Tag Archives: CentOS

Create Repo runs out of disk space in tmp

I have one CentOS 6 computer that is completely stand alone, as in no internet / network access. What? Yea, in this day and age, but I won’t get into the whys.

To keep this box updated, and to make additional packages available on demand, I manually copy the repository files to it, after which I rebuild the repo.

After the last update, besides running into a kernel panic that was easily fixed, I also ran into an issue with filling up the /tmp partition.

Basically, I ran into a “IOError: {Errno 28] No space left on device” and “Worker exited with non-zero vaule: 1. Fatal”. Picture below for your viewing pleasure.

out-of-space

I did a lot of searching for terms like createrepo and No space left on device, but didn’t have much luck finding anything. I was hoping to have createrepo save it’s temp files in another location, but no luck there.

Now, I’m not much of a Linux Admin, so forgive me if this is a clumsy fix, but the thought occurred to me that I could specify a different location for tmp, or even just change it temporarily, then I would be back in business.

I ran across this wonderful post that explained how I could do so using:

export TMPDIR=/path/to/other/directory

Someone in the comments mentioned that the command above would leave the change in the environment, and I assumed change it permanently, at least for that user. I wasn’t sure on that point, but they suggested using this instead:

env TMPDIR=/path/to/other/directory; command

I gave it a try, and it didn’t work for me. Now, I’ll admit here that I was running a script instead of a command, which could have made the difference.

What I ended up doing was to place the following at the beginning of the script:

export TMPDIR=/path/to/other/directory

Then the following at the end:

export TMPDIR=/path/to/original/directory

Evidently /tmp is the general location, and that’s what I set it to.

Kernel Panic in CentOS 6.7 After Kernel Upgrade

I ran into this issue after updating with Yum, on the next reboot, the system wouldn’t boot properly. In my case, the error I received was

Kernel panic – not synching:

VFS: Unable to mount root fs on unknown-block(0,0)

 Picture of the full error below:

error

In my case, I’m running CentOS 6.7 in a virtual machine running on Oracle VirtualBox.

You should be able to get around the error pretty easily, just reboot the computer, and when you see the “Press any key to enter the menu” pictured below, press a key.

press-key1

You should be presented with a list of the kernels you have installed, pictured below, select the previous one, and press enter.

previous-kernel

After doing that I was able to boot into CentOS.

I did a little digging, and found this bug report, which matched my error, the only difference I can see is that the kernel version was different.

The first solution I ran across was to run “yum reinstall kernel”, which is what I did, and the issue was resolved. Evidently in some cases the kernel doesn’t fully install, and in some cases fails transaction checks.

If the reinstall doesn’t work for you, keep reading the comments, there’s several other solutions, hopefully one of them will help you out. Feel free to let me know what worked for you.

Good Luck!