Search Results

Search found 24427 results on 978 pages for 'ec2 api tools'.

Page 19/978 | < Previous Page | 15 16 17 18 19 20 21 22 23 24 25 26  | Next Page >

  • Drive Mapped On Amazon EC2 Instance Startup Disconnected After Logon

    - by jsn
    I am launching an Amazon EC2 instance (Windows Server 2012), and on startup (though User Data field), it runs this powershell script: <powershell> NET CONFIG SERVER /AUTODISCONNECT:-1 # clear all prior connections net use * /delete /y > C:\delLog.txt 2>&1 # mount new drive net use R: \\dbHost\share /user:username pass /persistent:yes > C:\useLog.txt 2>&1 ipconfig /all > C:\ipLog.txt </powershell> When it launches and I connect to it through RDP, it shows in explorer "Diconnected Network Drive (R:). If I double click it, error message displays "R:\ is not accesible. Access id denied". Normally, it would ask me for credentials to reconnect. I need for this drive to be connected through the duration of the instance. delLog.txt contents: You have these remote connections: T: \\dbHost\share Continuing will cancel the connections. The command completed successfully. useLog.txt contents: The command completed successfully. ipLog.txt contents as expected. The net use commands works fine by itself, it connects. Anyone have any idea what could be wrong? There is only one account on these machines - Administrator. It is a samba share to a Linux server on a private network.

    Read the article

  • IPSec Tunnel to Amazon EC2 - Netkey, NAT, and routing problem

    - by Ernest Mueller
    Hey all, I'm working on getting an IPSec VPN working between Amazon EC2 and my on-premise. The goal is to be able to safely administer stuff, up/download data, etc. over that tunnel. I have gotten the tunnel up in openswan between a Fedora 12 instance with an elastic IP and a Cisco router that's also NATted. I think the ipsec part is OK, but I'm having trouble figuring out how to route traffic that way; there's no "ipsec0" virutal interface because on Amazon you have to use netkey and not KLIPS for the vpn. I hear iptables may be required and I'm an iptables noob. On the left (Amazon), I have a 10. network. Box 1 is privately 10.254.110.A, publically IP 184.73.168.B. Netkey tunnel is up. Box 2 is publically 130.164.26.C, privately 130.164.0.D And my .conf is: conn ni type= tunnel authby= secret left= 10.254.110.A leftid= 184.73.168.B leftnexthop= %defaultroute leftsubnet= 10.254.0.0/32 right= 130.164.26.C rightid= 130.164.0.D rightnexthop= %defaultroute rightsubnet= 130.164.0.0/18 keyexchange= ike pfs= no auto= start keyingtries= 3 disablearrivalcheck=no ikelifetime= 240m auth= esp compress= no keylife= 60m forceencaps= yes esp= 3des-md5 I added a route to box 1 (130.164.0.0/18 via 10.254.110.A dev eth0) but that doesn't do it for predictable reasons, when I traceroute the traffic's still going "around" and not through the vpn. Routing table: 10.254.110.0/23 dev eth0 proto kernel scope link src 10.254.110.A 130.164.0.0/18 via 10.254.110.178 dev eth0 src 10.254.110.A 169.254.0.0/16 dev eth0 scope link metric 1002 Anyone know how to do the routing with a netkey ipsec tunnel where both sides are NATted? Thanks...

