the task management page, it keeps updating the number of active tasks (as expected) but it does so on any page dealing with the Tasks including places where you may not want this to happen (search to name one instance I can imagine). This is now corrected for. I’m able to pull in get variables (the stuff in the url) so I can test if we’re on a page we want updates or not.
