Tag Info

New answers tagged

4

The cached reads aren't from the underlying device. They aren't representative at all of the performance of the different devices, neither of which could come anywhere near that kind of throughput. Ignore that number.


0

In general, when debugging state.highstate you should try: salt (from master) salt-call (from minion) Minion is reachable Ensure you don't have some firewall getting in the way, port 4505 and 4506 should be open to master. See also (http://docs.saltstack.com/en/latest/topics/tutorials/firewall.html) salt '*' test.ping -- Make sure minions respond to ping ...


1

Since allowing sudo nice is pretty much the same as granting root permissions, I would not consider doing that. I see two possible solutions: 1/ Allow the full command via sudo and pin it's parameters, like: %nicegroup ALL=/bin/nice --20 /bin/whatever But that works only if it is always the same program that needs nice'ing. 2/ Allow only renice via ...


1

No, there is not. nice is one of those commands that will run the process given in its own context - that is, if you allow sudo nice, any commands passed to nice will be executed as root. It may be possible to use /etc/limits.conf and use nice limits that way and run the script you want to be higher/lower priority, but it doesn't help with renice.


2

You are almost certainly running into mis-matched DNS entries. In your case using IPv6 transport, the AAAA and PTR records must match for the source address(es) of an MTA, also matching the EHLO name you are claiming to be.


0

the problem was i changed the htpasswd.users manually. so after running htpasswd.py -c -b /etc/nagios/htpasswd.users yamen yamen the problem gone :)


1

It might be worth to check whether the caching happens on the client and not server-side. Firefox will cache any 301 (PERMANLTY moved)-redirecteds for a few days. Use 302 (temporary) redirect. To fix "old targets" you can set up an html page redirecting to another domain (e.g. without www) etc. to prevent firefox from using its cache.


1

You can create a new iptables chain which can be separately flushed and refreshed. $ iptables -N AWS $ iptables -I INPUT 1 -j AWS From here, just add all of the IP ranges to the AWS chain. To refresh the rules, simply iptables -F AWS and re-populate. For example: $ iptables -F AWS $ iptables -A AWS -s 50.19.0.0/16 -j REJECT If you want to automate ...


1

This happens because the grub device map (/boot/grub/device.map) is not correct anymore, after exchanging a drive from raid. Simply run grub-mkdevicemap -n for updating the device map and grub-install for the corresponding devices.


-1

Create the file /etc/motd.tail and write there everything you need. The file /etc/motd will be generated upon every system boot based on the contents of /etc/motd.tail.


0

Node data is stored in the Chef server database in a format that isn't trivial to extract. If your Chef Server is still functional then you can simply run knife download nodes/NODENAME.json for a particular node or simply knife download nodes to download all of your nodes data to a nodes directory. If your Chef server is no longer functioning properly then ...


0

It appears to require a "bridged" adapter, which then lets you select the type of network card under the "advanced" settings. If you install the vbox extension pack then you can use an Intel network card to PXE boot. Other network adapters seem to PXE boot as well, probably without needing the extensions installed. The Intel card seems to be best ...


0

Your problem appears to be with WGA (Windows Genuine Advantage validation/anti-piracy). mpa.one.microsoft.com is one of the domains used by this DRM scheme, and an error message saying that domain can't be reached strongly suggests that a failure to reach this domain is causing the XP clients to kill the connection you're trying to make. So, your first ...


0

In case it helps... I had systematic 500 Internal Errors and had to struggle a lot to make php-fpm work with Apache and IPSCONFIG on a Ubuntu 14.04 LTS Trusty, build with Falko tutorial This is my working PHP-FPM configuration: /etc/php5/fpm/pool.d/www.conf user = www-data group = www-data listen.owner = www-data listen.group = www-data listen.mode = ...


1

You'll need to adjust the number of child processes you start to something lower. Depending on your configuration, that may be located in the httpd-mpm.conf or httpd.conf file.


0

Old question, but for posterity... I suspect you need to update your PATH to include the directory in which your ansible and ansible-playbook executables are located. Check if they're in your path on *nix: which ansible-playbook If not, fix that first, or try using the full path to the executable in the script. This may be of interest, ...


0

If you are using a text-editor to view the log files, that could be the cause; the "^@" characters may indicate that a line is too long (eg: in vim, turn on wrapping) The encoding is mismatched; either use a different text-editor to view the file, or change the encoding used by syslog.


1

For Debian OS Place your configuration file into /etc/openvpn, for example /etc/openvpn/client.conf. Prefix/comment out lines starting with "down" and "up" (#down and #up) - or delete them (these are calling external script) from client.conf Reload openvpn configuration /etc/init.d/openvpn reload /etc/openvpn/client.conf Check with ifconfig Do you see ...


