Skip to content
TORNLIFE More

Last Action update

Started by _SCOFIELD_ [1441750] on in Bugs & Issues.

11 replies · 211 views · thread synced · 9 days ago · View on torn.com
About this thread

Posts archived: 12 / 12 posts (100%) · the total is Torn's reply count + the opening post at the last fetch

Counted by TornLife from the archived posts.

Archived posts
12
Discussion span
→
People posting
4
Likes on archived posts
11
Posts by staff, officers and moderators
4
Authority score
59 / 100
Historical score
42 / 100
Story score
48 / 100
Engagement score
63 / 100

Most-liked replies

Mentioned in this thread

Dahn [2317379] ×1 TheCulturedOne [3631262] ×1 Muppet [3926388] ×1

_SCOFIELD_ [1441750]

Sometimes unlisting items from the item market does not update user's last action

 

 

Muppet [3926388]: 14:35:46 - 03/04/26 You removed a Dune Vest (Insurmountable) from the item market

 

at 2026-04-03 14:36:48 (API check) - User's Last Action: 2026-04-03 14:35:28

 

they unlisted an item but last action still showed from 18 seconds before

Mentions: Muppet [3926388]

CloudJumper [1636201]Staff

Are you certain that you are not viewing cached data from the API? The last action will either always update or never update. It can't bug out occasionally, unless there's something specific that the player is doing differently.

splent [2088243] Admin Staff

Hey, these specific selections you mentioned are cached by the global cache, however all API requests are cached by service cache.

I've described this in more detail on the API docs page now: https://www.torn.com/api.html 

 

If you need 100% fresh data for any selection, you need to ensure your request is unique. 

Probably the easiest way to do that is to pass `timestamp` query value with the value of current time in seconds, though it can also be achieved with other values or query parameters.

 

I hope that clarifies things a bit!

_SCOFIELD_ [1441750]

I do pass the timestamp query in all my API requests to get the latest data.

 

Edit: So the data should not have been cached.

_SCOFIELD_ [1441750]
  1. Dahn [2317379] Listing Vanished, Idle | Okay | Last Action: 2026-04-10 18:55:05, @ 2026-04-10 18:56:43 18373754905: {"name":"Vanguard Pants", "itemID":"1357", "listingID":64881002, "price":5000000000, "amount":1, "userID":2317379, "userName":"Dahn"}

     

    Dahn [2317379] Listing Vanished, Idle | Okay | Last Action: 2026-04-10 18:55:05, @ 2026-04-10 18:56:50 18086742881: {"name":"Vanguard Body", "itemID":"1356", "listingID":64322585, "price":12000000000, "amount":1, "userID":2317379, "userName":"Dahn"}

 

API pinged at 18:56:43 - that means he must have unlisted between 18:55:13 - 18:55:43

but Last action shows 18:55:05

 

If API is correct, then torn didn’t update last action.
Or API didn’t update.

Now sure which, but there’s a bug here

Mentions: Dahn [2317379]

_SCOFIELD_ [1441750]

TheCulturedOne: 19:50:47 - 10/04/26 You removed a Qsz-92 (Assassinate) from the item market

 

He unlisted at 19:50:47 - but last action on API was 19:50:33 - So Either Last Action didn't update on torn, or API didn't update it correctly

 

TheCulturedOne [3631262] Listing Vanished, Idle | Okay | Last Action: 2026-04-10 19:50:33, @ 2026-04-10 19:52:04 18500539357: {"name":"Qsz-92", "itemID":"248", "listingID":66401280, "price":950000000, "amount":1, "userID":3631262, "userName":"TheCulturedOne"}

 

 

 

 

Mentions: TheCulturedOne [3631262]

splent [2088243] Admin Staff

Hey, 

 

Thanks for the details. 

It looks like there was a temporary DB2 replication issue, spanning from 10-04-2026 05:00 UTC until roughly 11-04-2026 13:00 UTC, and a few more issues before 10th April, which could explain all of this.

I just tested item market interaction logic, and there are no issues with updating the last action, which means the DB contains correct values. 

 

Because API is only serving the DB values, I am not sure what could go wrong here.

There's no caching on the last action, and service cache can't last for more than 30s (otherwise all selections would have a problem), so I'm really inclined to believe this was a temporary issue related to db2 replication.

I'll close the bug for now, but if it happens again in the future please feel free to open a new thread again. 

 

Thanks!

splent [2088243] Admin Staff

Update! 

After this error was reported once again, I made some additional tests. It looks like I was wrong and API times could deviate up to 60s from the real values. 


This should now be fixed properly. 

Thanks!

Apatxe [2813453]

Doing the work to allow more mug bots? Why not leave that “error” as feature if the intent is to provide window of 60 seconds anyways for global cache? Torn is becoming a bot heaven…