    Read the article

  • APC fragmentation on EC2 Micro for Wordpress + W3TC

    - by Maarten Provo
    I'm trying to optimize APC for my Amazon EC2 Micro server running one Wordpress-site with W3TC. I've started with the settings advised by TechZilla in another topic but I keep getting high fragmentation with 50% of space being free. I've uploaded an image to http://www.maartenprovo.be/downloads/apc.jpg but I can't post it here since I need at least 10 reputation. What values can I optimize to prevent fragmentation? [apc] apc.enabled=1 apc.shm_segments=1 ;32M per WordPress install apc.shm_size=164M ;Leave at 2M or lower. WordPress does't have any file sizes close to 2M apc.max_file_size=2M ;Relative to the number of cached files apc.num_files_hint=1000 ;Relative to the size of WordPress apc.user_entries_hint=4096 ;The number of seconds a cache entry is allowed to idle in a slot before APC dumps the cache apc.ttl=7200 apc.user_ttl=7200 apc.gc_ttl=3600 ;Auto update chache files on change in WP-ADMIN or W3TC apc.stat=1 ;This MUST be 0, WP can have errors otherwise! apc.include_once_override=0 ;Only set to 1 while debugging apc.enable_cli=0 ;Allow 2 seconds after a file is created before it is cached to prevent users from seeing half-written/weird pages apc.file_update_protection=2 ;Ignore files apc.filters apc.slam_defense = 0 apc.write_lock = 1 apc.cache_by_default=1 apc.use_request_time=1 apc.mmap_file_mask=/var/tmp/apc.XXXXXX apc.stat_ctime=0 apc.canonicalize=1 apc.write_lock=1 apc.report_autofilter=0 apc.rfc1867=0 apc.rfc1867_prefix =upload_ apc.rfc1867_name=APC_UPLOAD_PROGRESS apc.rfc1867_freq=0 apc.rfc1867_ttl=3600 apc.lazy_classes=0 apc.lazy_functions=0

    Read the article

  • IPSec Tunnel to Amazon EC2 - Netkey, NAT, and routing issue

    - by Ernest Mueller
    I'm working on getting an IPSec VPN working between Amazon EC2 and my on-premise. The goal is to be able to safely administer stuff, up/download data, etc. over that tunnel. I have gotten the tunnel up in openswan between a Fedora 12 instance with an elastic IP and a Cisco router that's also NATted. I think the ipsec part is OK, but I'm having trouble figuring out how to route traffic that way; there's no "ipsec0" virutal interface because on Amazon you have to use netkey and not KLIPS for the vpn. I hear iptables may be required and I'm an iptables noob. On the left (Amazon), I have a 10. network. Box 1 is privately 10.254.110.A, publically IP 184.73.168.B. Netkey tunnel is up. Box 2 is publically 130.164.26.C, privately 130.164.0.D And my .conf is: conn ni type= tunnel authby= secret left= 10.254.110.A leftid= 184.73.168.B leftnexthop= %defaultroute leftsubnet= 10.254.0.0/32 right= 130.164.26.C rightid= 130.164.0.D rightnexthop= %defaultroute rightsubnet= 130.164.0.0/18 keyexchange= ike pfs= no auto= start keyingtries= 3 disablearrivalcheck=no ikelifetime= 240m auth= esp compress= no keylife= 60m forceencaps= yes esp= 3des-md5 I added a route to box 1 (130.164.0.0/18 via 10.254.110.A dev eth0) but that doesn't do it for predictable reasons, when I traceroute the traffic's still going "around" and not through the vpn. Routing table: 10.254.110.0/23 dev eth0 proto kernel scope link src 10.254.110.A 130.164.0.0/18 via 10.254.110.178 dev eth0 src 10.254.110.A 169.254.0.0/16 dev eth0 scope link metric 1002 Anyone know how to do the routing with a netkey ipsec tunnel where both sides are NATted? Thanks...

