[Official note] Why is ES4 divided into two stages?

From the first version of ES4 released to now, there has been a period of time, the vast majority of friends have been adapted to the ES4 brought about by the new package mode, but there are still some new and altar friends on the ES4 two-stage mode with puzzled or distorted. This post will explain the two-stage significance of ES4, hope that more friends can appreciate the design purpose of the ES4 two stage model.

1. What is encapsulation

Before we discuss the issue, let's start with a brief talk about what is encapsulation.

In fact, the word "encapsulation" is too abstract, and the original word "Sysprep" is a very figurative word, originally intended to "Prepare the System (SYS)" (Prep), in other words, the system is set to deploy (deploy) state.

The encapsulation itself is a very simple thing:
(1) for winxp/2003 and other NT5 systems, it is necessary to extract the Deploy toolkit from the System Source disk tool, extract Sysprep.exe and related program files, create C:\Sysprep directory, and copy Sysprep.exe and related files into it. Execute the C:\Sysprep\sysprep.exe, follow the instructions in the UI to complete the encapsulation;
(2) for win7/8/2008/2012 and other NT6 systems, packaging is more simple, directly run C:\Windows\system32\sysprep\sysprep.exe, follow the UI instructions to explain the operation, you can complete the package.

so the "encapsulation" is not complicated, and the real trouble and complexity are the various demands we put forward. For example, we need to enhance the scope of the system deployment, so that the integration of various disk controller drivers, we need more interfaces to invoke their own designated programs (Universal drive, runtime, package, optimization, etc.); we want to compress the C-drive software to save volume and release at deployment time; we want to make the deployment background look better, To show more explicit progress in the deployment, and so on. It is our demand that makes encapsulation quite complex, thus reducing the package success rate and stability to a certain extent.

In order to solve this problem, the two-step mode of ES4 was born through repeated thinking and practice.

2. Two stages of their respective purposes

The first stage of ES4 is to make the most basic package for the system, and to encapsulate the stable complete system as the primary purpose.
<ignore_js_op>
In the first phase, ES4 focuses on completing the encapsulation task without performing any operations that are not directly related to encapsulation. Thus the maximum possible maintenance of the System package stability and success rate.
After the first phase is completed normally, a basic system with deployable capabilities is completed.
The system can be backed up in PE immediately after a reboot, resulting in a system image with a basic deployable capability. This system can be used for deployment, but the scope and scalability of the deployment are limited.

The second stage of ES4 is to make various adjustments to the system, extending the scope of the system support and adding various functions as the primary purpose.
<ignore_js_op>
In the second phase, ES4 focuses on adjusting the system to what the wrapper needs to do, only performing the adjustment. This allows the system to diversify to the maximum extent possible without compromising encapsulation.
After the second phase is completed normally, a personalized system with wide deployment scope and versatility is completed.
This system can be backed up immediately under PE, resulting in a system image with strong deployment capabilities!

Thus formed, encapsulation is the package, adjustment is the adjustment of the Division of the situation. This will not affect the System Package success rate and deployment stability because of some system adjustment.

3. Additional benefits in two-stage mode

The two-stage model also brings 3 additional benefits:

(1) When a problem is found, you can more clearly determine whether the problem is "system encapsulation" or "system tuning"
If the image after the first phase is not deployed at all, the problem is in the first phase, the "System encapsulation";
If the image after the first phase is ready to be deployed and the image after the second phase is not deployed properly, the problem is in the second phase, "system tuning."

(2) based on the first stage image, you can try the second stage multiple times without re-encapsulation
The second stage has many options, and when some options are set up with a problem, just restore the first stage image and execute the second phase again based on the configuration file just now.
In this case, no re-encapsulation is required, and the package success rate is not affected.
But it's important to note that I'm talking about "second phase after recovering the first phase of the image" rather than performing a second phase of re-execution, re-execution, multiple executions, and the consequences are unknowable.

(3) Under the PE can be very good to circumvent system permissions
Windows has more restrictive permissions, and many operations cannot be done under the current system for security reasons, but encapsulation needs to change some system key values frequently, which is restricted by permissions and cannot be done well.
And the PE and the system you encapsulate can be said to be two completely different systems. Modifying another system under one system can be said to be unimpeded by 97% of the conditions.
In this way, we can better adjust the system, regardless of the target system permissions issues, and do not have to close some security restrictions under the current system to take unnecessary risks.

(1) (2) are based on the "completion of the first phase immediately after the re-boot into PE, making the first stage image", remember!

4. Other issues

(1) When the first phase is completed, what time does it enter PE?
After the first phase is complete, the system is already a system that has entered the "ready to deploy" state. At this point restart please immediately enter the PE, and back up the system, ready to start the second phase.
And in the first phase after the restart, did not enter the PE, and from the hard disk to boot the system, then the system has begun to deploy, at this time All is late, has deployed the system, how to perform the second phase?
A lot of new people in this circle, the first phase of execution, restart the deployment after the implementation, only remembered to enter the PE, after the ES4 found in the "Ready state" system.

(2) How to enter PE under virtual machine?
Can ask this question, at least will prove that you are not a qualified technician, you have a long way to go on the road of technology.

In the case of VMware virtual machines, when the virtual machine is turned off, locate the device.
Select the first optical drive (select one if your VM only has one optical drive set)
<ignore_js_op>
Pop-up page, set virtual machine light to drive with ISO image, click "Browse" button
<ignore_js_op>
Select the image file for the "USB flash drive installation assistant ISO" (again, please use the ISO version instead of the USB flash drive!) )
<ignore_js_op>
Enter the virtual machine BIOS to set the optical drive to the first boot
<ignore_js_op>
Starting with this, you can see the start page of the PE.

(3) Two stage package ≠ two times package
A two-stage package is a package divided into two stages, respectively, to carry out encapsulation and adjustment, thereby increasing the success rate, enhance stability.
Two-times encapsulation refers to deploying the encapsulated system and then encapsulating it again.
This is two completely different concepts! Do not confuse!

Well, that's it, after reading this article carefully, I think the altar friends of the two stages of ES4 have more understanding.

Two-phase encapsulation is an innovation in system encapsulation. System encapsulation from the beginning of the study, in addition to ES4 auxiliary tools are using a one-stage package, the system adjustment and encapsulation in the same time period.

Of course, I do not deny that the ES4 is more problematic than ever since the first attempt at phase two. In one system (PE) modify the contents of another system (the encapsulated system), in the operation of the system, can say that the previous code is useless, many of the previous functions need to be significantly adjusted or completely changed to perform normal. This makes many of the things that seem to be the same as before, the actual way of doing things.

But even so, ES4 try new ways, new ideas will not stop the pace. The two-stage benefits are obvious, now encountered the problem is because the two stages is not as mature as a stage, after all, a phase has been used for more than 10 years, and two stages since the ES4 has been more than 1 years time! Continue to try, continue to innovate, continue to improve, just ES4 to do things, a good idea, not because of a moment without perfect and give up!


Dare to innovate, only the future!

[Official note] Why is ES4 divided into two stages?