Permalink
Browse files

Merge branch 'source' of github.com:promet/promet.github.com into source

  • Loading branch information...
2 parents 28143e4 + 88d3070 commit f4f24563413f769466397bec8b802c6bd6e8dfd6 Greg Palmier committed Dec 10, 2012
@@ -9,7 +9,7 @@
<meta name="author" content="Promet OPS Team">
- <meta name="description" content="Marius Ducea <iframe src="http://markdotto.github.com/github-buttons/github-btn.html?user=mdxp&type=follow&count=true" allowtransparency="true" &hellip;">
+ <meta name="description" content="Marius Ducea <iframe src="http://ghbtns.com/github-btn.html?user=mdxp&type=follow&count=true" allowtransparency="true" &hellip;">
<!-- http://t.co/dKP3o1e -->
@@ -73,7 +73,7 @@ <h1 class="entry-title">Marius Ducea</h1>
<div id="author-top-links" style="margin:-10px 0 8px 0;">
- <iframe src="http://markdotto.github.com/github-buttons/github-btn.html?user=mdxp&type=follow&count=true"
+ <iframe src="http://ghbtns.com/github-btn.html?user=mdxp&type=follow&count=true"
allowtransparency="true" frameborder="0" scrolling="0" width="165px" height="20px"></iframe>
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
View
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
@@ -7,7 +7,7 @@
<div id="author-top-links" style="margin:-10px 0 8px 0;">
{% if author.github %}
- <iframe src="http://markdotto.github.com/github-buttons/github-btn.html?user={{ author.github }}&type=follow&count=true"
+ <iframe src="http://ghbtns.com/github-btn.html?user={{ author.github }}&type=follow&count=true"
allowtransparency="true" frameborder="0" scrolling="0" width="165px" height="20px"></iframe>
{% endif %}
@@ -57,7 +57,7 @@ You should see Solr access on your Tomcat admin/manager page (`http://hostname:8
**3.) Connect Drupal Site to Solr (multi-core setup)**
I already have a Drupal site with ApacheSolr module installed and SolrPHPclient library.
-
+<!--more-->
```
cp /path/drupal_site/sites/all/modules/apachesolr/schema.xml /usr/share/tomcat6/solr/conf/schema.xml
cp /path/drupal_site/sites/all/modules/apachesolr/solrconfig.xml /usr/share/tomcat6/solr/conf/solrconfig.xml
@@ -20,7 +20,7 @@ I'm sure almost everyone is aware at this point on the **failure** that caused A
The failure was specifically related to the **EBS** drives that made customer instances non responding, but also prevented them to start or stop new instances with the same EBS volumes (that probably 99% have tried immediately as they got paged). There are some sites that had failover mechanisms, but if they were in the same availability zone it was useless (something that looked like a good solution and fast and cost effective). Others, many startups, found out that they had no such mechanism at all, and that they depended way too much on the Amazon reliability. Until this issue, Amazon had a great uptime record; there were many issues but with individual instances, but not such a global issue. You would expect people running their application in the cloud to expect failures and be prepared and I'm sure most of them are compared with applications deployed in the regular datacenter, but apparently there is still much work to be done.
Overall I believe this showed (if we needed a reminder), that failures can happen and anyone can suffer from such a problem (Google had problems, Facebook the same, and Twitter is most of the time down, and now was just Amazon's turn). We need to be prepared and build and architect our applications with this in mind and be ready to failover. A great example of this is the twilio application design: [http://www.twilio.com/engineering/2011/04/22/why-twilio-wasnt-affected-by-todays-aws-issues/](http://www.twilio.com/engineering/2011/04/22/why-twilio-wasnt-affected-by-todays-aws-issues/)
-
+<!--more-->
Also I think Amazon will learn many things from this event, and hopefully one of them will be to _better communicate with their clients_. They can definitely improve on this, and not leave people to go to twitter or ec2 forums for the best updates on such problems instead from the source.