    Read the article

  • Amazon EC2 instance missing Network Interface

    - by Sergiks
    I am running Linux on a t1.micro instance at Amazon EC2. Once I noticed bruteforce ssh login attemtps from a certain IP, after litle Googling I issued the two following commands (other ip): iptables -A INPUT -s 202.54.20.22 -j DROP iptables -A OUTPUT -d 202.54.20.22 -j DROP Either this, or maybe some other actions like yum upgrade perhaps, caused the follwing fiasco: after rebooting the server, it came up without the Network Interface! I only can connect to it through AWS Management Console JAVA ssh client - via local 10.x.x.x address. Console's Attach Network Interface as well as Detach.. are greyed out for this instance. Network Interfaces item at the left does not offer any Subnets to choose from, to create a new N.I. Please advice, how can I recreate a Network Interface for the instance? Upd. The instance is not accessible from outside: cannot be pinged, SSH'ed or connected by HTTP on port 80. Here's the ifconfig output: eth0 Link encap:Ethernet HWaddr 12:31:39:0A:5E:06 inet addr:10.211.93.240 Bcast:10.211.93.255 Mask:255.255.255.0 inet6 addr: fe80::1031:39ff:fe0a:5e06/64 Scope:Link UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:1426 errors:0 dropped:0 overruns:0 frame:0 TX packets:1371 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:152085 (148.5 KiB) TX bytes:208852 (203.9 KiB) Interrupt:25 lo Link encap:Local Loopback inet addr:127.0.0.1 Mask:255.0.0.0 inet6 addr: ::1/128 Scope:Host UP LOOPBACK RUNNING MTU:16436 Metric:1 RX packets:0 errors:0 dropped:0 overruns:0 frame:0 TX packets:0 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 RX bytes:0 (0.0 b) TX bytes:0 (0.0 b) What is also unusual: a new micro instance I created from scratch, with no relation to the troubled one, was not pingable too.

    Read the article

  • Need a recommendation for shared storage on auto-scaling ec2 w/ scalr

    - by john h.
    I have come across so many answers to this question that I am completely lost! I am moving our 2 sites to a load balanced ec2 system with scalr as our cloud manager. Now the question is coming up about persistent storage for the user's uploaded content and other files. Could someone please give me a suggestion and possible a link to a tutorial for the following setup and goals. 2 websites (1 Forum, 1 ecommerce). 1 LB 1 App server (to scale out to as many as needed) 1 DB server (to scale out to as many as needed) Our sites will need to autoscale and according to what I am learning about scalr, that means as new instances load up, I need to run a script to set the basics up on that server (git,php mods, pull site from git, move keys, etc) What I don't understand is how should I handle user uploaded content like profile pictures, avatars, product images, themes, etc... Do I mount an EBS or s3fs folder to hold the websites (maybe /var/www/websitefolder) or do I do something like mount the avatar folders /var/www/websitefolder/images/avatars) I am not sure where to go with this. Could someone give me some detailed help? -John

    Read the article

  • Prevent Amazon EC2 Time zone from reverting back on yum update

    - by D.Tate
    I use an Amazon EC2 server instance that runs a distro called Amazon Linux AMI. (I've read that it is based on CentOS/Red Hat). My specific version is the 2012.09 release. Anyway, I was able to change the time zone about a week ago from the default UTC to America/New_York (which is EST/EDT). The command I used to change it was: ln -sf /usr/share/zoneinfo/America/New_York /etc/localtime ...thanks to this other Server Fault question. At that point, I was able to run date from the the command line, and it correctly displayed the EDT time. And even after EDT "fell back" to EST this past Sunday, I was pleased to find that running date still produced the correct local time. So that was great. However, after running a yum update yesterday, it seems that my time zone got reverted back to plain 'ol UTC. I even checked the last modified time of /etc/localtime file, and indeed it confirmed that it had been modified around the same time I had updated. Is there any way to prevent this from happening again, or will I be stuck resetting the time zone every time I do a yum update?

    Read the article

  • Map a URL bought with Dreamhost to Amazon EC2 (AWS)

    - by Edan Maor
    I have several URLs I purchased through Dreamhost. I'm starting to use Amazon's AWS, and I'd like to map the URLs to Amazon. This is something of a silly question, and I've already done the same thing several times to other services (mapping from Dreamhost to WebFaction). But for some reason when I tried to find the proper way to do the same mapping to Amazon, I find a lot of detailed writing talking about whether I should be using CNAME or A records, etc. So I wanted to ask in the simplest possible terms and hopefully get a simple, concrete answer: I bought a URL from Dreamhost, I have an EC2 server running on AWS (to which I already mapped an Elastic IP address). How do I make the URL map to AWS? And if there are several options, which one should I effectively be using? P.S. Meta-question - why are things so much more difficult with AWS? When I search Google for "Move from Dreamhost to WebFaction, I get very simple answers on how to do the mapping. In what way is AWS different?

