Fog server keeps going down

[szecca1]

@Wayne-Workman I’m sorry I am having a difficult time doing this, so I’m just going to skip it unless you have more instructions on how to do this.

[Wayne Workman]

@szecca1 https://wiki.fogproject.org/wiki/index.php/Vi

Also - use putty and just copy / paste. No need to type all that out.

In Putty, you can copy by selecting the text and then right clicking anywhere on the window.
In Putty, you can paste by simply putting the cursor where you want your text to go, and then right click anywhere on the window.

[szecca1]

@Wayne-Workman Thank you but clearly I did something wrong. This is my output when I followed your instructions

[Wayne Workman]

@szecca1 The cat command just outputs the contents of a file for viewing.

You run executibles in Linux with ./

Like ./installfog.sh ?

Try to run the script you just wrote. Like this:
cd /root
./monitor.sh
then output the contents of the log using the cat command.
cat /root/monitor.log

You can also look at the log using vi…
vi /root/monitor.log

[szecca1]

@Wayne-Workman I may have found my problem but not sure how to fix it

[Wayne Workman]

@szecca1 That’s not a big deal. What gets ran is what is stored on the disk.

Can you run the script? Is there output in the log?

[szecca1]

@Wayne-Workman No I tried to see if the file was in the /root folder and its not so i wanted to make sure that wasn’t a problem. I’ll just recreate it again

[Wayne Workman]

@szecca1 You’re using Fedora right?

Did you create the crontab event as root? the crontab -e command will create a cron event and run as the user who created the event, meaning, not created using root, no permission to the /root directory.

[szecca1]

@szecca1 I dont know what I am doing wrong but when I type /root then ls the file is not there and I continue to go back and resave it and still nothing

[Wayne Workman]

@szecca1 Switch to the root user like this: su root and then create the crontab event.

Apologies for not saying that earlier. I do everything as root.

[szecca1]

@Wayne-Workman apparently I don’t have the password for root

[szecca1]

@Wayne-Workman I was just doing everything through sudo and my password

[george1421]

For better or worse here, I have to be the voice of reason: At some point (in the near future) you have to make a decision to continue down the path to try to fix this install, or just abandon it and spin up a new fog instance and manually copy the files across.

IMO something is botched with this OS install. This system should not become unavailable like this. How much effort are you going to expend on trying to revive this system, where spinning up a new OS and fog install can be done in about 30 minutes.

[Wayne Workman]

@szecca1 Ok.

You can set this up under your username and not even need sudo or root permissions.

Log into the server using your normal user account. issue the pwd command as soon as you log in (present working directory). What does it say?

Change the paths in the script to point to that directory instead of /root/
Also, create the script in this directory as well.

[Wayne Workman]

@george1421 said:

For better or worse here, I have to be the voice of reason: At some point (in the near future) you have to make a decision to continue down the path to try to fix this install, or just abandon it and spin up a new fog instance and manually copy the files across.

IMHO, not having the root password is reason enough…

[szecca1]

@george1421 The problem isn’t getting the new OS installed, I have a NAS that also needs to be attached that Tom spent some time with me to get set up properly. If it was that simple I wouldn’t have even bothered to worry about it. The install of the OS was fine for about 5 months and then randomly this issue started occurring.

[Wayne Workman]

@szecca1 The synology NAS? FOG now fully supports those. No need for symbolic links or anything anymore.

[szecca1]

@Wayne-Workman If thats what you guys are recommending me do then fine I can do that. But what OS should I use? I don’t want to use Fedora and then have someone else tell me I should have used something else

[Wayne Workman]

@szecca1 How about do nothing for the moment… The server was not crashed this morning. Let it go. See what happens.

The below script and cronjob was just for information collecting.

Plus you can always just manually monitor top and free -m

[george1421]

Right now I’m not saying do or don’t start over. But this has been an issue for 11 days now. What is going on is not normal. There is something wrong with this build. It could be that you are using a NAS for storage for FOG (I simply don’t know since I have not setup this situation). All I’m saying at some time in the (near) future you will need to deiced to just “cut bait” and start over or continue on and spend more time on this and not on deployment.