i2s3

watch for new files and upload them to s3

this utility makes it very easy to get files into a s3 bucket:

mkdir /tmp/i2s3q
cp /etc/hosts /tmp/i2s3q/hosts
# or
echo 1234 > /tmp/i2s3q/1234.txt

that's it - object is now stored on s3!

if pulic/http permissions are set on the bucket, you can browse:

files are stored based on your local directory structure:

mkdir -p /tmp/i2s3q/foo/bar/baz
echo test > /tmp/i2s3q/foo/bar/baz/test.txt

https://s3.amazonaws.com/example_bucket/foo/bar/baz/test.txt

Usage:

-c, --config            config file [/etc/i2s3.cfg]
-h, --help              don't print this message

all below parameters (long versions) can be specified in a config file
with format 'key => value'.  command line arguments override config file.

-D, --debug             print debug messages
-Q, --quiet             hush console output

-d, --delete            delete local files after uloading to s3
-f, --foreground        stay foreground; dont fork
-p, --piddir            create pid file here
-q, --queue             queue directory to monitor
-r, --reprocesss        reprocess queue this many seconds (for failed uploads, et. al.) [300]
-s, --syslog_facility   send syslog messages to this facility (none if not specified)
-u, --user              user to run as

--s3_host               s3 server hostname      [s3.amazonaws.com]
--access_key_id         use this aws access key id
--secret_access_key     use this aws secret
--bucket                upload objects to this s3 bucket

NOTE: /not/ using --delete incurrs cost in cpu (and some ram). This is because each local file must be tracked to make sure we don't keep uploading non-changed files amazon during our queue check/reprocess. resources are fine when i tested at 10,000 files.

jkister/i2s3

i2s3

watch for new files and upload them to s3

Usage: