Sissle Quick User Guide

Quick Guide

Installation

Sissle requires no special installation. Unzip the Sissle.zip to any directory c:\program files\sissle for instance, right click on sissle.exe and produce a link. Copy the link to your desk top.

If you do not have Net 2.0 then download the set up file dotnetfix.exe

or use the official Microsoft site if you prefer or require a different language, other than English
Microsoft Net 2.0

Guide

The Amazon S3 system allows you to create 100 buckets and fill each bucket with as many keys as you like. The bucket is the highest level of creation, consider a bucket the same as a very very large disk. Keys are the equivalent of files and are unlimited in number but each key / file must be less than 5 GB. Bucket names must be unique within the S3 system. That is the bucket name must be unique across all users of S3, so for example trying to name a bucket, test, john, will fail but testy564, john.charles.smith could possibly succeed. There are no directories / folders on the S3 system, these are simulated for you by Sissle.

Once a bucket has been created Sissle will make the S3 system appears as an ordinary disk to you.

The use should be self evident but there are a few points to watch.

Bucket Creation

Create a new bucket by entering a name in the remote MkDir at the root level. A pop up menu gives you a choice of plain or back up mode. Plain means it will attempt to create a bucket with the name you have supplied. Back up mode will append some encrypted characters (MD5Hash of your access key) to the name you supplied to create a unique bucket name. These encrypted characters will not be seen while you operate the program.

Example
In plain mode if you try and create a bucket called "test" it will not work, as some other user already has a bucket called test.

In back up mode if you create a bucket called "test" it will be created as test but is actually prepended with a set of characters to make it fairly unique.

Note :

If you require the bucket to contain a web site or be open to the public then you want plain mode.
If you wish the bucket for your own use then the mode is irrelevant but I suggest back up mode.

If you make a mistake at this stage, you can not correct it other than deleting the bucket and its contents and starting again.

Transferring Files to and from S3

This a simple several stage operation.

From the main window select files and folders to transfer.
Initiate transfer, to or from S3, using the direction buttons.
This will open the transfer status window where finer editing of which files are required can be done. When complete the transfer can be started.
The transfer now takes place. Typically 10-50 files are transferred simultaneously, the overall progress is shown in a progress bar.When all transfers have been completed the transfer status window is closed and control is passed to the main window.

Transferring to and from S3 is simple:
Except segmented downloads.
Except setting the ACLs

Segmented DownLoads

The S3 service is very reliable and segmented downloads is a mechanism to add more reliability to downloads. If you are downloading very large files (e.g. 1 GB ) then segmented downloads provides:

- any failure will only require a repeat of the failed segment (default 10MB)
- there is a recovery process that can be applied to this file at any time (eg tomorrow)

As this is a specialized subject and requires a lot of description this is provided with its own page Segmented Downloads

ACLs

Each bucket and its content have individual access rights. These are provided by Access Control Lists, ACLs. The default is to set everything private to the owner, you. If this is what you want, e.g. for backups, then you can ignore ACLs altogether.

This another specialized area and has been given its own page, ACLs

Synchronization

There are 3 types of Synchronization, choose wisely, as deletions can occur.

Abbreviations:
LERDN - Local exists remote does not
RELDN - Remote exists local does not
LNR - Local newer than remote
RNL - Remote newer than local

Local is synchronization master.
LERDN - Copy local to remote
RELDN - Delete remote
LNR - Copy local to remote
RNL - Do Nothing
Remote is synchronization master
LERDN - Delete local
RELDN - Copy remote to local
LNR - Do Nothing
RNL - Copy remote to local
True Synchronization
LERDN - Copy local to remote
RELDN - Copy remote to local
LNR - Copy local to remote
RNL - Copy remote to local

For the first 2 choices above there exists 2 options

Local is synchronization master.
Inhibit remote deletions
Remote is synchronization master
Inhibit local deletions

Amazon S3 Failures

Any failures with S3 are repeated a number of times as set in the Option dialogue and also applies the time delay specified in the same Option dialogue. After these repeats have occurred a message box presents the error details and asks if you wish to retry or ignore this operation.

Under Options menu there is a Reset menu item. The reset is an admittance that problems have occurred and will most probably reoccur that require its use. This attempts to abort the S3 operations thread and return the GUI to a sensible state.

Advice

Transferring a large file

Set number of repeats low or even zero.
Set the ReadWriteTimeout large 20 min = 20 * 60 *1000 = 1200000

Transferring lots of small / medium files

Set number of repeats high 3/4/5.
Set the ReadWriteTimeout small 1 min = 60000

Modes

Create buckets in the correct mode Plain / Backup.
Not correctable latter.

ACLs

ACLs can be corrected / altered at any time, but it will save time if set before transfer.