Rubenerd Annexe (Closed)

I've retired all my Tumblr blogs, in light of the Yahoo purchase and recent censorship concerns.

I'll be doing all my writing on Rubenerd.com, where I'm also importing all my posts.

mdadm is a tool for managing Linux Software RAID arrays.

It can create, assemble, report on, and monitor arrays.

It can also move spares between raid arrays when needed.

It can be found at http://www.kernel.org/pub/linux/utils/raid/mdadm/ or any of the kernel.org mirrors.

There is a git repository at http://neil.brown.name/git/mdadm or git://neil.brown.name/mdadm.

FAQ and other documentation can be found at http://www.linuxfoundation.org/collaborate/workgroups/linux-raid

If you just supply the minimum options:

virt-install paravirtualized guests cannot install off cdrom media

It must default to PV mode. For HVM, use this surprising option:

Assuming your builders type is qemu, you may have seen this:

Qemu stderr: Could not initialize SDL(No available video device) - exiting

One solution in the Packer json was to disable graphics:

qemuargs: [ "-nographics" ]

Alternatively, start it headless:

"headless": "true"

Building a Packer image (how's that name for terrible Google juice):

==> qemu: Error launching VM: Qemu failed to start. Please run with logs to get more info.

So how do we run with logs? From Debugging Packer:

Packer has detailed logs which can be enabled by setting the PACKER_LOG environmental variable to any value

So in that case:

export PACKER_LOG=1

You can also define a logfile instead:

If you want to run scp on a non-standard port, you use the -P flag. So I did this:

scp file.img.xz -P 99999 [email protected]:/dest/path/

And what did I get?

ssh: connect to host example.com port 22: no route to host

Why was it ignoring my hypothetically impossible port 99999? Oh, wait.

scp -P 99999 file.img.xz [email protected]:/dest/path/

A useful set of flags is the delightful "plunt"

# netstat -plunt

For example, on a PostgreSQL box with custom ports:

Assuming you can access your database:

postgres=# SHOW data_directory

On Debian Wheezy with the official PostgreSQL repo, the default will show:

data_directory ------------------------------ /var/lib/postgresql/9.3/main (1 row)

So now when I restarted Postgres:

Error: Invalid line 603 in /etc/postgresql/9.3/main/postgresql.conf: \ >listen_addresses="localhost,10.0.1.2"

See the bug? Yes, double quotes instead of single. From postgresql.conf:

listen_addresses='localhost, 10.0.1.2'

Getting the following error when trying to start Postgres:

UTC LOG: invalid IP mask "md5": Name or service not known

Turns out, it's because I forgot to define the subnet mask for the allowed IP, which meant the security setting was being interpreted. Weird behavior, I would have expected it to search after a backslash.

An example for pg_hba.conf:

host replication postgres 10.0.1.10/24 md5

And an example of my bad one: