MB.'s questions

I had to add another IP address to our server (eth0:1 192.168.0.100) and all of the sudden ossec client stopped working.

On the client side I'm seeing this: 2014/02/19 02:31:28 ossec-agentd: INFO: Trying to connect to server (server_ip:1514). 2014/02/19 02:31:28 ossec-agentd: INFO: Using IPv4 for: server_ip. 2014/02/19 02:31:49 ossec-agentd(4101): WARN: Waiting for server reply (not started). Tried: 'server_ip'.

How can this be fixed? Most likely I have to specify the source ip or source interface on the client side but I can't figure out where.

Right now I'm storing apache2 logs locally:

ErrorLog     /var/log/apache2/one.error.log
CustomLog    /var/log/apache2/one.access.log common

How can I keep storing the logs locally but also forward them to a local rsyslog (who is pushing data to a remote rsyslog).

I'm trying to get a transparent https proxy setup working. I'm not looking to intercept or anything, I would like the https traffic to just be forwarded to the real host.

In squid I have something like: https_port 3130 cert=/etc/ssl/certs/host1.crt key=/etc/ssl/private/host1.key

However when I make a web request I'm prompted with the ssl certificate is invalid warning screen.

Am I doing something wrong? I tried adding 'transparent' to https_port and squid 3.4 won't start.

Management is asking me to tweak a couple of sites (http only is OK, https not needed) that employees access from our internal network. It would be something very basic, inserting some JS that displays a message. This would only be needed for some particular sites.

I was wondering what tool would you recommend for this? UNIX/Debian based.

I'm encountering the same problem as here: http://www.quora.com/Amazon-EC2/Is-it-possible-to-rescue-an-EBS-volume-which-has-marketplace-codes.

Basically I've locked myself out of the instance and I don't have any snapshots. I was working on a larger setup and for security reasons I tried setting up UsePam no in sshd_config which borked everything.

I used the official Debian Amazon Marketplace AMI. Now I'm not allowed to mount the volume somewhere else as a secondary device to fix my stupidity.

The solution listed in the quora thread isn't working. I tried it with two ubuntu images and the instance won't boot with that user data set.

Is there a recent workaround? I'm not trying to access anything secret since I have been using the stock Debian AMIs. I just want to fix that issue without having to re-do everything.

I've been trying to use s3fs on an EC2 server but it's really slow. I spent the last 5 hours uploading 100MB of small files.

Is there anything I can do to improve the performance or is it something related to S3fs? If not, what's an alternative I could use?

I have a new version of our app and it needs to be tested before placed everywhere in production.

There is one main machine that currently receives all requests and forwards them to an Apache using proxy_pass. I want to add another server, and forward only some GET requests to it.

Can nginx be configured to do that?