I have the same confusion. I don’t think anyone knows what that sentence is supposed to mean.
I have the same confusion. I don’t think anyone knows what that sentence is supposed to mean.
Bumbling Bayesian参与话题 33消息总数 209
该用户的聊天记录定位。
聊天记录
共 209 条，显示第 51-100 条One can still donate fees when the orderbooks are invalid and one can always donate to KT.
One can still donate fees when the orderbooks are invalid and one can always donate to KT.
Yeah, for your use case, if the price drops too low, you would prefer not to get any fills, right?
Yeah, for your use case, if the price drops too low, you would prefer not to get any fills, right?
The strike range definitions are underspecified. There are events where it is impossible from looking at the market strikes alone to figure out what the ranges are. One has to know the granularity of the source.
The strike range definitions are underspecified. There are events where it is impossible from looking at the market strikes alone to figure out what the ranges are. One has to know the granularity of the source.
They forced us to deal with fractional contract counts to make things easier on casual gamblers and still cannot put a little elbow grease into the UI.
They forced us to deal with fractional contract counts to make things easier on casual gamblers and still cannot put a little elbow grease into the UI.
The dumb things I experience with fractional contract counts on the UI have to be visible to some casual users.
The dumb things I experience with fractional contract counts on the UI have to be visible to some casual users.
Yeah, they could easily fix that silliness if they wanted.
Yeah, they could easily fix that silliness if they wanted.
That isn't fair of me. It might not be laziness. Maybe they really have nobody who can figure out how numbers work in JavaScript.
That isn't fair of me. It might not be laziness. Maybe they really have nobody who can figure out how numbers work in JavaScript.
Indeed, lots of UI laziness, like the Kalshi web UI currently.
Indeed, lots of UI laziness, like the Kalshi web UI currently.
Bumbling Bayesian
2026-06-29 02:16:31
↗
Kalshi
Kalshi API订单修改出现order_id_and_client_order_id_mismatch等错误
Kalshi
Kalshi API订单修改出现order_id_and_client_order_id_mismatch等错误
Yeah, and `order_side_mismatch`. Definitely could be a problem on my side, but I have no idea why the behavior has changed.
Yeah, and `order_side_mismatch`. Definitely could be a problem on my side, but I have no idea why the behavior has changed.
Bumbling Bayesian
2026-06-29 02:08:19
↗
Kalshi
Kalshi API订单修改出现order_id_and_client_order_id_mismatch等错误
Kalshi
Kalshi API订单修改出现order_id_and_client_order_id_mismatch等错误
Cool. Now I am seeing a new error code, `order_ticker_mismatch`.
Cool. Now I am seeing a new error code, `order_ticker_mismatch`.
Bumbling Bayesian
2026-06-29 02:05:39
↗
Kalshi
Kalshi API订单修改出现order_id_and_client_order_id_mismatch等错误
Kalshi
Kalshi API订单修改出现order_id_and_client_order_id_mismatch等错误
It looks like there just might be significant lag now between what Get Orders reports for client_order_id and what the exchange has. My system kept restarting and falling over until I took out the client_order_id that was not matching.
It looks like there just might be significant lag now between what Get Orders reports for client_order_id and what the exchange has. My system kept restarting and falling over until I took out the client_order_id that was not matching.
Bumbling Bayesian
2026-06-29 01:56:12
↗
Kalshi
Kalshi API订单修改出现order_id_and_client_order_id_mismatch等错误
Kalshi
Kalshi API订单修改出现order_id_and_client_order_id_mismatch等错误
Yes, I am passing in the client_order_id. It is not yet clear to me why it is not matching. The first log indicated that I received an error on the amend, so I did not use the updated_client_order_id, but it seemed to be applied anyway.
Yes, I am passing in the client_order_id. It is not yet clear to me why it is not matching. The first log indicated that I received an error on the amend, so I did not use the updated_client_order_id, but it seemed to be applied anyway.