1

What may be happening is that the Apache 2 init script has higher priority than the Nginx script. You can either change the priorities (but then the lower priority server may not "get its ports"), or just change the init scripts that get run. Change the priority with: update-rc.d nginx defaults update-rc.d apache2 start 30 2 3 4 5 . stop 70 0 1 6 Just ...


0

At high level and functionally they're mostly equivalent. In both cases the functionality will allow you to block or grant access to certain network ports to/from remote networks, regardless of how the server itself is configured. The underlying technology could be fundamentally different i.e. rather than the state-full packet inspection that is the more ...


0

More than likely your disks are not yet correctly attached to your instance. When you add additional disk space to an EC2 instance, your disk isn't immediately mounted on your Linux. ITKBCentral has a blog post with a tutorial on how to move your running instance to a larger disk. If this isn't your current situation, please provide more information about ...


3

You should use a different server block for each "type" of domain you manage, so that you don't have to use if statements to test the $host variable. # Domains to serve only via HTTPS server { server_name domaintoredirect1.tld domaintoredirect2; listen 80; return 302 https://$host$request_uri; } server { server_name domaintoredirect1.tld ...


0

Any attempt at creating an if statement to return the https version of the site will result in a redirect loop (as your condition, server_name is not changing after the redirect, so will always match the condition and redirect again...). With that said, what I would recommend would be to set up individual configuration files per virtual host/subdomain (I ...


2

As it turns out I was very close to a working config. Changing the redirect line to read redirect prefix / code 301 if old_url made it work as expected. I wrote a blog post which outlines the problem in a little more detail: http://fromanegg.com/post/105250564587/how-to-rewrite-and-redirect-with-haproxy


3

Create the directory /etc/default/grub.d if it doesn't exist already. Create a file /etc/default/grub.d/myextraoption.cfg adding to the variable you want (Append to it only, with an extra space. You want to be careful to not clobber or mangle any existing data there.): GRUB_CMDLINE_LINUX_DEFAULT="${GRUB_CMDLINE_LINUX_DEFAULT} extra-option" Run ...


0

I strongly suggest you to use micro-caching. Example : http://www.howtoforge.com/why-you-should-always-use-nginx-with-microcaching http://reviewsignal.com/blog/2014/06/25/40-million-hits-a-day-on-wordpress-using-a-10-vps/ I recently set up micro-caching on my box. With apache benchmark it holds up to 50 000 connections , CPU goes only to 6%. No timeouts, ...


0

I had the same problem and after searching I found out that this solution worked for me. I just had to rename all /dev/md/* devices to /dev/md* devices in /etc/mdadm/mdadm.conf and run update-initramfs -u to update the initramfs.


0

There are various methods of solving the issue you are encountering, you can try the following methods: METHOD 1 :: FRESH INSTALL sudo apt-get update sudo apt-get upgrade sudo apt-get remove --purge `dpkg -l | grep php | grep -w 5.5 | awk '{print $2}' | xargs` sudo apt-get purge apache2 php5 libapache2-mod-php5 sudo sed -i.bak "s/trusty/precise/g" ...


0

Get inspiration from answer at similar thread on SO The problem is this location directive for php-fallback location / { # This is cool because no php is touched for static content try_files $uri @rewrite; } location @rewrite { # Some modules enforce no slash (/) at the end of the URL # Else this rewrite block wouldn't be ...


0

You can try https://github.com/CHH/phpenv There is one more version manager https://github.com/wilmoore/php-version, but I would give a try to phpenv, because I am using rbenv and it works fine.


0

Down the Network Interface Without knowing what version of Ubuntu you're running, probably the simplest thing you could do is to just bring down the network interface (assuming you've only got one): sudo ifdown eth0 Use the following to bring it back up: sudo ifup eth0


0

Ok, problem solved. This is one of the biggest WTF's of my history with computers. By the way, one problem with sites like ServerFault is that if people don't know the answer right away they will lose interest and one has to solve their problem alone anyway. Not complaining, just a thought. So I tried to renew my IP using dhclient in verbose mode. ...


1

You could try using the following commands: (this is to disable all Internet Connections in Ubuntu) nmcli nm enable false and to enable it again: nmcli nm enable true Hope it helps! :)


0

Read the fine manual for error 2003 at http://dev.mysql.com/doc/refman/5.0/en/access-denied.html


0

Per https://help.github.com/articles/working-with-ssh-key-passphrases/#platform-mac it seems that Mac OSX caches the key in its own keychain. This in turn becomes an alternate data-source for ssh-agent.


