Jump to content

Deploy Plans for the week of 2017-07-17


Caleb Linden
 Share

You are about to reply to a thread that has been inactive for 2430 days.

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

Recommended Posts

  • Lindens

Second Life Server:

https://wiki.secondlife.com/wiki/Release_Notes/Second_Life_Server/17#17.07.11.327548

Scheduled Tuesday 2017-07-18 03:00-09:00 PDT

 

Second Life RC BlueSteel:

https://wiki.secondlife.com/wiki/Release_Notes/Second_Life_RC_BlueSteel/17#17.07.17.327657

Scheduled Wednesday 2017-07-19 07:00-10:30 PDT

 

Second Life RC LeTigre:

https://wiki.secondlife.com/wiki/Release_Notes/Second_Life_RC_LeTigre/17#17.07.17.327657

Scheduled Wednesday 2017-07-19 07:00-10:30 PDT

 

Second Life RC Magnum:

https://wiki.secondlife.com/wiki/Release_Notes/Second_Life_RC_Magnum/17#17.07.13.327620

Scheduled Wednesday 2017-07-19 07:00-10:30 PDT

Link to comment
Share on other sites

I've kind of lost track here, so I wonder if anybody knows: Is the Magnum version different from the other RCs because it includes that fix for DJ boards? Or is that fix everywhere now, and the difference is something else? (I don't see that fix showing up in the other, non-Magnum release notes, but not sure that means anything.)

Link to comment
Share on other sites

The changes that will affect the DJ boards is what's rolling to main channel today - the simulator OS update.
http://wiki.secondlife.com/wiki/Release_Notes/Second_Life_Server/17#17.07.11.327548

The new code rolling to magnum is just "internal fixes".
Looks like they changed their mind about what was being rolled to Magnum this week at the last minute: http://wiki.secondlife.com/w/index.php?title=Release_Notes%2FSecond_Life_RC_Magnum%2F17&diff=1205991&oldid=1205990

  • Like 1
Link to comment
Share on other sites

2 hours ago, Whirly Fizzle said:

The changes that will affect the DJ boards is what's rolling to main channel today - the simulator OS update.
http://wiki.secondlife.com/wiki/Release_Notes/Second_Life_Server/17#17.07.11.327548

The new code rolling to magnum is just "internal fixes".
Looks like they changed their mind about what was being rolled to Magnum this week at the last minute: http://wiki.secondlife.com/w/index.php?title=Release_Notes%2FSecond_Life_RC_Magnum%2F17&diff=1205991&oldid=1205990

Thanks for that Whirly... but I guess I'm easily confused because "

  • Scheduled for 2017-07-18 for Main Channel
  • OS upgrade (Nothing is changing functionality-wise)

" isn't the  "WARNING, THIS WILL BREAK EXISTING CONTENT" I was expecting.

Link to comment
Share on other sites

20 minutes ago, Whirly Fizzle said:

Well, hopefully not too much content will break.
As far as I'm aware, most of the big sellers of the DJ boards, including SHX,  have an update ready that should work fine on the new main channel version.

Yes, that's all well and good but Saii specifically asked folks on Main Server NOT to update until the DJ boards broke.  Now broken DJ boards don't fit my definition of "not much" and it was scant comfort that Oz made great efforts to work out just what WAS happening on Magnum (though we are all grateful that the effort WAS made).

Link to comment
Share on other sites

Logged in today to find my SHX song boards "broken" showing the http script errors as expected after the latest region update on the main channel. I applied the song board updates I received from SHX  but the script error is still there and the boards remain broken. Anyone else had this issue?

Link to comment
Share on other sites

I can't seem to get logged-in to the jira right now*  to check for recent, relevant reports, so I'm just gonna throw this out in hopes somebody else may have seen it:

Shortly after the weekly update, I noticed that my traffic maps weren't working. There's some internal http service inside, so I started trying to debug them -- the http stuff was working fine -- eventually tracing it as far as something inside Experience KVP, most likely failure of llUpdateKeyValue to actually update anything that other scripts could see. Before I could get any deeper into debugging, everything magically started working again, and apparently continued working for days -- until now, when it's failing again.

I'm loathe to delve in if it's a system thing that will cure itself, so... anybody else see this? (Might be limited to KVPs written in one sim and read in another, but haven't gotten that far into it to know for sure.)

*OpenID response from id.secondlife.com is timing out -- presumably unrelated to my XP KVP problem.

EDIT TO ADD: Well, I got back into the jira and don't see anything relevant. The KVP stuff still isn't working for me again, so I guess I'll try to debug further when I can get more than a couple minutes in-world.