Bumbling Bayesian
2026-06-29 01:51:27
↗
Kalshi
Kalshi API订单修改出现order_id_and_client_order_id_mismatch等错误
Kalshi
Kalshi API订单修改出现order_id_and_client_order_id_mismatch等错误
(Looks like my previous discussion of the idempotence problem has been deleted in Kalshi's new data retention process.)
(Looks like my previous discussion of the idempotence problem has been deleted in Kalshi's new data retention process.)
Bumbling Bayesian
2026-06-29 01:37:14
↗
Kalshi
Kalshi API订单修改出现order_id_and_client_order_id_mismatch等错误
Kalshi
Kalshi API订单修改出现order_id_and_client_order_id_mismatch等错误
This seems to have something to do with the REST API being more flaky today and Amend Order not being idempotent. My system tries to amend an order, gets an error, retries the amend, and discovers that the amendment really did happen, despite the error.
This seems to have something to do with the REST API being more flaky today and Amend Order not being idempotent. My system tries to amend an order, gets an error, retries the amend, and discovers that the amendment really did happen, despite the error.
Bumbling Bayesian
2026-06-29 01:31:48
↗
Kalshi
Kalshi API订单修改出现order_id_and_client_order_id_mismatch等错误
Kalshi
Kalshi API订单修改出现order_id_and_client_order_id_mismatch等错误
Ah! This looks like the old idempotence problem.
Ah! This looks like the old idempotence problem.
Bumbling Bayesian
2026-06-29 01:05:28
↗
Kalshi
Kalshi API订单修改出现order_id_and_client_order_id_mismatch等错误
Kalshi
Kalshi API订单修改出现order_id_and_client_order_id_mismatch等错误
I am getting a new error code, `order_id_and_client_order_id_mismatch`, upon amend. Debugging now, but wondering if something changed.
I am getting a new error code, `order_id_and_client_order_id_mismatch`, upon amend. Debugging now, but wondering if something changed.
Shouldn’t be. But right now, the web UI continues to serve up NaNs and infinities. I doubt they have processes in place to handle sanitizing user inputs universally.
Shouldn’t be. But right now, the web UI continues to serve up NaNs and infinities. I doubt they have processes in place to handle sanitizing user inputs universally.
I would not be surprised if they never provide a way to name accounts due to the XSS risk.
I would not be surprised if they never provide a way to name accounts due to the XSS risk.
Yeah, so how much of the problem is Kalshi scalability issues and how much is bad behavior?
Yeah, so how much of the problem is Kalshi scalability issues and how much is bad behavior?
Is the guy who trades fewer RFQs doing something other than a bonafide trade? I don't understand the other possible motivations.
Is the guy who trades fewer RFQs doing something other than a bonafide trade? I don't understand the other possible motivations.
My v2 logic has the same as my v1 did, where I look for `not_found` and handle it special.
My v2 logic has the same as my v1 did, where I look for `not_found` and handle it special.
I thought v1 order endpoints were already gone.
I thought v1 order endpoints were already gone.
Yikes. That is confusing. I thought it was only about performance.
Yikes. That is confusing. I thought it was only about performance.
perps
perps
Your link was for margin, not predictions.
Your link was for margin, not predictions.
I see it on the website.
I see it on the website.
My bad. It is still there.
My bad. It is still there.
The website is generated from that asyncapi.yaml file.
The website is generated from that asyncapi.yaml file.
I see `status` in `userOrderPayload` appearing in asyncapi.yaml in my git repository on 2026-02-17 and it is gone on 2026-04-10.
I see `status` in `userOrderPayload` appearing in asyncapi.yaml in my git repository on 2026-02-17 and it is gone on 2026-04-10.
