Tag Info

Hot answers tagged

29

Update 09/04/2014 1:00AM EST Amazon has stated that all Elastic Load Balancers have been updated and are now longer vulnerable. They recommend rotating certs as well. Update 08/04/2014 2:56PM CST Amazon has stated that all Elastic Load Balancers except those in US-EAST-1 have been updated, and the vast majority of those in US-EAST-1 have been updated. ...


7

Danger! Please please do not just turn off your firewall here and think it's a solution. You were on the right track, except you turned a debugging step into a solution. What should we do instead? How do I truly fix this? Well you were right that you were only accepting everything from the lo interface. So we just need to allow this port to be accessed ...


5

It should be: ec2-54-208-90-77.compute-1.amazonaws.com ...which resolves correctly. Additionally, that record is not a CNAME. It's an A record. ~ dig ec2-54-208-90-77.compute-1.amazonaws.com ; <<>> DiG 9.8.3-P1 <<>> ec2-54-208-90-77.compute-1.amazonaws.com ;; global options: +cmd ;; Got answer: ;; ->>HEADER<<- ...


4

Logstash + ElasticSearch + Kibana. That is the combination you need. Sounds like you're looking towards AWS, so build up a modest EC2 cluster for this - perhaps a single logstash "router/broker" box and two Elasticsearch cluster nodes. Keep a reasonable amount of data "online" in ES indexes, and then archive older indexes to S3. ElasticSearch supports this ...


3

Is there a way to assign the public elastic IP address directly to the eth0 interface? No, there is not. Internet traffic to/from EC2 instances always traverse the Elastic IP 1:1 NAT infrastructure. If not, does anyone know a work around to using Vyatta in a VPC configured with an ipsec VPN? I have all manner of IPsec operating in VPC ...


3

Nah you can't, there's nothing to refer to anyway (e.g. logical ID). Just create your own main table ;-). This is probably one of the reason it can't be used: One way to protect your VPC is to leave the main route table in its original default state (with only the local route), and explicitly associate each new subnet you create with one of the custom ...


3

Unfortunately you can't do this globally. However, for each AWS product that supports it, you typically can limit access to a certain region. For instance, for EC2, you can do the following: { "Statement":[{ "Effect":"allow", "Action":"RunInstances", "Resource":"*", "Condition":{ "condition":{ "ec2:Region":"us-west-1" ...


2

Am I doing something wrong or is the AWS VPC ec2 instances not able to reach any of the AWS managed services (s3/sns/sqs) without a public route to the Internet Gateway in the routetable? That's correct. If you need to use those services, your instances will need EIPs or you'll need a NAT host in your VPC. A VPC is truly private, and it operates in ...


2

Let me ask - why do you care? I understand the logical reasoning, but it would not be something I would expect AWS to do - I would do those blocks in my application level. There are a lot of thigns that may go wrong and I would want full control over those blocks. So, for me it would be irrelevatn whether AWS does it - I would run my own decision logic.


2

Heavy usage Apache 2.4 on a t1.mirco instance is like humpty dumpty sitting on a couple toothpicks. He's too big and heavy! So of course he's squashed the toothpicks to pieces and rolled down the hill. You have at least two basic solutions to the root cause of this problem: Get more server resources, and less steal time. Either with AWS, or go to Digital ...


2

AWS optimize connections going out and not from inside to inside. If you're using the public IPs it's giving you, likely your connection is leaving AWS (or at least getting to the edge router), then routing back into AWS. If you have 2 instances that are internal, use the private IP. Then the connections go between local switches (that's a bit over ...


1

The A-record alias you create has to be the same as the name of the bucket, because virtual hosting of buckets in S3 requires that the Host: header sent by the browser match the bucket name. There's not really another practical way in which virtual hosting of buckets could be accomplished... the bucket has to be identified by some mechanism, and that ...


1

You are on the right track. ELBs can be SSL termination proxies. In this scenario you set up HTTPS one time on the ELB and install the certificate there and not on the instances. Traditionally for SSL you needed the web server name to match the CN in the cert. This hasn't been the case for a while now. See ...


1

You could the innodb_file_per_table (https://dev.mysql.com/doc/refman/5.5/en/innodb-parameters.html#sysvar_innodb_file_per_table). Unfortunately you will get a file per table and not per DB (I don't know about an option for this in MySQL). With this you could link each table-file per symlink to different EBS-Volumes which are mounted to the instance. You ...


1

I've corrected the problem by removing perl path statements in the windows environment variables. They were at the beginning of the string from an old install. The strawberry install added it's own path statements at the end. This allowed the new strawberry intall to search the paths in the correct order.