Edited by Qie Niangao
Link to comment
Share on other sites

This is strange.  Given what Lucia reported above I am wondering just how many restarts and rolls some LeTigre sims actually had, because late last evening UK time, Wirtz, a LeTigre mainland sim was on 17.07.11.327548, the Main Server build.

Given all this could we have some clarification about just WHAT WENT WRONG, please Linden Lab?

 

@Qie Niangao:

Following some unpleasant spamming of a few JIRAs, the whole shebang was made Read-Only except for new entries or folk commenting on their own JIRAs

my guess is that this also has not worked as intended. (!)

ETA: Having said that it is working fine for me.  I guess LL just don't like you Qie! ;)

Edited by Ayesha Askham
additional information
Link to comment
Share on other sites

Well, I guess ignore my problem. I forced a one-time unchecked llUpdateKeyValue() that's usually checked, and for now things are working again. Probably it's just a condition I never handled correctly, but honestly, I don't know what to make of the fact it's worked fine for years then gotten stuck twice in as many days. Maybe it will all become clear eventually.

Link to comment
Share on other sites

Since directly after the "OS Upgrade" restarts on the 18th I keep getting disconnected on every main server region I try.
My home parcel is on a main server region and I usually stay there logged in 24/7 since last November with no real problems. Now after about 20 minutes on average I get disconnected with following messages:
 

WARNING: httpFailure: LLEventPollResponder::error: <5> got 498:
WARNING: httpFailure: Forcing disconnect due to stalled main region event poll.


I have tried other regions too with the same result. Also reset the router and got a new IP.
Then yesterday, thinking I was safe on RC regions I went to a BlueSteel place and got disconnected the same way. So I am in exile on a Magnum region since the 18th now (again, constantly logged in).
 
Perhaps related, teleporting is much slower than I am used to, with the following messages during the teleporting process:
 

2017-07-23T19:50:44Z WARNING: httpFailure: Unexpected HTTP error.  status: 498, reason:

 

Link to comment
Share on other sites

After trying other viewers I realized the disconnect itself happens on the client.
If you use Singularity and notice similar problems maybe try to remove lines 220 to 224 in https://github.com/singularity-viewer/SingularityViewer/blob/master/indra/newview/lleventpoll.cpp
 
 This does not change the observation that something on the 18th changed that causes it. I logged out during the rolls and logged back in directly after and it started immediately as described in my previous post.
 
I also get warnings in the old Radegast viewer that I have never seen before:
 

Unrecognized internal caps exception from https://sim8859.agni.lindenlab.com:12043/cap/84563704-5756-2613-2a06-740d7e342682:
   at System.Net.WebConnection.HandleError(WebExceptionStatus st, System.Exception e, System.String where)
   at System.Net.WebConnection.ReadDone(IAsyncResult result)


And Storm gives these that look similar (but no idea if it is normal as I usually do not use it):
 

WARNING: llcorehttp/_httppolicy.cpp(437) : stageAfterCompletion: HTTP request 0x178b4160 failed after 0 retries.  Reason:  Timeout was reached (Easy_28)
WARNING: llmessage/llcorehttputil.cpp(282) : onCompleted:  
--------------------------------------------------------------------------
 Error[Easy_28] cannot access url 'https://sim8859.agni.lindenlab.com:12043/cap/267e5f1d-974a-cf35-8aea-648888cb27a0' because Timeout was reached

 
This after teleporting:
 

WARNING: llmessage/llcircuit.cpp(1075) : checkCircuitTimeout: LLCircuitData::checkCircuitTimeout for 216.82.49.48:13016 last ping 105.223644s seconds ago.
WARNING: llmessage/llcircuit.cpp(1085) : checkCircuitTimeout: LLCircuitData::checkCircuitTimeout for 216.82.49.48:13016 still dead, dropping.


And sometimes this. 502 seems to be a typical timeout but could be related in this case:
 

WARNING: llcorehttp/_httppolicy.cpp(437) : stageAfterCompletion: HTTP request 0xd60aa30 failed after 0 retries.  Reason:  Bad Gateway (Http_502)
WARNING: llmessage/llcorehttputil.cpp(282) : onCompleted:  
--------------------------------------------------------------------------
 Error[Http_502] cannot access url 'https://sim10293.agni.lindenlab.com:12043/cap/6cc4e003-783a-db9b-dde2-98e15c64f282' because Bad Gateway


After disabling the timeout code on Singularity I am logged in for over 12 hours now and I notice no problem except parcel traffic, parcel banlist and parcel object owner list not loading. Script memory lists work.
 
Weird is how the issue seems rather uncommon afaik.
Radegast and Storm print the warnings about every minute on main AND magnum.
Singularity works as usual on magnum but always times out on main for that particular cap address.

I am in Europe btw.

Link to comment
Share on other sites

You are about to reply to a thread that has been inactive for 2430 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...