2 questions regarding image capture/creation.



    1. What is this replicate option? alt text Does it duplicate the image for redundancy/backup? Essentially I want to know if it takes up storage space.

    2. When capturing an image, I noticed FOG’s storage space would go down, but after completion it returned to it’s normal value (after recapturing the same image). Does FOG store a temporary file of the in-progress capture before overwriting the image on the server? I want to understand the process of capturing better, as I had always assumed if I captured over even a bit the image became corrupt even if cancelled before completion.

    Thanks!



  • @Tom-Elliott I see, the storage nodes themselves are not meant for capacity but redundancy and throughput? Is mounting a network storage location viable to expand my capacity?


  • Senior Developer

    @salted_cashews
    I think you’re thinking too much into this.

    If you have 1 storage group, the master node replicates its “replicatable” images to all storage nodes in the group.

    If you have 2 storage nodes in one storage group, the images that are set to replicate will replicate from the master node to the other node. The space is not “divied” up. The master node will always be the node images captured are loaded to.

    You could, I suppose, copy specific images from the master node and place it across other nodes so long as those images are not set to be replicated, but again, this does not increase your actual storage space. It just distributes it to the other nodes.

    The simplest method, add more space to your master storage node.



  • @Tom-Elliott So I would like to expand our total storage capacity for images, as our main server (and node) are close to max capacity. Would adding a storage node across the network work for this or would I physically need to add more drives? E.g. node 1 has images a-z, but I need more space for 1-9. If I add a separate physical node via the network (node 2), would I be able to keep a-z on node 1 and 1-9 on node 2 and still be able to capture/deploy as normal? I would just have to specify which node to capture/deploy from this point?


  • Senior Developer

    @salted_cashews Storage node’s can “see” the images, but the images are loaded from the local node in use. Meaning, if you don’t replicate the image across your storage nodes, and a machine uses a node that has NOT had the image replicated to it, it will fail to perform the task. We have some checks and balances of course so hopefully a machine will not load a storage node that does not have the image files available to it.

    I don’t understand what you mean if “that would simply expand the size of your available storage”?

    The storage is reported on the local storage node, not the total of all nodes in a storage group.



  • Being that I don’t think this warrants another topic thread, I was curious if I added another storage node and set the images to not replicate, if that would simply expand the size of my available storage? If a node isn’t master it can still see other images and I can still store images on that node to be used correct?



  • @Sebastian-Roth Sweet. Thanks!


  • Developer

    @salted_cashews said in 2 questions regarding image capture/creation.:

    If I only have one node does it just ignore the option even if checked then?

    Yes, pretty much. Definitely does not use more space on your single node.



  • @Tom-Elliott said in 2 questions regarding image capture/creation.:

    1. Yes, it replicates the image to other storage nodes, if you have multiple.
    2. Yes. It captures the image into /images/dev/<macofmachinecapturedfrom>. Once the image is captured and complete, it copies the image from /images/dev/<macofmachinecapturedfrom> to /images/<imagePathName>. If /images/<imagePathName> already exists, it will overwrite it.

    If I only have one node does it just ignore the option even if checked then?

    This is game changing, I no longer need to fret if I start a capture by mistake haha.

    Thank you for such a quick and concise response, you guys are on top of it.


  • Senior Developer

    1. Yes, it replicates the image to other storage nodes, if you have multiple.
    2. Yes. It captures the image into /images/dev/<macofmachinecapturedfrom>. Once the image is captured and complete, it copies the image from /images/dev/<macofmachinecapturedfrom> to /images/<imagePathName>. If /images/<imagePathName> already exists, it will overwrite it.

Log in to reply
 

450
Online

5.9k
Users

13.3k
Topics

125.1k
Posts