Yep, that certainly worked. We were avoiding that just so we did not lose all of the location associations that the hosts already had, but if this solves the issue, we will plow through it.
Thanks for the info! This took care of the issue.
Yep, that certainly worked. We were avoiding that just so we did not lose all of the location associations that the hosts already had, but if this solves the issue, we will plow through it.
Thanks for the info! This took care of the issue.
We are currently running the latest dev-branch FOG server 1.5.10.1629 on Ubuntu Server 20.04.6 LTS.
We have seven FOG Nodes added to the system and we have seven locations defined in the locations tab. I cannot seem to assign a location to a storage node anymore. When we first installed the servers, everything went well. We lost a hard drive in a server and had to rebuild it. After reconnecting the storage node to the FOG server, it pulls images as expected, but I cannot assign that location to the storage node.
When I select the correct storage node and click Update on the location, it gives a success message, but listing all locations just returns a * indicating that it will pull from any node, which is not good for us as these are on different WAN links.
I also cannot seem to add a new location. Attempting to add a new location gives me the same success message, but it does not show up in the list of locations.
Is there any way to edit the database to assign this storage node to the location?
I’m wondering if it has to do with spaces in the FOG Node names? Reading the forums here I found out that I should be able to click on the names of the Nodes on the Kernel Versions page to see their actual kernel versions but those menus do not open either. All of our Nodes and server itself have spaces in the names of them.
Seems no hints here unfortunately. Tried updating the location again and immediately checking logs and this was all that was reported:
redacted:~$ tail /var/log/php7.4-fpm.log
[13-Apr-2023 15:20:00] NOTICE: [pool www] child 49963 exited with code 0 after 11202.532821 seconds from start
[13-Apr-2023 15:20:00] NOTICE: [pool www] child 57441 started
[13-Apr-2023 15:20:00] NOTICE: [pool www] child 49961 exited with code 0 after 11205.012798 seconds from start
[13-Apr-2023 15:20:00] NOTICE: [pool www] child 57442 started
[13-Apr-2023 15:20:00] NOTICE: [pool www] child 49964 exited with code 0 after 11202.725741 seconds from start
[13-Apr-2023 15:20:00] NOTICE: [pool www] child 57445 started
[13-Apr-2023 15:20:00] NOTICE: [pool www] child 49962 exited with code 0 after 11205.098800 seconds from start
[13-Apr-2023 15:20:00] NOTICE: [pool www] child 57446 started
[13-Apr-2023 15:20:13] NOTICE: [pool www] child 49969 exited with code 0 after 11209.884619 seconds from start
[13-Apr-2023 15:20:13] NOTICE: [pool www] child 57455 started
redacted:~$ tail /var/log/apache2/error.log
[Thu Apr 13 00:00:35.653198 2023] [mpm_prefork:notice] [pid 968] AH00163: Apache/2.4.41 (Ubuntu) OpenSSL/1.1.1f configured -- resuming normal operations
[Thu Apr 13 00:00:35.653216 2023] [core:notice] [pid 968] AH00094: Command line: '/usr/sbin/apache2'
We are using the Location plugin with storage nodes across 6 different locations. We had a hard drive fail in one of the storage nodes and I rebuilt it yesterday. The storage node is up and active, but I cannot assign it to the correct location on the Locations tab. I can click the location, tell it to use the correct storage node, and I get a successful message when clicking save, but when going back to locations, it does not have the storage node assigned, only * next to the location. I even tried creating a brand new location and when I clicked save, it did not appear in the location tab even though I got a success message.
All storage nodes/DB server are running Ubuntu 20.04 and running dev-branch 1.5.10.4.
Updated and he has confirmed it worked. Captured from the same host twice just to make sure.
Thank you so much for looking into this!
@george1421 said in Cannot Capture Image - Failed to open stream:
@hummela ok then on the fog server would you run these two commands.
On the destination images
ls -la /images
ls -la /images/Dell7090Staff
and then on the captured directories
ls -la /images/dev
ls -la /images/dev/7486e211bc42
Ignore the fact that the mac address changed - he tried capturing again from a different host (bad troubleshooting there, I know). The capture still failed in the same way with this host.
I had him try to capture to the same image again and it failed in the same way - Failed to open stream.
Just as a bit of an update -
I manually deleted the /images/Dell7090Staff directory and had him try capturing the image again. This time, the image captured successfully. It definitely seems like an issue overwriting an existing image with a new one.
Confirmed that there is a folder in /images/dev named 7486e211be88, which has the files in it.
FTP moving was interesting. An image had previously been uploaded using this folder name in /images (something the other tech failed to tell me). So this was actually the second capture attempt for the same image. My FTP rename was unsuccessful the first time until I realized there was already a directory with the same name in the /images folder. I deleted that directory and the rename was successful. It seems that it has to do more with overwriting an existing image with an updated one.
I can confirm I am using the FOS kernel 5.15.34, as I tried updating that to see if that had been the issue.
both /images and /images/dev are owned by fogproject with group root, as are the directories inside.
One of my coworkers has approached me saying he cannot capture an image from one of our new Dell OptiPlex 7090 computers. From what I can tell, it uploads the image correctly to the /images/dev folder but fails when trying to move it to the /images folder. He told me that uploading an image from a different computer (a Dell Latitude 3520) works as intended. I have tried logging into the storage node that is being uploaded to with ftp and created a file/moved it successfully, but the OptiPlex will not play nice. Something curious that I noticed, it seems that the capture process is uploading as root instead of fogproject. The folder in /images/dev will be owned by root, and after the transfer to the /images folder will remain as root (in the case of the Latitude), the OptiPlex will put files in the /images/dev folder as root but will not complete the move.
Other info:
Running FOG server dev-branch 1.5.9.147 on a VM (10.1.0.27) with locations enabled, tftpboot to this server
Running FOG storage node dev-branch 1.5.9.147 on a physical server (10.1.0.115), uploading correctly to this server
Screenshot of the error that FOG spits out when it fails: