Difficulty Adding Storage Node



  • Hey All,

    I’ve searched and read many of the Articles on the Forums and Internet related to my problem.
    I think I’ve changed files too many times and I guess im not sure what Config File I need to look at and change or correct at this point.

    Essentially at first I was unable to locate the “password” for the storage node because the screen didnt pop it up… Eventually I found an article that told me where it was and I put it in and didnt work… I had already changed Values in that Config File, I changed Files in the opt/fog/services/etc/config.php file, I changed the MYSQL in the Web GUI Config Settings…

    At this point my Primary Machine still images and works but it is unable to communicate to the Storage Node with the following Logs: I realize the Failed Login notice, but I have no clue if this is on the Storage Node End, GUI End, MYSQL End or what!

    [12-18-12 11:02:22 am] * [12-18-12 11:02:22 am] Finding group members.
    [12-18-12 11:02:22 am] * [12-18-12 11:02:22 am] Found: 1 other member(s).
    [12-18-12 11:02:22 am]
    [12-18-12 11:02:22 am] * [12-18-12 11:02:22 am] Getting my image root.
    [12-18-12 11:02:22 am] * [12-18-12 11:02:22 am] My Root: /images
    [12-18-12 11:02:22 am]
    [12-18-12 11:02:22 am] * [12-18-12 11:02:22 am] Starting Sync.
    [12-18-12 11:02:22 am] * [12-18-12 11:02:22 am] Syncing: Fog Storage G5
    [12-18-12 11:02:22 am] * [12-18-12 11:02:22 am] SubProcess -> Removing old file `5000XP’

    [12-18-12 11:02:22 am] * [12-18-12 11:02:22 am] SubProcess -> Transferring file `5000XP’

    [12-18-12 11:02:56 am] * [12-18-12 11:02:56 am] SubProcess -> mirror: Fatal error: max-retries exceeded

    [12-18-12 11:02:56 am] * [12-18-12 11:02:56 am] SubProcess -> Transferring file `ElemGX240’

    [12-18-12 11:02:59 am] * [12-18-12 11:02:59 am] SubProcess -> mirror: Login failed: 530 Login incorrect.

    [12-18-12 11:02:59 am] * [12-18-12 11:02:59 am] SubProcess -> Transferring file `HSMCGX260’

    [12-18-12 11:03:03 am] * [12-18-12 11:03:03 am] SubProcess -> mirror: Login failed: 530 Login incorrect.

    [12-18-12 11:03:03 am] * [12-18-12 11:03:03 am] SubProcess -> Mirroring directory `COW5102W7’

    [12-18-12 11:03:03 am] * [12-18-12 11:03:03 am] SubProcess -> Making directory `COW5102W7’

    [12-18-12 11:03:36 am] * [12-18-12 11:03:36 am] SubProcess -> mirror: Access failed: 550 Failed to change directory. (/images/COW5102W7)

    [12-18-12 11:03:36 am] * [12-18-12 11:03:36 am] SubProcess -> Mirroring directory `ELMCDC7800W7’

    [12-18-12 11:03:36 am] * [12-18-12 11:03:36 am] SubProcess -> Removing old file `ELMCDC7800W7/d1p2.img’

    [12-18-12 11:03:36 am] * [12-18-12 11:03:36 am] SubProcess -> Transferring file `ELMCDC7800W7/d1p2.img’

    [12-18-12 11:04:10 am] * [12-18-12 11:04:10 am] SubProcess -> mirror: Fatal error: max-retries exceeded

    [12-18-12 11:04:13 am] * [12-18-12 11:04:13 am] SubProcess -> Mirroring directory `HS6005LabV4W7’

    [12-18-12 11:04:15 am] * [12-18-12 11:04:15 am] SubProcess -> mirror: Login failed: 530 Login incorrect.

    [12-18-12 11:04:15 am] * [12-18-12 11:04:15 am] SubProcess -> Mirroring directory `HSMCDC7800W7’

    [12-18-12 11:04:19 am] * [12-18-12 11:04:19 am] SubProcess -> mirror: Login failed: 530 Login incorrect.

    [12-18-12 11:04:19 am] * [12-18-12 11:04:19 am] SubProcess -> Mirroring directory `Teacher6005V3W7’

    [12-18-12 11:04:22 am] * [12-18-12 11:04:22 am] SubProcess -> mirror: Login failed: 530 Login incorrect.

