Update satisfactory image for 1.0 release

[ikiris]

Bump satisfactory docker server image and make port changes nessecary to accommodate 1.0 port changes.

Has been tested personally and works.

[ikiris]

The data is literally what zfs snapshots are for man. I've got the game going to a host path for that reason. It's nice and safe.

On Wed, Sep 18, 2024, 12:52 Niborian @.***> wrote:

@.**** commented on this pull request.

In stable/satisfactory/10.1.1/ix_values.yaml https://github.com/v3DJG6GL/truecharts_archive/pull/12#discussion_r1765476975 :

@@ -0,0 +1,58 @@ +image:

• repository: wolveix/satisfactory-server
• pullPolicy: IfNotPresent
• tag: v1.8

You make a good point, and I agree that rollbacks can be a quick fix. However, rollbacks generally only revert the app—not the content or save files. If an update changes the data format or structure, there’s a chance that the save files or even backups could be affected, especially if the app does automatic migrations or overwrites older backups.

The chance of something breaking is low, but personally, I wouldn’t take that risk. With my luck, I’d end up losing all the hours I’ve put into the game save. So I’d rather wait a bit longer or just use my own catalog while waiting, rather than skipping the version lock and hoping for the best.

But that’s just my opinion—everyone has different risk tolerance!

— Reply to this email directly, view it on GitHub https://github.com/v3DJG6GL/truecharts_archive/pull/12#discussion_r1765476975, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABFAZVVUR2DHO3CKQH35MA3ZXG4WVAVCNFSM6AAAAABOFVP7M6VHI2DSMVQWIX3LMV43YUDVNRWFEZLROVSXG5CSMV3GSZLXHMZDGMJTGM4DEOJSGQ . You are receiving this because you authored the thread.Message ID: @.***>

[Niborian]

Yea true, you’ve got a really good point. I didn’t mean to bother you if it came across like that 😅, but I’m gonna add the hash it is even more safety— even tho the chance for something to go wrong is really low, just my opinion. It’s not really my choice if this is accepted anyways 😅😁, but I might be a bit too cautious—better safe than sorry!😄

[ikiris]

Do you not realize this is just the version of the chart, and it keeps the actual game updated via whatever is current on steam? Like dude you don't seem to understand the actual risk profile here.

On Wed, Sep 18, 2024, 13:31 Niborian @.***> wrote:

Yea true, you’ve got a really good point. I didn’t mean to bother you if it came across like that 😅, but I’m gonna add the hash it is even more safety— even tho the chance for something to go wrong is really low, just my opinion. It’s not really my choice if this is accepted anyways 😅😁, but I might be a bit too cautious—better safe than sorry!😄

— Reply to this email directly, view it on GitHub https://github.com/v3DJG6GL/truecharts_archive/pull/12#issuecomment-2359149958, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABFAZVRS4PQMBYAQ52JXOQ3ZXHBJTAVCNFSM6AAAAABOFVP7M6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDGNJZGE2DSOJVHA . You are receiving this because you authored the thread.Message ID: @.***>

[Niborian]

I get that the risk is really low, like 1 in a billion (just a random number I used, not tied to the actual risk). I’ve actually used the stable tag alone for Factorio just to get updates ASAP, so I totally understand that approach. But after reading more about using hashes, I’ve come to see the pros and cons. The risk isn’t just personal—it’s something everyone shares (unless you’re using your own catalog).

For example, if the risk is 1 in a billion and 2.3 million people have pulled the Docker image, and let’s say all of them are actively using it, the probability of someone running into a problem is around:

• 0.23% for 2.3 million users.
• 0.05% for 500k active users.

Now, if the app is running 24/7, on average, something could theoretically go wrong in about:

• 18.1 years for the 2.3 million total users.
• 83.3 years for the 500k active users.

But, of course, probability means it could happen at any time, earlier or later than expected.

This highlights why risk management is important, especially for larger user bases. When it’s just a small group (say 1-10 people), taking on personal risk might be fine. But when you have larger numbers, those risks start to accumulate. Even a low probability can affect someone when spread across millions of users.

For me, I don’t mind taking the extra step to use the hash and mitigate that small risk. Losing a world wouldn't be fun, at least for me. I didn’t mean to come across the wrong way, but your response felt a bit sharp. I’m just sharing my perspective, and I respect that you have yours.

[ikiris]

This is a pseudoscientific word salad that does not accurately reflect how the math works at all for release management risk at scale. I’m done engaging with you on this.

On Wed, Sep 18, 2024 at 15:55 Niborian @.***> wrote:

I get that the risk is really low, like 1 in a billion (just a random number I used, not tied to the actual risk). I’ve actually used the stable tag alone for Factorio just to get updates ASAP, so I totally understand that approach. But after reading more about using hashes, I’ve come to see the pros and cons. The risk isn’t just personal—it’s something everyone shares (unless you’re using your own catalog).

For example, if the risk is 1 in a billion and 2.3 million people have pulled the Docker image, and let’s say all of them are actively using it, the probability of someone running into a problem is around:

• 0.23% for 2.3 million users.
• 0.05% for 500k active users.

Now, if the app is running 24/7, on average, something could theoretically go wrong in about:

• 18.1 years for the 2.3 million total users.
• 83.3 years for the 500k active users.

But, of course, probability means it could happen at any time, earlier or later than expected.

This highlights why risk management is important, especially for larger user bases. When it’s just a small group (say 1-10 people), taking on personal risk might be fine. But when you have larger numbers, those risks start to accumulate. Even a low probability can affect someone when spread across millions of users.

For me, I don’t mind taking the extra step to use the hash and mitigate that small risk. Losing a world wouldn't be fun, at least for me. I didn’t mean to come across the wrong way, but your response felt a bit sharp. I’m just sharing my perspective, and I respect that you have yours.

— Reply to this email directly, view it on GitHub https://github.com/v3DJG6GL/truecharts_archive/pull/12#issuecomment-2359387677, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABFAZVUPSOCU2KKVUTKA6R3ZXHSDFAVCNFSM6AAAAABOFVP7M6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDGNJZGM4DONRXG4 . You are receiving this because you authored the thread.Message ID: @.***>