1

Try the following link. Add a config file in your root inside .ebextensions directory. Then add this. files: "/etc/httpd/conf.d/vhost.conf": mode: "000644" owner: root group: root encoding: plain content: | NameVirtualHost *:80 <VirtualHost *:80> DocumentRoot "/var/app/current/" <Directory ...


1

I had this same problem, and I solved it by putting the elastic IP into my /etc/hosts with a name entry that matched my server hostname and the name I was using in DNS, and ran chef-server-ctl reconfigure. Edit: this solution then broke cookbook uploads for me, and I had to relearn the good old AWS lesson that it's the private IP you must put in /etc/hosts. ...


1

Right now, you can't: https://forums.aws.amazon.com/thread.jspa?messageID=505301&#505301 ... at least they are aware of people wanting this. My guess is this will be available when they add custom Option Groups for PostgreSQL.


1

I just ran the upgrade to 2012 R2 Datacenter and it successfully activated against Amazon's KMS server. At first I tried Standard since the 2012 AMI was Standard, but it would not activate. It may not yet be officially supported, but it seems to work fine. I did it by mounting the ISO I downloaded from Microsoft VLC.


1

S3 is a storage service. Code of any sort will not execute on it. There are cheaper options than AWS if you just want to run a tiny, non-critical contact form script. Shared hosting might run you as low as a buck a month, or a small Digital Ocean instance runs $5/month. If your current website has PHP support and it's still not working, though, you're ...


1

After installation, Amazon doesn't do anything to your instances anymore so I don't think they have functionality to provide for this scenario. You could hack up a script to loop over all instances and check if his key is there (it's not pretty, but it'll work :) ): for INSTANCE in $(ec2-describe-instances | grep INSTANC | grep running | awk '{print $4;}') ...


1

You could simply use a master-slave configuration for MySQL, simply specify the SSD machine as master, and replicate off of it. However, I strongly recommend using RDS unless you have a truly unique setup. RDS allows for provisioned IOPS and will take care of a lot of things you need to manually do for a mysql server.


1

This trick has saved me a few times. If you don't know if the machine is live or not... use these steps. Find the public IP address of your local machine (PC you are on). In the AWS EC2 Dashboard, find the security group attached to your VPS/EC2. Add a rule for inbound/outbound ALL Traffic to/from your IP address to the public address of the EC2. ...


1

Since you're on the free tier, I'm assuming you're using a t1.micro instance. This particular instance lacks the ephemeral storage (the gigs of "temporary" space) and must be EBS-backed, as you see there. Ephemeral storage is very useful for worker servers but very bad for permanent storage because this gets wiped if the server is shut down or terminated. ...


1

You can use the instance metadata documented here http://docs.aws.amazon.com/AWSEC2/latest/UserGuide/AESDG-chapter-instancedata.html below I wrote a simple piece of java code which prints the running instance-id import java.net.URL; import java.net.URLConnection; import java.util.Scanner; public final class Main { public static void main(String[] args) ...


1

If I understand the documentation correctly, when you have versioning enabled deleting the file simply reverts the "latest" version back one version number. This however does not give the ability to restore an entire bucket. This makes the previous versions in S3 not suitable for your needs (i.e, recovery from deletion). Keep a backup someplace else as well ...


1

Well, it basicly cannot be done. Question is, why would you want to do that? You have it always on the same hostname, so it is always accessible under the same endpoint all the time from the Internet. The only workaround, which we once used, when we wanted to grant external developer access to RDS instance, was making stunnel on development EC2 instance with ...


1

As EEAA mentioned, you can't actually assign the elastic IP to an interface on your instance. That's not how EC2 publicly-routed IPs work. I'm not sure how much it will help you, but if you do want to know the IP programatically, you can query it from the metadata "API": curl -s http://169.254.169.254/latest/meta-data/public-ipv4 If an elastic IP is ...


1

It sounds like what they expect you to add is a record like the following (as written in the normal master file format): _amazonses IN TXT "the-very-special-key" You don't specify what this management tool that you refer to offers in terms of UI but the name where you want to add the record is _amazonses (if specified as relative to the zone), the type is ...


1

You need to be more specific. Can an EBS volume be detached from one EC2 instance and attached to another EC2 instance? Yes, as long as both instances are part of the same Availability Zone. EBS volumes can be thought of the same way as a portable Hard Drive. Now the caveat to this part of your question is, is this EBS volume the root volume, or was it added ...



Only top voted, non community-wiki answers of a minimum length are eligible