    Read the article

  • EC2 configuration for medium load service on Django

    - by Luberg
    I have created a very basic Django application which puts an email to the database (Coming soon page for a startup). I launched a t1.micro instance to try out which load it can carry out. Nginx+FastCGI from Django+sqllite/postgres - tried both. blitz.io test gave me a pretty unhappy result (just 100 users within 1 minute): This rush generated 542 successful hits in 1.0 min and we transferred 809.01 KB of data in and out of your app. The average hit rate of 8.81/second translates to about 761,612 hits/day. You got bigger problems though: 87.28% of the users during this rush experienced timeouts or errors! I tried both to put varnish, disabled Debub mode in django and started fastcgi in threaded mode - nothing helps. This is not gonna be a super highload page - just a coming soon page to save email of subscribers, it should carry at least 500-1000 users at the same time in peak... I believe t1.micro is super small for that, but I also have tried small instance - not better result.. Please let me know should I use something different from Amazon EC2, or to pick smth better than t1.micro, or I that is definetely a configuration issues?...

    Read the article

  • Hadoop streaming job on EC2 stays in "pending" state

    - by liamf
    Trying to experiment with Hadoop and Streaming using cloudera distribution CDH3 on Ubuntu. Have valid data in hdfs:// ready for processing. Wrote little streaming mapper in python. When I launch a mapper only job using: hadoop jar /usr/lib/hadoop/contrib/streaming/hadoop-streaming*.jar -file /usr/src/mystuff/mapper.py -mapper /usr/src/mystuff/mapper.py -input /incoming/STBFlow/* -output testOP hadoop duly decides it will use 66 mappers on the cluster to process the data. The testOP directory is created on HDFS. A job_conf.xml file is created. But the job tracker UI at port 50030 never shows the job moving out of "pending" state and nothing else happens. CPU usage stays at zero. (the job is created though) If I give it a single file (instead of the entire directory) as input, same result (except Hadoop decides it needs 2 mappers instead of 66). I also tried using the "dumbo" Python utility and launching jobs using that: same result: permanently pending. So I am missing something basic: could someone help me out with what I should look for? The cluster is on Amazon EC2. Firewall issues maybe: ports are enabled explicitly, case by case, in the cluster security group.

    Read the article

  • File permission woes on an Ubuntu ec2 instance

    - by Pardoner
    I've set up an amazon ec2 instance and I'm have some file permission issues. I've created myself a new user and added myself to the following groups: adm:x:4:me,ubuntu sudo:x:27:me www-data:x:33:me,www-data ssh:x:108:me admin:x:111:me ubuntu:x:1000:www-data,me me:x:1001:me but when I cd /var/www I can't do simple commands without doing sudo. So I chown -R www-data:www-data /var/www to ensure that I'm in the owning group but I still have to type sudo for everything. If I sudo su www-data it works fine. Since I'm in the www-data group shouldn't I have the same privilages as www-data? One strange thing I'm noticing is that when I ls -l it list the owner but not the group names. Could this possibly be part of the issue? Is is posible for a directory to not be part of a group? drwxr-xr-x 4 www-data 4.0K Oct 24 16:39 . drwxr-xr-x 14 root 4.0K Oct 10 16:58 .. drwxrwxr-x 9 www-data 4.0K Oct 23 04:03 admin.mywebsite.com drwxrwxr-x 2 www-data 4.0K Oct 4 00:29 mywebsite.com drwxrwxr-x 9 www-data 4.0K Oct 23 04:03 staging.mywebsite.com Edit : It appears I had some alias messing with my ls command. By calling \ls -l I can see that all my files are in the correct group.

    Read the article

  • EC2 kernel decision and issues with creating a new machine with my AMI

    - by roacha
    I could really use some advice. I started a new instance on EC2 using Amazon's AMI and during the deployment process I selected a Kernel ID of "Use Default". I then configured my server the way that I wanted to and took a snapshot of it. I then created my own AMI to create new servers with. When I try and create a new server with this AMI the server fails to start and I get the error: EXT3-fs: sda1: couldn't mount because of unsupported optional features (240). Which appears to happen because I am selecting a kernel id of "Use default" again when building my second server. I have read that in order for this to work I need to choose the same kernel id that was used in my original server. I have deleted my original server and don't know what it was using. What is the best process to follow in order to not have these issues? Should I choose "Use Default" for my original server? How do you know which kernel it selected? Then should I just document this and always specify this during the deployment of my next servers using my custom AMI? OR should I choose a custom kernel id during the initial build and always use this one moving ahead hoping Amazon never retires it? Thanks for any advice!

