Coins will not disappear. However the exchange may decide to use a chain which you do not want to use, and thus prevent you from using coins that you do want to use.

That is incorrect. There can still be one or more splits. However these splits may not be persistent. Those miners who are not following the BIP 91 rules will find their block orphaned. Since there may be 20% or more (because of variance, BIP 91 could activate without 80%) of miners who are not following the BIP 91 rules, they could end up mining a small fork of the chain briefly before the rest of the miners mine a chain longer than theirs and causes a reorg. Furthermore, with the prevalence of spy mining, miners who are enforcing BIP 91 may accidentally end up extending a non-BIP 91 chain because they are mining before validating.

The same issue would happen if BIP 148 had majority hash rate support, but because it does not, BIP 148 would likely result in a persistent two chain split.

You are asking us to predict the future; we can't do that. BIP 91 currently has almost 80% hash rate support. With some luck and variance, it could activate soon. Whether it actually does waits to be seen. We cannot predict the future.

Furthermore, even if BIP 91 does activate, miners may not be actually enforcing it so it may not even activate segwit.

If BIP 91 activates, then it will require all blocks to signal for segwit. This is the same as what BIP 148 requires after August 1st, so, assuming that there are no chain splits because of BIP 91, BIP 148 nodes will follow the same chain, and after August 1st, both will be enforcing the same rules.

Also my website is tracking BIP 91 signaling: https://www.btcforkmonitor.info/ as well as monitoring for any chain splits.

This thread is talking about neither of those things. BIP 91 is unrelated to UASF or UAHF. Neither the UASF or the UAHF will trigger the warning that you see here as they do not do miner signaling; rather they activate via flag day. However BIP 91 activates via miner signaling, so we get the warning here about that.

Multiple chain splits can occur when the 80% of miners orphan the blocks of the 20% who are not signaling for segwit. While this may not result in a persistent chain fork, there can and likely will be mini forks and longer orphan chains. This problem is also exacerbated by spy mining (where miners begin building on a new block before validating it) and by the short confirmation window of BIP 91. The short window means that BIP 91 could activate with less than 80% hash rate support.

Your questions is incredibly vague. Can you please be more specific? Can you tell us what wallet software or service you are using?

You are completely missing the nuance. Very few of the Core developers have said that they support BIPs 148/149 with the current political situation. Re-read what those statements say: "Prefer" it is what he would choose if it was only up to him and no outside influences. That means that if BIP 148/149 existed in a vacuum with no outside influence (e.g. from miners) and everyone went along with it, they would prefer that as it is an activation via flag day. Are you seriously that confused about the meaning of written English? Are you seriously misconstruing "prefer" and "acceptable" as meaning "want it now as it is now" even when the definition on the page says otherwise?

Since BIP 148/149 do not exist in a vacuum and they are not the proposals accepted by everyone, nearly all Core developers do not endorse it. There have been three attempts to merge BIP 148 into Core, and all of them have been rejected by the very same people you claim are endorsing BIP 148. Those Core developers are not saying that they are happy with BIP 148/149 as they are now, only that if those were the only two proposals and existed in a vacuum and without any contentiousness that they would then support them.

It's a general information thread about "warnings that you will see" so I put it in Bitcoin Discussion for more visibility.

Read https://bitcointalk.org/index.php?topic=2027513.0

Check out the website here: https://www.btcforkmonitor.info/

This is a website that I put together over the weekend in preparation of potential BIP 91 activation and BIP 148 activation. If the site detects a chain fork, there will be a big red alert at the top saying as such. Each node will also include the block hash and height at which the chain that it is following diverged from another node's chain. In the event of multiple chain forks, the most recent one will be shown for each node.

The website will consider any chain split of 2 or more blocks deep as a chain split. This is to avoid cases of orphaned blocks and of latency between the nodes.

The website will update its data from the nodes every 10 seconds. However you will need to refresh the page to see if there has been any changes.

---

The source code for this website is available at https://github.com/achow101/forkmon. Pull requests are welcome.

The website and its source code is licensed under the MIT License.

read https://bitcointalk.org/index.php?topic=2027513.0

yes.

Where am I complaining about the 80% activation threshold? The purpose of this thread is to inform people about the warning that they will see in Core, not complaining about BIP 91.

The table does not explain any nuance and conditions under which people find BIPs 148 or 149 as preferred.

If the majority of Core actually supported BIPs 148 and 149, then BIP 148 support would have been merged into Core, but it has not.

btc1 is the client, not BIP 91. BIP 91 is the proposal. Anyways, by Core, I mean any derivative of Core.

August first has nothing to do with Segwit2x. Segwit2x can cause a chain split before August 1st. The effects of a segwit2x chain split and a BIP 148 chain split on August 1st will be fairly similar, so read this thread: https://bitcointalk.org/index.php?topic=2012799.0

You can't. The best you can do is to make a crawler to try to figure out what nodes are accepting incoming connections. You can't know about the nodes that only have outbound connections though.

Segwit2x is using BIP 91. BIP 91 specifies that 269 of the last 336 blocks (~80% of the last 336 blocks) must signal for BIP 91 on bit 4. Once there is enough signalling, it will be locked in for a short grace period of 336 blocks. Then it will activate. Once BIP 91 activates, all blocks must signal for segwit on bit 1. This means that segwit will then activate via its original deployment on bit 1. Segwit2x is not actually activating a different segwit, they are activating something else which will force the activation of segwit.

There may be forks since 20% of the hash rate won't be support BIP 91 and thus will likely not be signalling bit 1 for segwit. With some luck and spy mining, forks could be made from miners extending a chain off of a block which does not signal bit 1. However Bitcoin Core will allow you to choose the chain to use (via the invalidateblock RPC command) and any of the forks will be valid to Core.

If BIP 91 activates successfully and everyone who claims they will enforce it actually enforces it, then there shouldn't be a fork.

Yes. Coins at rest are not at risk.

Any recent version of Core should be fine, even ones that don't have segwit support.

Yes. There currently is no fork, so you can safely send your coins.

Add reindex=1 to your bitcoin.conf file. Then start Core. Once it has started, remove that line from the bitcoin.conf file. What this will do is reindex the database, which means that it is reading the blockchain data from disk and rebuilding its databases. It will appear as if it is redownloading the entire blockchain, but it is not. Make sure that your computer does not forcibly shut down during this time as an unclean shut down will corrupt the database and require it to be reindexed again.

Yes, it will work with Bitcoin Core 0.13.1+

No. There is no need to touch your wallet files or the blockchain data. All you need to do is download and install Bitcoin Core 0.14.2. The install will overwrite the previous install but it will not touch anything in the datadir (folder containing the wallet and blockchain stuff).

269 of the last 336 blocks (~80% of the last 336 blocks) must set bit 4 of the version number field in order for BIP 91 (first half of segwit2x) to activate.

No, that is not a guarantee that we will be getting segwit2x.

Segwit2x is not guaranteed to activate on July 21st. It is not activated via flag day, rather it is activated via miner signalling. July 21st is the day that those who signed the New York Agreement for Segwit2x are supposed to begin signalling and running segwit2x nodes. But that is not a guarantee that activation will happen that day nor is it a guarantee that everyone will actually do so.

No, that is completely incorrect. The threshold for BIP 91 (the first half of segwit2x) is 80%, not 50%. Secondly, a 100 block period is much too short to determine anything. It can be highly affected by variance and luck. You need to much longer timespan to determine how much support there actually is. Even the 336 block window for BIP 91 is very short and could be affected by variance so there could actually be less than 80% hash rate support and BIP 91 would still activate.