    [12-18-12 11:04:35 am] * [12-18-12 11:04:35 am] SubProcess ->
    [12-18-12 11:04:35 am] * [12-18-12 11:04:35 am] SubProcess -> Complete


  • Moderator



  • Chad,

    Will give those instructions a try. Could you tell me how to create a dummy file to test with? I thought I saw you post something in another thread about Linux being able to create a gibberish file with a size limit…

    Thanks!


  • Moderator

    For tests, try to push a 10mb,100mb, and 1gb dummy file from the master node to the storage node and see if you have issues with FTP after changing ports and if needed, after changing switches.


  • Moderator

    It’s strange that from the storage node to master, there is a an intermittent connection time out issue but not from the master to the storage node. Can you try different ports on the switch or a different switch just to see if the same thing happens.



  • Currently both are connected to the same 3com 10/100mb switch. Our network uses no VLans. Dhcp is supplied by our windows server.


  • Moderator

    Explain how the fog servers are connected via the network? Same switch, vlan, etc…



  • Chad,

    Thank you for your Reply on Christmas Eve… Im stick at the Office waiting for a package trying to figure this out…

    I tried your FTP Suggestions and something to note:

    On Storage Node:
    I was able to successfully FTP to its own IP Address (10.75.200.4) with ID:fog/Pass: fcebfb (password it assigned)
    I was able to FTP to Master Server (10.75.200.5) with fog and f0g (what I setup as the Main Login on the server), however about 50% of the time it would say connection time out… With that being said when I did get FTP’d in, I Changed Dir to /images, I created a Dir named Test (it did it), but then when I wanted to see all the DIrectories Listed, the connection timed out again.

    On the Master Server:
    I was able to successfully FTP to its own IP Address (10.75.200.5) with fog/f0g
    I was able to successfully FTP to Storage Node (10.75.200.4) with fog/fcebfb (password that was assigned with NO connect issues)
    I was able to Create a Directory and get a list of directories.


  • Moderator

    From your log file, it appears you have multiple issues with the storage node. It helps to understand the process of replication between the main fog server and the storage nodes.

    The main FOG server asks the SQL server for a list of storage nodes in its storage group. It them tries to connect to each one using the username “fog”, and the password specific to each storage node. The storage node password was created by the installer when you setup the storage node. It’s located in the .fogsettings file on the storage node in the /opt/fog or /opt/fog-setup directory.

    After you setup the storage node and get the password value out of the file, you login to the main fog server web UI and define a new storage node in the same storage group as the main fog server. You have to put in the ip address, username, and password. The username is almost always “fog” and the password is usually “fs####” where #### is a number between 0000 and 9999.

    After you define the storage node, you should start seeing the main fog server try to push its /images out to the storage nodes using lftp. It only syncs the differences from the main fog server to the storage node.

    You can kick start this process by stopping and restarting the FOGImageReplicator service and you can track this by tailing the /opt/fog/log/fogimagereplicator.log file. I’m not sure on the exact name and location of that file but it’s somewhere in the /opt/fog/log directory on the MAIN fog server.

    You can verify FTP is working between the two FOG servers by going to a command line and trying to connect via FTP to the storage node using the “fog” username, and the password from the .fogsettings file on the STORAGE node. You should be able to browse, created, and delete files in the /images directory on the storage node. If this doesn’t work, then we need to troubleshoot FTP.



  • I’ve also followed these two Links to try and fix it myself today:

    http://fogproject.org/forum/threads/unable-to-replicate-image-on-storage-node.615/

    and Today I rebuilt the Fog Storage Node from Stratch and same error (this time I didnt use fog as the main login so the install script gave me a user ID and password), so I am inclined to believe it is an Issue with the Main Fog Server and some way I have the passwords Setup or messed up. I am running Fog .32 on Ubuntu Ver. 11.10. Maybe TFTP Settings producing these errors?

    Any Advice would be great… I have recieved 116 PCs that will need to be imaged in about 2 weeks and would hate to have to try and Single Cast each of them with this 2 Disc Main server, it would be cruel and unusual punishment. My Storage Node is a DL380 G5 with 8 Discs in Raid configuration so it would spank them out in no time.

    Thank you in advance.


Log in to reply
 

332
Online

38724
Users

10554
Topics

99908
Posts

Looks like your connection to FOG Project was lost, please wait while we try to reconnect.