    Read the article

  • Amazon EC2: Instances, IPs and a wordpress blog (LAMP)

    - by JustinXXVII
    I had a link to my blog posted on Reddit yesterday and MySQL crashed on my EC2 Micro instance. I know I didn't have that many visitors because I used a marketing link that tracks hits. The link got 167 hits over the course of the last 18 hours, and MySQL crashed twice. So anyway, 167 visits is not a lot, so I've done some short term optimizations like restricting the number of Apache threads to limit the MySQL calls. I also set up WP Super Cache to serve static content. Soon I'm going to offload all of my images to S3 or CloudFront. So this leads me to my question. If this doesn't seem to help, and if i have another traffic "spike", how do AMIs work when you have a MySQL database? I think I understand that if you have more than one instance and assign the same Elastic IP to both of them, the incoming traffic gets distributed among both. But what happens when the MySQL database gets updated on one of the instances? I just need to wrap my mind around what happens when I create an AMI and then launch a new instance to help with traffic. Thanks for your suggestions.

    Read the article

  • Replicated filesystem and EC2 MySQL

    - by El Yobo
    I'm currently investigating migrating our infrastructure over to run on Amazon's EC2 and am trying to figure out the best way to set up a MySQL service. I'm leaning towards running our own MySQL instances, rather than going with Amazon's RDS, but am still considering the best approach for performance and cost on the instance itself. In order to have persistent data, the MySQL data needs to be on an EBS volume (with some form of striped RAID, e.g. RAID0 or RAID10) to improve persistence. However, EBS IO is limited by the network interface (gigabit, so a theoretical maximum of 128 MB/s), while the ephemeral volumes have no such problem. I did see a suggestion for running two MySQL servers on an instance, with a master running on the ephemeral disk (which we would also RAID) and a slave storing changes to an EBS volume, but this has some additional overhead and complexity (two servers). What I was imagining is using some form of replicated file system such that I could have a filesystem on top of a RAID0 of ephemeral volumes to maximise performance all changes from the above immediately replicated to another RAID1 volume backed by multiple EBS volumes to ensure no data loss The advantages of this would be best possible IO performance for the DB server; no network delay in IO decreased IO on EBS volumes (as all read IO will be done on the ephemeral volumes) so decreased cost good data security, as it's backed onto redundant EBS volumes However, I haven't seen an appropriate system to replicate all changes from one volume to the other; is there a filesystem, or any other approach, which will do this? The distributed file systems, e.g. GlusterFS, DRBD etc seem to focus on replicating disks between servers, can they be set up to do what I'm interested in here? I also haven't seen anything about other's taking this approach. Do I have a solution in need of a problem here (i.e. is performance good enough, so this whole idea is redundant)? Is there some flaw in the plan?

    Read the article

  • ec2 spot instance for daily processing task

    - by chaft
    I don't have much experience as a sysadmin or with amazon aws, so I hope someone can explain in simple terms or refer me to a good guide on how to achieve the below. I have a system running on ec2 and amazon rds getting data in and saving it to the db. I need to run a script once a day (at the end of the day) to process all that data and prepare a daily report. This process will take approximately an hour to run. It needs to run on a high memory instance.. From what i've read so far, I guess the best way to do it is to have a high memory spot instance run every day, set it up to execute the script on startup and and shut down when done. Is that the right way to do it? If so, how to do it? how to tell the spot instance to run every day? through a cron job on the other server or is there a better way? How to set it up to run the script on startup? through cloudinit? Any help would be appreciated. One last thing, the job is not very time sensitive as long as it runs every day.. thanks