I've also seen different providers (I will not give names but they are everywhere on twitter) come out and offer their services to people hit by this. I don't believe this is the best approach to sell your very performant cloud solution or barebone datacenter servers or whatever, and probably it does worse for their reputation making them look like coyotes on a fallen prey. I'm sure like me, most people felt disgusted on their offers at this time, and it will definitely not make me look at their offering again.
@@ -31,7 +31,7 @@ aptitude install libpam-pwdfile
```
Next create your password file (example: _/etc/vsftpd_users.conf_) which contain the list of virtual users and password hash. You can use [htpasswd](http://httpd.apache.org/docs/2.0/programs/htpasswd.html) that's included in Apache or go to this website, [htaccesstools.com](http://www.htaccesstools.com/htpasswd-generator/) to generate the list of users and passwords.
-
+<!--more-->
Once you have the list of users (example: _demo1_ and _demo2_), you have to create their virtual directory as well, let's say I want to set ftp users home directory inside _/home/ftpusers_. This will be their home directory and they can't see or navigate outside of their home dir. Make sure they're writable as well.
```
@@ -38,7 +38,7 @@ vi /etc/rsnapshot.conf
```
Here's my rsnapshot config changes and/or additions:
-
+<!--more-->
```
snapshot_root /backups/ #this is a mounted partition
interval daily 1 # i removed all other schedule and one daily copy only
@@ -32,7 +32,7 @@ Overview of "bare-metal" restore process:
create, format, and mount logical volumes to use on the vm that you want to restore or re-create, based on its original vm disk properties such as size, names, etc.
copy all files from backup location to the newly mounted lvm disk
-
+<!--more-->
adjust necessary xen vm configs (if any, like new ip address) and start/create the vm using the volume with restored contents or files, and you're done.
In the restore example below I'll refer to the vm as "domain.com" and we will restore the backups to a different volume with the same config as the original. Then we'll start the vm, domain.com, using the restored volume.
@@ -31,7 +31,7 @@ The core of this funcionality is provided by the sendgrid gem which originates f
#### Redmine Configuration File
If your Redmine install is something like /var/www/redmine, your configuration file will be /var/www/redmine/config/configuration.yml. If you have a relatively clean Redmine install, chances are you might not have one or it's a default template with a lot of examples. You might find an example detailing how to integrate with GMail in the example file. Our set up will be similar. Paste the following into your configuration.yml file and change the necessary parameters.
-
+<!--more-->
```
production:
delivery_method: :smtp
@@ -0,0 +1,59 @@
+---
+layout: post
+title: "HowTo: Bulk Import DNS Zones to Softlayer DNS using Cpanel"
+date: 2012-11-13 01:08
+comments: true
+author: gerold-mercadero
+published: true
+status: publish
+categories:
+- cpanel
+- HowTo
+- dns
+---
+
+This guide covers the process for doing bulk import of existing DNS zones from a Cpanel or BIND server to Softlayer's DNS servers (*ns1/ns2.softlayer.com*). This is useful if you need to add hundreds of zones, if for example you are moving your DNS hosting servers to Softlayer's free DNS hosting. Currently, Softlayer doesn't have a tool to do bulk import of zones with existing records. Although you can create a default zone template and bulk add zones but this doesn't fit for existing zones.
+
+### Requirements:
+
+1.) DNS zones on a Cpanel server (WHM or DNSOnly).
+
+If you are not using a Cpanel server then you can stand up a Cpanel VPS this process and install Cpanel DNSOnly as it is free or install Cpanel WHM and automatically it will have 15-day trial license, or you can get a one month WHM VPS license for $15/month.
+DNSOnly should be fine and easier/faster to install but you cannot view the DNS zones from its UI, although you can view or edit the zone files directly (*ex: /var/named/domain.db*).
+Then import your zones to Cpanel server. There are lots of guides or discussions online in doing this. Some mentioned that you can move the named.conf and zone files as-is and adjust the formatting to match Cpanel's.
+In our case we were running Cpanel server so we copied straight our "/var/named/*.db" files to Cpanel DNSOnly to import zones.
+
+2.) Softlayer API key for your account.
