[BUG-229035] Inconsistent llGiveInventory to objects recently rezzed by llRezObject

[sl-service-account]

What just happened?

This issue involves a simple breedable. The parent object rezzes a child when a menu control ("Harvest") is clicked. Instead of completing further configuration, nothing happens until the menu control is clicked again. At this point the process completes but with an error ("object not found in inventory"), although the object is present and copyable.

Give a long time frame (minutes, not seconds), the process will sometimes complete on its own but for certain specimens data transfer between parent and child fails.

What were you doing when it happened?

Harvesting a breedable.

What were you expecting to happen instead?

Normally, the parent rezzes a child object ("ovule") which initiates handshaking with the parent. The parent then gives the child a copy of itself (another "ovule") and transfers various configuration data ("genetic code"). At this point, the ovule completes configuration and relabels itself as a "seed."

Other information

Some of the breedables transfer all data in one communication. If a second menu click (as described above) is done, this communication takes place. More advanced specimens transfer a second data string; this second communication fails.

We began observing this about two weeks ago, and it is not happening on all server versions. We replicated the issue on the Blue sim on mainland. The server version that we've observed the problem on is in the Environment field, and we're collecting info from customers about their server versions.

Links

Related

• [ ] https://github.com/secondlife/jira-archive/issues/6912

Original Jira Fields

| Field | Value | | ------------- | ------------- | | Issue | BUG-229035 | | Summary | Inconsistent llGiveInventory to objects recently rezzed by llRezObject | | Type | Bug | | Priority | Unset | | Status | Accepted | | Resolution | Triaged | | Reporter | BlueNote Cyberstar (bluenote.cyberstar) | | Created at | 2020-07-05T16:38:34Z | | Updated at | 2020-07-17T18:27:10Z | ``` { 'Build Id': 'unset', 'Business Unit': ['Platform'], 'Date of First Response': '2020-07-05T12:28:25.612-0500', "Is there anything you'd like to add?": "Some of the breedables transfer all data in one communication. If a second menu click (as described above) is done, this communication takes place. More advanced specimens transfer a second data string; this second communication fails.\r\n\r\nWe began observing this about two weeks ago, and it is not happening on all server versions. We replicated the issue on the Blue sim on mainland. The server version that we've observed the problem on is in the Environment field, and we're collecting info from customers about their server versions.", 'ReOpened Count': 0.0, 'Severity': 'Unset', 'System': 'SL Simulator', 'Target Viewer Version': 'viewer-development', 'What just happened?': 'This issue involves a simple breedable. The parent object rezzes a child when a menu control ("Harvest") is clicked. Instead of completing further configuration, nothing happens until the menu control is clicked again. At this point the process completes but with an error ("object not found in inventory"), although the object is present and copyable. \r\n\r\nGive a long time frame (minutes, not seconds), the process will sometimes complete on its own but for certain specimens data transfer between parent and child fails.', 'What were you doing when it happened?': 'Harvesting a breedable.', 'What were you expecting to happen instead?': 'Normally, the parent rezzes a child object ("ovule") which initiates handshaking with the parent. The parent then gives the child a copy of itself (another "ovule") and transfers various configuration data ("genetic code"). At this point, the ovule completes configuration and relabels itself as a "seed."', 'Where': 'http://maps.secondlife.com/secondlife/Blue/60/196/1250', } ```

[sl-service-account]

Lucia Nightfire commented at 2020-07-05T17:28:26Z

Sounds like another victim of https://jira.secondlife.com/browse/BUG-228939

[sl-service-account]

BlueNote Cyberstar commented at 2020-07-05T23:59:04Z

Thanks Lucia. The description of BUG-228939 is consistent with the problems I've observed. I will note that the problems at my mainland parcel (a "testing site" for the store) appeared about the time Second Life Server 2020-06-12T19:06:40.543526 was pushed out. Handshaking had up to that point addressed the problems that we had back in the debacle in November and December.

Since this occurs on some sims but not all, is there a suggested workaround? Most of our customers are having no problems, but if the bug is promulgated in the next server update, it's likely that they will. 

[sl-service-account]

BlueNote Cyberstar commented at 2020-07-09T22:32:27Z, updated at 2020-07-09T22:35:40Z

Following the rolling restart on 8 July 2020 (which also updated to Server 2020-07-01), the issues appear to be resolved, at least on the Blue sim. Of course, we've no idea whether the restart alone resolved the problem or if the newer Server version fixed it, but regardless, things are working as expected now.

We are now checking with customers who reported similar issues at other sims over the past couple of weeks.

Thanks!

[sl-service-account]

BlueNote Cyberstar commented at 2020-07-17T18:27:11Z

Following the rolling restart on 15 July 2020 (which updated to Server 2020-07-10) we are once again observing failure to transfer inventory & communicate data to child objects following llRezObjects.

This has been observed at Blue sim & Trance sim (owned by the developer of the breedable).