    Read the article

  • How to interpret number of URL errors in Google webmaster tools

    - by user359650
    Recently Google has made some changes to Webmaster tools which are explained below: http://googlewebmastercentral.blogspot.com/2012/03/crawl-errors-next-generation.html One thing I could not find out is how to interpret the number of errors over time. At the end of February we've recently migrated our website and didn't implement redirect rules for some pages (quite a few actually). Here is what we're getting from the Crawl errors: What I don't know is if the number of errors is cumulative over time or not (i.e. if Google bots crawl your website on 2 different days and find 1 separate issue on each day, whether they will report 1 error for each day, or 1 for the 1st, and 2 for the 2nd). Based on the Crawl stats we can see that the number of requests made by Google bots doesn't increase: Therefore I believe the number of errors reported is cumulative and that an error detected on 1 day is taken into account and reported on the subsequent days until the underlying problem is fixed and the page it's crawled again (or if you manually Mark as fixed the error) because if you don't make more requests to a website, there is no way you can check new pages and old pages at the same time. Q: Am I interpreting the number of errors correctly?

    Read the article

  • Google Webmaster tools Incorrect rel-alternate-hreflang implementation warning message

    - by Noam
    I'm getting this warning msg. in Google webmaster tools Incorrect rel-alternate-hreflang implementation In particular, there seems to be a problem with missing or incorrect bi-directional linking (when page A links with hreflang to page B, there must be a link back from B to A as well). This msg. seems pretty straight forward, but when checking their example pages, I'm not finding anything wrong. I'm using alternate for translation of main site menu, titles, etc.. In each page I have this: <link rel="alternate" hreflang="en" href="http://mydomain.com/page" /> <link rel="alternate" hreflang="jp" href="http://ja.mydomain.com/page" /> <link rel="alternate" hreflang="ko" href="http://ko.mydomain.com/page" /> <link rel="alternate" hreflang="th" href="http://th.mydomain.com/page" /> <link rel="alternate" hreflang="es" href="http://es.mydomain.com/page" /> <link rel="alternate" hreflang="pt" href="http://pt.mydomain.com/page" /> I've double checked this exists in all the 6 pages. This is the first time I've seen this msg although I've implemented this at least 6 months ago, and implementation hasn't changed. Is there any way to check a specific set of pages for these things? Am I missing something in my implementation? We're auto-redirecting people from a location to their specific language, and give them an option to manually change this. I've also just found out about the suggestion for Vary HTTP header - is that relevant and important here?

    Read the article

  • Fix 403 errors in Google Webmaster Tools

    - by Justin
    Hi Team, I have a domain that has "fallen off a cliff" for searches in Google. Searches that used to be in position 1-4 are now gone from page 1. The same search in Bing shows the typical position expected (top 5 results). In reviewing Google Webmaster Tools, I am seeing two problems: 1. The Sitemap is reporting two errors: General HTTP error: HTTP 403 error (Forbidden) URLs not accessible However, the URL they provide as "no accessible" is accessible. I can click the link Google provides and it works fine. There are 6,000 crawl errors of type 403. Again, most of these pages that have 403 are accessible in my browser (tried various browsers as well). About half are from January, the other half from November. There are no IP-specific firewall rules on ports 80 and 443 that could block the goolgebot Using the user agent switcher add-on for FF I confirmed that the page loads when the user agent is the googlebot I an confirm that most of the pages reported as 403 are accessible. A search of just "site:thedomain.com" does confirm there are over 9,000 in the index. But most searches don't return the site. I believe the 403 issues are the cause of the fall in search rankings, but I can't seem to find any information online with ideas about how to address this. Any ideas? jpe

    Read the article

  • What tools should every programmer know?

    - by acidzombie24
    What are some tools every programmer know about? Some examples i thought were Source control. (No explanation needed) Profiler. Many could go without but its good to know how to use one when the occasion arise What else? I was thinking a bug report software but i havent used one so i wasnt sure. Should programmers know how to use TRAC? I remember in the past a person telling me if i was making a shared library i should know (Some Name) which generates docs from the source (in that case C++). What was that called and what else should i know about? -edit- what about team management software? any software you could not live without in a specific project would be a good mention. I'll also mention i use VMs frequently during the prototype or end phase to see if there are any issues on a clean XP or linux distro and if i forgot anything in my redistribution. I cant imagine the end/release and testing phase of a project without a VM.

