# Advice needed: migrating off the s3 gateway **URL:** \ **Category:** Elasticsearch\ **Created:** [October 6, 2013, 11:33am UTC](https://discuss.elastic.co/t/advice-needed-migrating-off-the-s3-gateway/13859 "2013-10-06T11:33:23Z")\ **Posts on this page:** 3\ **Page:** 1
**Author:** ![Paul\_Sanwald](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/paul_sanwald/32/2076_2.png) [@Paul\_Sanwald](https://discuss.elastic.co/u/Paul_Sanwald)\ **Post date:** [October 6, 2013, 11:33am UTC](https://discuss.elastic.co/t/advice-needed-migrating-off-the-s3-gateway/13859/1 "2013-10-06T11:33:23Z")
Hi Everyone, I know the S3 gateway is deprecated, and I'm looking to move away from using it. but, because of the way I deploy and run elasticsearch, I'm not sure the best way to do this; and was hoping to get some advice from the group. I'll try to give a detailed explanation of my deployment, bear with me. I have a web service that uses elasticsearch as its datastore. This web service runs using amazon's elasticbeanstalk infrastructure; this allows me to autoscale and do some other nice things around managing deployments, but it does mean that instances get swapped around as AWS needs to, much like heroku, I can't rely on my instances having static IPs, and I can't rely on persistent local storage. When I set up my service (which is a java web service running under tomcat), instead of running elasticsearch as a service, I used a plugin to start up the node directly in my webservice. This has worked very well for us, as I just use the S3 gateway, and when I restart my service, the data is just read off of S3. The documentation says I should move to the local gateway, but, it doesn't seem like this will work with the way I have my deployment architected: I don't have persistent local storage on elasticbeanstalk, unless I do something fancy using an AMI, which is a possibility. Do I just need to move off of elastic beanstalk entirely, and use standard EC2 instances with manual deployment, if I want to use the local storage gateway? Is there another gateway that will give me shared, non-local storage? thanks very much for reading, and for any advice you might have. best, --paul -- You received this message because you are subscribed to the Google Groups "elasticsearch" group. To unsubscribe from this group and stop receiving emails from it, send an email to [elasticsearch+unsubscribe@googlegroups.com](mailto:elasticsearch+unsubscribe@googlegroups.com). For more options, visit [https://groups.google.com/groups/opt\_out](https://groups.google.com/groups/opt_out). ---
**Author:** ![Oliver\_K](https://avatars.discourse-cdn.com/v4/letter/o/e95f7d/32.png) [@Oliver\_K](https://discuss.elastic.co/u/Oliver_K)\ **Post date:** [December 2, 2013, 7:45pm UTC](https://discuss.elastic.co/t/advice-needed-migrating-off-the-s3-gateway/13859/2 "2013-12-02T19:45:45Z")
Any news on this? Have you got any good solution? I am planning the production release of an webapp based on php, mysql (AWS RDS) and elasticsearch. I would really prefer to use the same beanstalk cluster for the webapp and elasticsearch. Running another additional cluster for the elasticsearch nodes sounds like more administrative work afterwards and also mean more costs. What is the best practice structure for using Beanstalk on Amazon EC2 and also having a good backup solution if an instance goes down... Am Sonntag, 6. Oktober 2013 13:33:23 UTC+2 schrieb Paul Sanwald: > Hi Everyone, > I know the S3 gateway is deprecated, and I'm looking to move away from > using it. but, because of the way I deploy and run elasticsearch, I'm not > sure the best way to do this; and was hoping to get some advice from the > group. I'll try to give a detailed explanation of my deployment, bear with > me. > > I have a web service that uses elasticsearch as its datastore. This web > service runs using amazon's elasticbeanstalk infrastructure; this allows me > to autoscale and do some other nice things around managing deployments, but > it does mean that instances get swapped around as AWS needs to, much like > heroku, I can't rely on my instances having static IPs, and I can't rely on > persistent local storage. When I set up my service (which is a java web > service running under tomcat), instead of running elasticsearch as a > service, I used a plugin to start up the node directly in my webservice. > This has worked very well for us, as I just use the S3 gateway, and when I > restart my service, the data is just read off of S3. > > The documentation says I should move to the local gateway, but, it doesn't > seem like this will work with the way I have my deployment architected: I > don't have persistent local storage on elasticbeanstalk, unless I do > something fancy using an AMI, which is a possibility. > > Do I just need to move off of elastic beanstalk entirely, and use standard > EC2 instances with manual deployment, if I want to use the local storage > gateway? Is there another gateway that will give me shared, non-local > storage? > > thanks very much for reading, and for any advice you might have. > > best, > > --paul -- You received this message because you are subscribed to the Google Groups "elasticsearch" group. To unsubscribe from this group and stop receiving emails from it, send an email to [elasticsearch+unsubscribe@googlegroups.com](mailto:elasticsearch+unsubscribe@googlegroups.com). To view this discussion on the web visit [https://groups.google.com/d/msgid/elasticsearch/54a46941-a047-45bd-bd1b-846f9c8dd81c%40googlegroups.com](https://groups.google.com/d/msgid/elasticsearch/54a46941-a047-45bd-bd1b-846f9c8dd81c%40googlegroups.com). For more options, visit [https://groups.google.com/groups/opt\_out](https://groups.google.com/groups/opt_out). ---
**Author:** ![system](https://us1.discourse-cdn.com/elastic/original/3X/1/a/1ac57faf039f6b580b3f104ef42a2a89e41014de.png) [@system](https://discuss.elastic.co/u/system)\ **Post date:** [July 6, 2017, 2:03am UTC](https://discuss.elastic.co/t/advice-needed-migrating-off-the-s3-gateway/13859/3 "2017-07-06T02:03:48Z")