"Too many error returns for script" since last region restart

[Samara Sharkfin]

I just put it here and watch as i can't do much.

Since the last region restarts I have an old soundscape objects that since today gives me the error "Too many error returns for script." every couple of seconds. Tried that in 3 different regions with the same result.

I can't reset it because it's no mod, and re-rezzing does not help either.

I guess it's something with the last Region update:   Second Life Server 2021-02-01.555570

 

[Whirly Fizzle]

1 hour ago, Samara Sharkfin said:

I just put it here and watch as i can't do much.

Since the last region restarts I have an old soundscape objects that since today gives me the error "Too many error returns for script." every couple of seconds. Tried that in 3 different regions with the same result.

I can't reset it because it's no mod, and re-rezzing does not help either.

I guess it's something with the last Region update:   Second Life Server 2021-02-01.555570

 

Looking at Rider's comment on BUG-230126 - "Too many error returns for script." error from a HUD while on regions on Second Life Server 2021-01-20.555189
it's likely to be a script problem the creator needs to fix.
Where is the soundscape object rezzed?

@Rider Linden

[Samara Sharkfin]

2 hours ago, Whirly Fizzle said:

Where is the soundscape object rezzed?

Thanks for replying Whirly. I read the JIRA now and understand that it might be a scripting problem, but that appeared only lately and the objects are rezzed since years. That's why i am maybe to trigger happy to shoot right at the region update 🙂
Meanwhile i rezzed some different ones of those ambient sound objects from the same maker, and the error seem to appear as some sound loops are played (but no sounds sound to be missing). Also it is more sporadic as i thought at the beginning (i swear it was stubbornly reproducible), it might go minutes until the error start to happen, then it repeats, probably as the loops loop.

So not sure if you really want to see them. If yes, i send you an NC just in case, and because they are on private land.

[Rider Linden]

2 hours ago, Whirly Fizzle said:

Looking at Rider's comment on BUG-230126 - "Too many error returns for script." error from a HUD while on regions on Second Life Server 2021-01-20.555189
it's likely to be a script problem the creator needs to fix.
Where is the soundscape object rezzed?

@Rider Linden

Yes.  Whirly is correct. This is a script issue that the creator will need to address. 

The script is making repeated HTTP calls to a server which is returning 500 class status responses(see below). In order to trigger that warning it needs to make quite a few requests in a very short amount of time with no non-500 responses.  In more poetic terms:  The script is sitting in the back seat of the car asking "arewethereyet?arewethereyet?arewethereyet?arewethereyet?howboutnow?arewethereyet?"

HTTP response can be broken down into 4 broad ranges:

• 200: All is right with the world
• 300: What you are looking for is over there
• 400: I don't know what you want or you asked wrong. Try again
• 500: Server is on fire

 

[Samara Sharkfin]

 

6 minutes ago, Rider Linden said:

500: Server is on fire

Okay understood.

I need fire truck ambient sounds. Got it!

Thanks!

[Rider Linden]

3 minutes ago, Samara Sharkfin said:

 

I need fire truck ambient sounds. Got it!

 

 

[Anaya Twine]

I am getting this as well since last night "Too many error returns for script" from a HUD

(I am on the new Firestorm btw) I have send the creator a notecard and waiting..

[AWalphaomega]

Same issue.  I don't think it should be placed on the shoulders of the script creators when things that worked for years suddenly don't, right after a region update.

 

Mine is Nuko AO 1.2.  the builder has not responded.

Edited February 11, 2021 by AWalphaomega

[Luhkey]

On 2/11/2021 at 7:50 AM, AWalphaomega said:

Same issue.  I don't think it should be placed on the shoulders of the script creators when things that worked for years suddenly don't, right after a region update.

 

Mine is Nuko AO 1.2.  the builder has not responded.

A friend of mine has the same problem with Nuko AO and the creator has not responded.  Are you sure this is a creator issue?  It was working fine, according to my friend, just a day or so ago, and suddenly is not.  Wouldn't that be a server issue that has broken something that previously worked fine?

[Tonk Tomcat]

Got those error messages from an old and abandoned E2V delivery system what i still use for some older products. I guess you will reveive those error messages, because the used server does not exist anymore. Because you don't receive any respond from the Nuka HUD creator i guess he left and the server is gone, so probably it will not be fixed anymore.

[Ardy Lay]

Could it be SL is now reporting errors that were previously silently ignored?

[animats]

I got that last night from ZiSwim, an old swim HUD. I wonder if it's trying to phone home to a long-dead server. The creator is gone from SL.

[Oz Linden]

On 2/15/2021 at 9:58 AM, Ardy Lay said:

Could it be SL is now reporting errors that were previously silently ignored?

Exactly so.

Since uplift, all http-out requests from simulators go through shared proxies (as I explained in another thread). It used to be that scripts that ignored error responses and repeated the requests only impacted the performance of that one region, but now they also impact both the performance and reliability of other scripts across the grid making requests, but they also harm the reputation of SL (causing hosting providers to block access from our services).

By sending these error reports, we're trying to get owners to discontinue using scripts that are just repeating requests that don't work.

[bigmoe Whitfield]

6 hours ago, Oz Linden said:

Exactly so.

Since uplift, all http-out requests from simulators go through shared proxies (as I explained in another thread). It used to be that scripts that ignored error responses and repeated the requests only impacted the performance of that one region, but now they also impact both the performance and reliability of other scripts across the grid making requests, but they also harm the reputation of SL (causing hosting providers to block access from our services).

By sending these error reports, we're trying to get owners to discontinue using scripts that are just repeating requests that don't work.

if anything I have on the grid, with any of my accounts,  caused the issue,  LL can disable it and point me to which one owns it and I'll go fix it.  I'm bad with remembering many things.

[Coffee Pancake]

Might be an idea to put in some kind of grid wide error cap shared by all instances of a script (identified by author, date & crc), once the cap is hit no more http.

Scripted products that depend on external data (often beyond the authors control) can't be expected to work flawlessly forever. This isn't really breaking old content, the content was broken.

[animats]

11 hours ago, Coffee Pancake said:

Might be an idea to put in some kind of grid wide error cap shared by all instances of a script (identified by author, date & crc), once the cap is hit no more http.

Useful only if that feeds into Marketplace, and marks the product as defective.

[arabellajones]

On 2/18/2021 at 4:35 PM, Oz Linden said:

Exactly so.

Since uplift, all http-out requests from simulators go through shared proxies (as I explained in another thread). It used to be that scripts that ignored error responses and repeated the requests only impacted the performance of that one region, but now they also impact both the performance and reliability of other scripts across the grid making requests, but they also harm the reputation of SL (causing hosting providers to block access from our services).

By sending these error reports, we're trying to get owners to discontinue using scripts that are just repeating requests that don't work.

Thanks.

This particular issue really does need clearly documenting in the SL Wiki

I was lucky, the item which has started spewing those error messages is one with scripts I can read and modify.

Many are not.

[Oz Linden]

On 2/19/2021 at 1:11 AM, Coffee Pancake said:

Might be an idea to put in some kind of grid wide error cap shared by all instances of a script

Unfortunately, that's technically fairly difficult

[iceing Braveheart]

since you guys like amazon services so much why not?

https://aws.amazon.com/lumberyard/

https://en.wikipedia.org/wiki/Amazon_Lumberyard