    Read the article

  • Tools for game script / storyboard

    - by Pietro Polsinelli
    I am searching for a tool that will help in writing a game script. By "script" I mean the text core of a storyboard - without the drawing drafts, which may or may not be there (yet). What I'm thinking of will let write a piece of text of the script, define a simplified workflow from that step, and then define the text of next steps, and so on. Searching online, I found Inform http://inform7.com/ ("A Design System for Interactive Fiction Based on Natural Language") which in theory is exactly what I am searching for, but trying to use it it has this model of a space (a dungeon, a library) where you are picking up objects and exploring them. In my case I am designing more a Sims like game, the flow is entirely different. Considering non specific software, mind mapping tools miss the linearity of the process. What I am writing is a directed graph - simply a work-flow, but the way I want to design it is more text based than work-flow based. SO what I'm doing now is using a text editor, which I'll transform directly in code. Any suggestions?

    Read the article

  • Working with Google Webmaster Tools

    - by com
    My first question is about Crawl errors in Google Webmaster Tools. Crawl errors is devided into few sections. One of them is HTTP. I assume that all broken links in HTTP was somehow found by crawler, this is not the links from sitemap. If this was found by scanning all sitemap pages for links, why it doesn't mention what was the source page, like in sitemap section with column Linked From. And what the meaning of Linked From, I thought if the name of section is sitemap, therefore all URLs should be taken from sitemap, so why there is Linked From? The second question, what is the best way to trreat searching on the site. How come the searching result page are getting indexed? Because of the fact that all searching result page are getting indexed, I have to many page in Linked From. What's the right practice? Question three: In order to improve response time in WMT, can I redirect all crawler's requests to designated free web server? Is this good practice? Question four: How should I treat Google Analytics Code (with parameters PageView, PageLoadTime), in the case user request non existing page, should I render Google code or not? Right now I use Google Analytics Code on the common template page, such that every page, also non existing page with error message contains Google Analytics Code, it seems like it has influence on WMT.

    Read the article

  • How do I mount an EBS root volume to a windows instance in Amazon EC2

    - by Kyle
    So basically, I created a large windows server for development, and then I created a micro windows server for production. I set up everything how I wanted it on my development server, and then i unmounted the drives, and mounted them to my micro server. Now I'm trying to get back into my large windows development server, and I'm getting the error. Invalid value 'i-4896ce28' for instanceId. Instance does not have a volume attached at root (/dev/sda1) this error pops up when I try to start my large windows server. I've remounted the drives to the large development server, and I still get this message. I'm not really sure what to do, I've read other posts and everyone is giving these almost like command line arguments and talking about other tools, and I really have no clue what any of that means, or where I even have an option to enter any commands without be logged into a specific instance.

    Read the article

  • Simple Backup Strategy for Amazon EC2 instances / volumes?

    - by minerj
    You have entered Introductory Backups for Amazon EC2 EBS-backed Windows Images 010... I have been browsing my brains out to find a simple backup strategy for our single windows 2008 server running SharePoint Services. This is an EBS-backed image of one server with one data volume. I don’t need anything exotic. I only need a “daily” backup (losing a day’s worth of data is not catastrophic). We have created and saved an EBS backed AMI image (Windows 2008) we are comfortable using. We started off making backups by simply creating a new EBS AMI image. This is really simple, but the running server is put offline during the first 10 – 15 minutes of creating the image – not ideal. The standard way of creating backups would seem to be creating snapshots of volumes attached to a running instance. Again it’s pretty simple and the server remains usable during the snapshot generation. The apparent Catch-22 is that you can’t simply launch a new instance directly from a snapshot. I know how to bundle a running instance to S3 storage and then register the AMI from the S3 bucket. This does allow me to capture a backup of a running instance and, if the running instance is lost, register the AMI from the S3 bucket and launch the new AMI to recover the instance, but this seems really convoluted and it seems ridiculous to have to juggle back and forth between the AWS Console and the S3 Organizer plug-in for Firefox to get this accomplished. (Please don't mention the command line approach, this is an 010 level course). From playing around with EBS-backed images, the following approach appears to work for me (all done within the AWS Console): 1.For your backups, simply snapshot the system volume (/dev/sda1) as needed. 2.If you lose your running instance, do the following: a.Create a new volume from your last snapshot backup b.Launch another instance of your starting AMI (must be EBS-backed) c.Stop this instance. d.Detach the existing system volume from the new stopped instance and discard. e.Attach the newly created volume as system volume (/dev/sda1) to the stopped instance. f.Re-start the new instance. I have tested this out a couple of times and it seems to work for me. Question: Is there anything wrong with this approach?

