Is it possible to use a variable in an nginx upstream directive?
upstream appserver {
server unix:/socks/$host.sock
}
I've search my little heart out and can't find the answer.
A) Is it possible?
B) Is this a terrible idea?
Logcheck currently sends lots of emails with messages like this
Jun 6 19:31:44 <hostname> kernel: [UFW BLOCK] IN=eth0 OUT= MAC=<mac-address> SRC=<source-ip> DST=<destination-ip> LEN=40 TOS=0x00 PREC=0x00 TTL=116 ID=28729 DF PROTO=TCP SPT=56681 DPT=80 WINDOW=16652 RES=0x00 ACK FIN URGP=0
According to this Q&A this packet is blocked because it is optional.
Is it reasonable to have the following /etc/logcheck/ignore.d.server/ufw
^\w{3} [ :[:digit:]]{11} [._[:alnum:]-]+ kernel: \[UFW BLOCK\].*ACK FIN.*
I've read just about everything there is online about improving postgres performance, but the "right" values for SHMMAX and SHMALL still elude me.
The consensus seems to be SHMMAX = total_memory/4 and SHMALL = total_memory/2 are safe starting values.
However, SHMALL can be measured in pages or bytes and I can't find any info about which is used on Ubuntu.
Does Ubuntu (or more generally Debian) use pages or bytes for SHMALL?
What is the best way of managing which services start when?
service --status-all lists all services, but does a poor job of the service status (most are "unknown") and does not provide any method of disabling/enabling services.
I read that modifying/deleting the symlinks in the rc*.d folders is not advised, because they can be recreated when their package is upgraded.
I have been researching log rotation for my server which hosts ~5 fairly high traffic sites. From what I can tell, my options are to use logrotate or to use piped logging with either rotatelogs or chronolog.
logrotate requires a restart of apache and both SIGHUP and SIGUSR1 restarts are less than ideal on high traffic sites, because either you drop a bunch of connections or you need to delay compressing the old log until all child processes have died naturally. Also, downtime can be quite significant if compression is enabled. Would using logrotate - without compression and with graceful restart - and compressing old logs after the fact be the best way to minimize downtime?
chronolog and rotatelogs sound promising, but are not well documented. I couldn't find examples of using either in combination with vhost specific logs. The chronolog website says, "when the expanded filename changes, the current file is closed and a new one opened". Is this globally? Or is that per AccessLog, CustomLog or ErrorLog directive?
Is there a significant difference between chronolog and rotatelogs?
I have a django app and I want to to setup Varnish on a server in front of it. In another serverfault thread somebody suggested putting Nginx in front of Varnish.
Should I put Nginx in front of Varnish on the caching server? If so, should I use Nginx on the app server?
I am trying to migrate a site to a new server (centOS5 and plesk 9.2.1)
I would like to get the site running on a subdomain before pointing the root domain to the new server.
In plesk I created example.com and the subdomain setup.example.com and then I pointed setup.example.com to my new server. That works.
plesk setup folders as follows:
- example.com/
- example.com/httpdocs/
- example.com/subdomains/
- example.com/subdomains/setup/
- example.com/subdomians/setup/httpdocs
At the moment, setup.domain.com shows the index file in example.com/subdomains/setup/httpdocs
I would like setup.example.com to serve the files out of example.com/httpdocs
Here is what I have tried. - using a server alias in plesk instead of a subdomain - using a subdomain, then modifying setup.example.com's DocumentRoot in httpd.include to be example.com/httpdocs and restarting apache
I've searched everywhere, is there an easy way to achieve this?