2

It looks like you are getting the Enter password: prompt when connecting to your mysql server from your php server but nothing after that. If so then the mysql user account settings for the root user are probably denying the connection. Have a look at your root mysql user account and make sure the Host field is set to allow the connection. From (MySQL) ...


1

If you are trying to login using SSH with the root account, by default on Ubuntu it'll just say no like you are experiencing. If you have to allow root access (highly dangerous), change 'AllowRootLogin' to yes. A much better approach would be to create a new user: $adduser serveradmin $passwd serveradmin (enter the password for the new account) Then you ...


0

Add the following in to your /etc/varnish/default.vcl sub vcl_recv { remove req.http.X-Forwarded-For; set req.http.X-Forwarded-For = client.ip; } Update your apache configuration to change your CustomLog to the following: LogFormat "%{X-Forwarded-For}i %l %u %t \"%r\" %>s %b" varnishcombined CustomLog /var/log/example.com-access_log ...


3

Postfix refused relay to mydomain.com because isn't listed as local or virtual domain. The first step is define the domain class (postfix.org/ADDRESS_CLASS_README.html). myhostname is the machine name. mydestination is a list with domains hosted and users listed on passwd.


4

IIRC the PHP mail function on Linux typically defaults to using the sendmail command. Sendmail (or any local replacement SMTP server) needs not be explicitly configured nor be running as a daemon to work. For your messages to delivered reliably though, configuring and running sendmail (or replacement) in daemon mode is probably recommended. For instance ...


2

Why not replace the config file before installing the update? If it loads on restart, it won't use the "wrong" config file till after the update restarts it . . .


3

I recently wrote a long answer on why you'd want partitions in the first place here. In short: use LVM and not native partitions. two LVM partitions (one for swap and the second with a root file system containing everything) is not necessarily bad. more partitions can be beneficial, but if you do, start with minimally sized ones and leave space ...


0

If you need to supply a password and use ssl you can do something like this. #!/bin/bash -e USER=$1 MYPASSWORD=$2 IRC_SERVER=$3 IRC_PORT=$4 CHANNEL=$5 MSG=$6 ( echo NICK $USER echo USER $USER 8 * : $USER sleep 1 echo PASS $USER:$MYPASSWORD echo "JOIN $CHANNEL" echo "PRIVMSG $CHANNEL" $MSG echo QUIT ) | ncat --ssl $IRC_SERVER $IRC_PORT The script should ...


2

First off, I recommend not breaking the RAID. The performance and reliability gains of RAID-1 over 2 independent disks are worth it. If you're not particularly well-versed in Linux, the default partition scheme that Ubuntu suggests is generally fine. If you have other needs, then here's a good starting point. Adjust as necessary, for example if you have a ...


1

rx_fifo_errors = Total number of rx_queue_*_drops rx_queue_*_drops = Number of dropped packets per queue Sounds like the RX interrupts aren't allocating buffers fast enough, resulting in the adapter dropping packets. Check and increase the ring buffer. # ethtool -g eth0 Ring parameters for eth0: Pre-set maximums: RX: 4096 RX Mini: 0 RX Jumbo: ...


1

I know this is old, but my recommendation would be: for i in seq -f %02g 0 99 ; do mkdir $i ; done the -f %02g ensures it stays at least two characters, such as 00 or 99, and will still allow 3 character numbers past 99 so if you have 100 it will not become 001. It will be 00-99 100 instead of 001-100 such as the -w does.


0

As per the grub documentation, you may find it useful to edit the /etc/grub.d/40_custom file with grub-mkconfig; /etc/grub.d/40_custom is particularly useful for adding entire custom menu entries; simply type the menu entries you want to add at the end of that file, making sure to leave at least the first two lines intact. Perhaps you could grab one ...


1

What you're after is findmnt. For example: $ findmnt -rn -S UUID=xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx -o TARGET /mnt/mountpoint or $ findmnt -rn -S PARTUUID=xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx -o TARGET /mnt/mountpoint If nothing is mounted matching that UUID, nothing is output and the return code is 1 (failure), otherwise, the mountpoint is output ...


0

Very simple fix! Here's to anyone wondering how this can be done: 1) Open /etc/ufw/before.rules and insert this rule above COMMIT on the last line -A ufw-before-output -m owner --uid-owner {user} -p {protocol} --dport {port} -d {ip} -j ACCEPT Fill in the values for each as so: {user} - user you want to allow this (www-data) {protocol} - udp or tcp ...


2

Your application fails to start when both node.js and nginx are told to use port 3000 because only one application can bind to the port (considering they both are trying to bind to INADDR_ANY, i.e. 0.0.0.0). You can solve this situation by adding the proxy location in nginx config, like location /socket.io/ { (or something more appropiate to your setup) ...



Top 50 recent answers are included