+
+ You will need this later to connect your Cpanel DNS server in a cluster setup with Softlayer's DNS servers. Cpanel have DNS cluster functionality with SoftLayer and VPS.NET.
+
+### The Import Process:
+
+1.) Configure your Cpanel DNS cluster to connect to Softlayer's DNS servers.
+<!--more-->
+- Login to Cpanel WHM or DNSOnly, then go to *Cluster/Remote Access* ->> *Configure Cluster*.
+- select *Enable DNS clustering*
+- On *Backend Type* dropdown, you will see *cPanel* only. Then click on *(Show All)* link next to it and the page will reload and you will see two more options: *Softlayer* and *VPS.NET*. Select *Softlayer* and click on *Configure button.
+[![cluster-setup](http://linuxsysadminblog.com/images/2012/cluster-setup.png)](http://linuxsysadminblog.com/images/2012/cluster-setup.png)
+- On *SoftLayer DNS Remote Configuration* page, enter your Softlayer API user and key. On *DNS role* choose *Synchronize changes* from dropdown list and click on *Submit*.
+[![softlyer-dns-config](http://linuxsysadminblog.com/images/2012/softlyer-dns-config.png)](http://linuxsysadminblog.com/images/2012/softlyer-dns-config.png)
+
+2.) Synchronize DNS Records across clusters. DNS Cluster is now your Cpanel server and Softlayer servers.
+
+- From Cpanel WHM or DNSOnly admin, go to * DNS Functions* ->> *Synchronize DNS Records*
+- Select *Synchronize all zones to all servers* and check the box that says *Synchronize zones that are not configured on this server*. Then hit *Synchronize* button.
+[![sync-dns](http://linuxsysadminblog.com/images/2012/sync-dns.png)](http://linuxsysadminblog.com/images/2012/sync-dns.png)
+- The dns sync process will run and it will take some time depending on the number of zones and records that you have. Wait until it's finish and you're done.
+- Once this starts you can login to your Softlayer portal and go to *Public Network* ->> *Domain Name System* and you will start seeing zones and see the progress on how many are imported.
+
+*That is all. You can shutdown or remove your Cpanel DNSOnly/WHM server when if not needed.*
+
+### Notes:
+
+- Cpanel have DNS cluster functionality with SoftLayer and VPS.NET so you might be able to use this guide to bulk import zones to VPS.NET.
+- At this time of writing, we encountered a problem with Cpanel DNS cluster integration with Softlayer and we cannot modify the dns records that were imported initially. When we update them it reverts to the original records from initial import. We disabled the DNS cluster setup from Cpanel and we update our DNS zones/records right from Softlayer portal.
+- Above info refers to BIND server as the default DNS server on Cpanel.
+
+
+
+
@@ -0,0 +1,39 @@
+---
+layout: post
+title: "Xen: Add Extra Partitions to Guest OS"
+date: 2012-11-21 18:55
+author: gerold-mercadero
+published: true
+comments: true
+categories:
+- xen
+---
+
+
+Here's one way to add and mount additional LVM partition in xen guest machines.
+
+1.) Create logical partition and format it. Example below creates "40GB" partition and "ext3" format in "vg0" volume group.
+
+```
+lvcreate -L 40G -n extra_storage vg0
+mkfs -t ext3 -v /dev/vg0/extra_storage
+```
+
+2.) Update xen config to add new partition to guest os. Assign device name to it, example "xvda3". Then re-create the VM to recognize the new partition/device. Note: reboot doesn't read the new xen configuration.
+
+```
+disk = [
+ 'phy:/dev/vg0/extra_storage,xvda3,w',
+```
+
+3.) After you reboot, login to VM and mount the new partition. New drive or partition shoud be in "/dev/xvda3."
+
+```
+ mount /dev/xvda3 /extra_space
+```
+
+Add it to "/etc/fstab" to automount on reboot.
+
+```
+ /dev/xvda3 /extra_space ext3 noatime,nodiratime,errors=remount-ro 0 1
+```
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.

0 comments on commit f4f2456

Please sign in to comment.