Maybe you are thinking of "Max 500k market subscriptions per session."? https://docs.kalshi.com/changelog#june-18-2026-9
Maybe you are thinking of "Max 500k market subscriptions per session."? https://docs.kalshi.com/changelog#june-18-2026-9
Thanks for this. I had not encountered and so didn't have `fill_or_kill_insufficient_resting_volume` in my enumeration of reverse-engineered error codes.
Thanks for this. I had not encountered and so didn't have `fill_or_kill_insufficient_resting_volume` in my enumeration of reverse-engineered error codes.
It would certainly not be the first time that components within Kalshi are unable to coordinate to satisfy a legal or contractual obligation. There is simply insufficient leadership at the top.
It would certainly not be the first time that components within Kalshi are unable to coordinate to satisfy a legal or contractual obligation. There is simply insufficient leadership at the top.
https://docs.kalshi.com/welcome
https://docs.kalshi.com/welcome
I am not having any trouble with it. Which properties are different?
I am not having any trouble with it. Which properties are different?
Expect new properties on new types to use the fixed point representation and to not need the _fp or _dollars suffix to disambiguate, because they won't have both forms.
Expect new properties on new types to use the fixed point representation and to not need the _fp or _dollars suffix to disambiguate, because they won't have both forms.
As I have said before, it would be great if the API served up its current OpenAPI spec.
As I have said before, it would be great if the API served up its current OpenAPI spec.
They often are representing something we are expected to migrate to.
They often are representing something we are expected to migrate to.
The docs do not necessarily represent what is currently deployed/executing.
The docs do not necessarily represent what is currently deployed/executing.
I’m going to try it! I need hurricane supplies anyway.
I’m going to try it! I need hurricane supplies anyway.
This looks intriguing…
This looks intriguing…
But if your requests are for the same market, they are probably going to be serialized at some point anyway.
But if your requests are for the same market, they are probably going to be serialized at some point anyway.
I think it really depends on your use case. If you can have multiple requests in flight in parallel and don’t wait for the response, it can be faster to take advantage of the load balancer.
I think it really depends on your use case. If you can have multiple requests in flight in parallel and don’t wait for the response, it can be faster to take advantage of the load balancer.
I tested various approaches using the CloudFront hostname a while back and found it could be faster in some cases to use the batching.
I tested various approaches using the CloudFront hostname a while back and found it could be faster in some cases to use the batching.
You may be misunderstanding what V2 means in this context. There is a V2 of the order management API under /trade-api/v2/, which also has the deprecated order management endpoints that worked up until a couple days ago. https://docs.kalshi.com/api-reference/orders/create-order-v2
You may be misunderstanding what V2 means in this context. There is a V2 of the order management API under /trade-api/v2/, which also has the deprecated order management endpoints that worked up until a couple days ago. https://docs.kalshi.com/api-reference/orders/create-order-v2
These guys are trying to help you. They know more than you. Consider listening.
These guys are trying to help you. They know more than you. Consider listening.
Maybe he has too much capital and would rather the exchange hold onto it for him instead of doing more trading.
Maybe he has too much capital and would rather the exchange hold onto it for him instead of doing more trading.
They say that it is impossible for them to do anything different on the UI. Maybe the front end team is at war with the API team. Sounds rough.
They say that it is impossible for them to do anything different on the UI. Maybe the front end team is at war with the API team. Sounds rough.
Scalability is hard. Let's go shopping.
Scalability is hard. Let's go shopping.
At 2026-06-20T00:30:48.713643Z, I got an error code of `get_milestones:_failed_to_get_milestones:_failed_to_deallocate_cached_statement(s):_read_tcp_10.0.82.251:49288->10.0.10.37:5432:_read:_connection_reset_by_peer` when trying to call Get Milestones.
At 2026-06-20T00:30:48.713643Z, I got an error code of `get_milestones:_failed_to_get_milestones:_failed_to_deallocate_cached_statement(s):_read_tcp_10.0.82.251:49288->10.0.10.37:5432:_read:_connection_reset_by_peer` when trying to call Get Milestones.