Jump to content

Why are we still waiting?


Aishagain
 Share

You are about to reply to a thread that has been inactive for 1734 days.

Please take a moment to consider if this thread is worth bumping.

Recommended Posts

It is difficult to know what is going on with the servers. Inara and I blog what is known. But, often the Lindens aren't sure. Simon Linden at the Server-Scripting UG Tuesday said he didn't know what all was in the updates rolling to the RC channels.

Rider Linden knew that some of his fixes were going to the RC channel today. But, made no statement on which channel.

Also, there are things changing that the Lindens say their work didn't change... For instance Script Run % is better or a continuing problem depending on who you talk to. But, no fixes for that issue have rolled. With Tuesday's update I have had fewer problems with disconnects when teleporting or crossing region boundaries. Before that disconnecting was an aggravating problem for me that was growing since the fix pushed 6 or 8 weeks ago. Did the servers change? Yeah, but according to the Lindens the update should not have affected that problem. So, was it the servers, my network, or my computer? No way to know.

My thinking is there is no single Linden that knows what is happening with the servers that is talking about it.

  • Like 1
  • Thanks 1
Link to comment
Share on other sites

I would not expect them to detail internal fixes for various reasons, they might be coding oversights that got noticed, they might be hooks going in for future-updates, somebody discovered a back-door planted yonks ago and closed it.... Some stuff you just don't tell to the passengers in case they panic

But yes, I'm intrigued that some of the script percentage improvements happened on their own, but equally, the problem sort of crept up all on it's own in the first place, didn't it?

Link to comment
Share on other sites

On 8/21/2019 at 6:53 PM, Grumpity Linden said:

"Internal Script Improvements" are rolling to all RCs today.  Sorry we dropped the ball on release notes. 

@ Grumpity Linden and Bugsly Linden

Internal Script Improvements may sound nice, but for technically savvy users or region owners or region managers it's just a simple generic term. Remember the old days when Oscar wrote the release notes. The information was much more extensive and informative. This should be reintroduced, even if it involves a bit more effort.

  • Like 3
Link to comment
Share on other sites

4 hours ago, Miller Thor said:

Internal Script Improvements may sound nice, but for technically savvy users or region owners or region managers it's just a simple generic term.

Your point is well taken; more information would definitely be welcome. Just in case it's of interest, this specific change was discussed at the Tuesday Server User Group. When asked which script events are involved in the improvements at this time:

Quote

[2019/08/20 12:13]  Rider Linden: Sensors and listens are the big ones.
[... later...]
[2019/08/20 12:45]  Rider Linden: The only ones that are visible this time around.  (there are some other things under the hood too).  Some of the other events and scheduling overhead will require more thought and I'd like to let them cook in my head for a while before I tackle them.

Now, as far as I know, no RC version with these improvement is currently running on the main grid (Agni); rather, it appears the RC regions have reverted to the release that was installed on the main server channel. 

That's another opportunity to improve communications: when the announced server installations need to roll back or change, it would be helpful if there were some information about the changes.

I get it that there may be some reluctance to announce where new vulnerabilities may be exposed, but I'm guessing most folks who read the release notes and these threads are motivated to help, and would benefit from knowing better when, where, and how to participate.

  • Like 1
  • Thanks 3
Link to comment
Share on other sites

My...I wondered when Oskar's name would resurface, Miller!

Considering how much better Wirtz seemed to run for the hours that the second set of script run changes were in place (530270), granted after a further restart and how much WORSE it is now, some sort of explanation for the "rollback" would be welcome and reassuring that LL actually were "on it".

It's not that I don't think that Rider knows what he's doing, I just don't understand the logic of the past few days' changes.

Edited by Aishagain
Additional text
Link to comment
Share on other sites

 Aishagain to know that your too young in SL. The Deploy notes Oskar Linden wrote looked like so 

 and so it was really informative for everyone.

The Logic behind some Rollbacks we had during the last Deploys maybe results from found Issues witch have been found after Linden updated the Regions.

The Ways in the IT are sometimes strange but its good when Issues got found and are now on the Way to get resolved and we have a stabile Regionsoftware running until the repaired new Software comes with the next Deploy.

Link to comment
Share on other sites

@Miller Thor:  This account may be too young to remember Oskar, but I've been around SL since 2008, and my memory (such as it is) goes back well before Oskar's time.

The current Region Software may be stable, but it has a serious memory leak which means that unless LL restart Wirtz two or three times a week, we can use the rubber bands we get in profusion as rapid-fire catapults! Also script run staggers down from 25 to under 10% quite quickly.

Link to comment
Share on other sites

@AishagainI just looked at this region. With about 9000 active Scripts, the normal Amount for a good Performance is 2000 to 3000, i'm not surprised when the value slowly moves down. I monitored the area for about 20 minutes and the value fluctuated between 25 and 40%. On average, it was about 32%. With such a load of approximately 16.5ms script time in the current state of things a normal value. I can only guess at the current state of things to a daily restart and should after the first restart the value after about 5 minutes below 20% and remain so is the region again to start until the value settles at about 40%. But even after a hopefully coming soon and stable update, I advise to continue to restate the region once a day because even the best update will probably hardly a massive performance increase in this massive script load result.

Edited by Miller Thor
Link to comment
Share on other sites

You are about to reply to a thread that has been inactive for 1734 days.

Please take a moment to consider if this thread is worth bumping.

Please sign in to comment

You will be able to leave a comment after signing in



Sign In Now
 Share

×
×
  • Create New...