    Read the article

  • Connectivity issues with dual NIC machine in EC2

    - by Matt Sieker
    I'm trying to get some servers set up in EC2 in a Virtual Private Cloud. To do this, I have two subnets: 10.0.42.0/24 - Public subnet 10.0.83.0/24 - Private subnet To bridge these two, I have a Funtoo instance with a pair of NICs: eth0 10.0.42.10 eth1 10.0.83.10 Which has the following routing table: Kernel IP routing table Destination Gateway Genmask Flags Metric Ref Use Iface 10.0.83.0 * 255.255.255.0 U 0 0 0 eth1 10.0.83.0 * 255.255.255.0 U 203 0 0 eth1 10.0.42.0 * 255.255.255.0 U 202 0 0 eth0 loopback * 255.0.0.0 U 0 0 0 lo default 10.0.42.1 0.0.0.0 UG 0 0 0 eth0 default 10.0.42.1 0.0.0.0 UG 202 0 0 eth0 An elastic IP is attached to the eth0 interface, and I can connect to it fine remotely. However, I cannot ping anything in the 10.0.83.0 subnet. For now iptables is not set up on the box, so there's no rules that would get in the way (Eventually this will be managed by Shorewall, but I should get basic connectivity done first) Subnet details from the VPC interface: CIDR: 10.0.83.0/24 Destination Target 10.0.0.0/16 local 0.0.0.0/0 [ID of eth1 on NAT box] Network ACL: Default Inbound: Rule # Port (Service) Protocol Source Allow/Deny 100 ALL ALL 0.0.0.0/0 ALLOW * ALL ALL 0.0.0.0/0 DENY Outbound: Rule # Port (Service) Protocol Destination Allow/Deny 100 ALL ALL 0.0.0.0/0 ALLOW * ALL ALL 0.0.0.0/0 DENY   CIDR: 10.0.83.0/24 VPC: Destination Target 10.0.0.0/16 local 0.0.0.0/0 [Internet Gateway ID] Network ACL: Default (replace) Inbound: Rule # Port (Service) Protocol Source Allow/Deny 100 ALL ALL 0.0.0.0/0 ALLOW * ALL ALL 0.0.0.0/0 DENY Outbound: Rule # Port (Service) Protocol Destination Allow/Deny 100 ALL ALL 0.0.0.0/0 ALLOW * ALL ALL 0.0.0.0/0 DENY I've been trying to work this out most of the evening, but I'm just stuck. I'm either missing something obvious, or am doing something very wrong. I would think I'd be able to ping from either interface on this box without issue. Hopefully some more pairs of eyes on this configuration will help. EDIT: I am an idiot. After I bothered to install nmap to run some more tests, I discover I can see the ports, and connect to them, pings are just being blocked.

    Read the article

  • Google Analytics API - Super simple?

    - by Jens Törnell
    Google Analytics API - Too complicated? I've read about Google Analytics API but heard of others that it is a bit complicated to make it work. I use PHP. Copy / paste example My question is if there is a copy / paste example anywhere on the web for getting a stats curve of the latest month, or just the numbers for that period? Important I need to use the new Google Analytics API version for 2012. The other one is going to die soon.

    Read the article

< Previous Page | 15 16 17 18 19 20 21 22 23 24 25 26  | Next Page >