Posts made by Kevin Talbot

@Sebastian-Roth yes and master was not ticked! I have ticked it but that has not cured the issue!

@Kevin-Talbot I just made the node a master but that has not cured the problem!

@george1421 said in Updating Database failed! Failed to open stream!:

node. Only the master node in a storage group can be used to capture im

I’ve always captured to a node, I have a separate storage group with one node in it which is the master for that storage group!

But I think you may have just nailed the problem…when I moved the node I renamed it and re-ip’d it and something tells me I may have deleted the old “named” node and added the newly “named” node but not clicked “master”

@Sebastian-Roth said in Updating Database failed! Failed to open stream!:

and storage node have changed as well?

Thanks for the reply Sebastian!

Only the node subnet has changed, I re-ip’d it as per the book!

@george1421 Hi yes very much still an issue thanks for replying!

I can deploy from and capture to the master fog server perfectly fine.

I can deploy from but NOT capture to the node

Struggling to get the screenshot to upload fails at 97% grrr…

Can anyone help please?

I have rolled back the passwords and now back to: Updating Database failed! Failed to open stream: no such file or directory

However, I can actually Capture to the master fog node, just not to the separate node!

Must be permissions I think?

Hi Guys please help!

Master Fog on VM Running Version 1.4.4 SVN Revision: 6077and single Node on pc, both running Ubuntu server 16.04.3

Recently physically moved the Node to new building and subnet. Renamed the node and re-ip’d it.

Also put a bigger Hard Drive in the Node

DEPLOY WORKS FINE FROM NODE !!!

But on Capture at 100% - Getting Database failed! Failed to open stream: no such file or directory

ran the following on the node and rebooted:

sudo chmod -R 777 /images

Also just changed all the passwords to match in the following locations and rebooted both:

Web Interface -> Storage Management -> [Your storage node] -> Management Username & Management Password
Web Interface -> FOG Configuration -> FOG Settings -> TFTP Server -> FOG_TFTP_FTP_USERNAME & FOG_TFTP_FTP_PASSWORD
The local ‘fog’ user’s password on the Linux FOG server
Server file: /opt/fog/.fogsettings -> password (For recent FOG Trunk versions only. 1.2.0 does not have this setting. 1.3.0 will though.)
Server file: /opt/fog/.fogsettings -> username (For recent FOG Trunk versions only. 1.2.0 does not have this setting. 1.3.0 will though.)

But now getting:

Updating Database failed! Login incorrect

Please help as I think I am making it worse?

Ok everybody I think I have nailed it however I don’t know if I can do anything about it??

Turns out we changed the model of SSD we put into our oldest client machines (Dell Optiplex 380’s)

I put a brand new (out the packet) “OLD MODEL” SSD in the same client machine I’ve been testing and deployed in 4 minutes flat…compared to the new model SSD which takes 15mins!!!

Question is can I do anything about it as the new model SSD is supposedly faster than the old model SSD

Both Kingston:

Old model = SHFS37A/120G
New model = SA400S37/120G

Can anyone think of anything that can be done as we have around 150 of these new model SSD’s…grrrr ???

It’s the same issue as this: http://www.edugeek.net/forums/o-s-deployment/103004-tasks-slowing-down.html

But I don’t know much about the init file?

@wayne-workman The old rig could deploy an image at a constant 5gb/min and that was an optiplex 360 with 2gb ram and a 1tb mechanical 7200 hard drive running ubuntu desktop 12 and fog1.2

I think this all started when I started upgrading ubuntu and fog on the above old hardware to ubuntu 14 and fog 1.4.4

I thought the 1tb hard drive had a bad block so thought I would start fresh, however the new rig I build today which is the latest ubuntu server 16.04.3 and fog 1.4.4 with all updates on much newer/faster hardware is deploying much much slower!

I think this is a fog issue but can’t pin point it yet! I have seen mention of replacing the init image with a previous versions init file???

@sebastian-roth The node server and client machine both have brand new SSD drives taken out of their packets today!

Hi Tom thanks for the reply!

It’s a brand new node, I built it today and its been rebooted multiple times.

I have rebuilt the whole thing from scratch as it was all on ageing hardware.

Now have VM running Ubuntu server 16.04.3 and fog 1.4.4 all bang up to date. (database)
Also Dell Optiplex 3020, i3 chip with 8gb ram, new SSD, running the same (node)

I have tried different compression levels and types, different patch cables, rebooted main switch etc etc. Deployments start at different levels due to compression but all end up coming down to 3gb/min which is no good.

On the old hardware I used to get constant gb/min speed for the duration of the deploy.

I have now taken to plugging the node and host into a seperate gig switch and isloating once started imaging and it’s doing the same thing!

Very frustrating, I must be missing something very simple somewhere?

I thought that had cured it as it started at over 8gb/min but it’s now down to less then 3gb/min at 20%

Please help?

I am just trying Partclone Zstd on level 10 to see what happens! It was on the default Partclone Gzip level 6.

Sorry yes I mean’t Gb/min!

Brand new Ubuntu Server 16.04.3 running fog 1.4.4

i3 chip, 8gb ram and SSD

40gb image used to deploy to one host in approx 6 min on the old version of fog.

IGNORE THIS I HAVE FIXED IT!!!

one final “sudo chmod -R 777 /images” on the node cured it!