Jump to content

Deploys for the week of 2013-01-28


Maestro Linden
 Share

You are about to reply to a thread that has been inactive for 4094 days.

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

Recommended Posts

  • Lindens

All four channels will be rolled, this week.  We have one new project in RC, and two previously RC'ed projects which are returning with bug fixes.

 

Second Life Server (main channel)

The main channel is getting the region crossing performance project which was on BlueSteel and LeTigre last week.  This project makes sim performance smoother when objects and avatars cross between regions. 

https://wiki.secondlife.com/wiki/Release_Notes/Second_Life_Server/13#13.01.14.269010 

Scheduled Tuesday 2013-01-29 07:00-11:00 PST

 

Second Life RC BlueSteel:

BlueSteel is getting a new project, which adds server support for normal and specular maps.  This blog posthas more information about the project.  The next step is to release a project viewer, which will allow users to try out this new functionality.

https://wiki.secondlife.com/wiki/Release_Notes/Second_Life_RC_BlueSteel/13#13.01.26.269542

Scheduled Wednesday 2013-01-30 07:00-11:00 PST

 

Second Life RC LeTigre:

LeTigre is getting the maint-server project that was on Magnum 2 weeks ago.  The main change for this version is a bug fix for a places search issue which popped up in the previous RC iteration.

https://wiki.secondlife.com/wiki/Release_Notes/Second_Life_RC_LeTigre/13#13.01.25.269523

Scheduled Wednesday 2013-01-30 07:00-11:00 PST

 

Second Life RC Magnum:

Magnum is getting the interest list improvement project.  This projectshould reduce the bandwidth usage of viewers due to object updates, and should improve simulator performance, especially in sims with many connected avatars.  This week's update addresses an issue where some users experienced high packet loss, leading to issues such as not seeing avatars or objects inworld.

https://wiki.secondlife.com/wiki/Release_Notes/Second_Life_RC_Magnum/13#13.01.28.269574

Scheduled Wednesday 2013-01-30 07:00-11:00 PST

 

We will be monitoring this thread as the code gets released, so feel free to note any observations you have about the server updates.  If you have a specific bug you'd like to report, please file a Jira

Link to comment
Share on other sites

Could anyone please tell us what this server thing means for users in SL?

Me and friends have huge issus since 1 week, we freeze, crash, can´t rezz objects etc. Is really annoying. Nowhere to read about if this is normal during this server rollings. Please give some information, more understandable for SL Users. Thanks, would be very nice :)))

Link to comment
Share on other sites

  • Lindens


WolfBaginski Bearsfoot wrote:

Foolishlu, perhaps, I am assuming that with threaded region-crossing in the Main Channel, the code is also in all the RC versions.

Yes or no?

Hi WolfBaginski, that's correct - as a rule, the RC versions also include all the changes from the same week's main channel version.

Link to comment
Share on other sites

  • Lindens


L0RAINE wrote:

Could anyone please tell us what this server thing means for users in SL?

Me and friends have huge issus since 1 week, we freeze, crash, can´t rezz objects etc. Is really annoying. Nowhere to read about if this is normal during this server rollings. Please give some information, more understandable for SL Users. Thanks, would be very nice
:)
))

Hi L0RAINE, depending on which server channel you're in ('About Second Life' shows this), you might be seeing the networking issues that some have reported on Magnum.  The comments here explain the issue (as we know it) and give details about a potentially fixed version of the server:

http://community.secondlife.com/t5/Second-Life-Server/Deploys-for-the-week-of-2013-01-21/m-p/1856701#M9120

Link to comment
Share on other sites

Another thing that got promoted from BlueSteel regions is a crash mode with agent departure from regions upon their send home, ban and/or ejection.


I expect the frequency of these crashes will now increase grid-wide thanks to there being no resolution on this issue yet. I'm already seeing an increase now.

References:

https://jira.secondlife.com/browse/SEC-1204

Related:

https://jira.secondlife.com/browse/SEC-1215

Link to comment
Share on other sites

  • Lindens


Lucia Nightfire wrote:

Another thing that got promoted from BlueSteel regions is a crash mode with agent departure from regions upon their send home, ban and/or ejection.

 

I expect the frequency of these crashes will now increase grid-wide thanks to there being no resolution on this issue yet. I'm already seeing an increase now.

References:

Related:


Hi Lucia, I checked out those Jira issues you cite, and they also happened in Second Life Server 12.12.18.268345.

Link to comment
Share on other sites

I saw SEC-1215 happening on 12.12.18.268345.

 

But the crashing after use of moderation functions I've only seen on version 13.01.14.269010.

I run sandboxes in many regions, and we have them on all channels.

Last two weeks crashes after moderation function use were only happening in regions on server version 13.01.14.269010.

Since Main Channel began using version13.01.14.269010 today, I've had three crashes on the same main channel region immediately after send homes and temp bans were made.

 

Link to comment
Share on other sites

Hello folks..I am having the strangest problem...ever since my sim received the new weeks code this morning...I cannot TP out of it, or cross to the connecting sim by flying. I have trouble logging into the sim and during the log-in i see the message " requesting region capabilities...attempt 1...2...and then sticks on 3 a bit before finally logging me in. I can log into any other sim just fine and TP around to various other sims normally. However... my alt can login my sim just fine and TP in and out as if there was no problem. Also I have a blank friends and group list when I log into my sim...the alts are fine. I am stumped...anyone have any ideas? I have restarted the sim several times with no change.

Link to comment
Share on other sites

OK I seem to have figured out how to remedy my problem. I remembered that when the code deploy came...I wasnt paying attention and so got logged off by the sim restart. Well..I thought maybe I should do that again and see what happens.

So this time I restarted the sim with me in it...got logged off. Now the problem is fixed! Yayyyy

Link to comment
Share on other sites

Theres also issues with 13.01.14.269010 version regions while they are caching.

During region caching, parcel data is not updated for agents that teleport into the region. It remains that of the last parcel you were on in the region you came from. Land owners or staff cannot change parcel settings, check prim counts, return prims via the prim owner's list, check the access list or banlist or add/remove agents via the access list or banlist.

Agents are able to fly +/- 128 meters outside of the region boundaries and cannot region cross into other regions. Attempting to teleport out results in a crash.

https://jira.secondlife.com/browse/BUG-1516

Link to comment
Share on other sites

Since around 9 oclock slt last evening, I and an number of otther people have been plagued with a series of crashes that occur on what appears to be a two hour cycle.  I have checked my isp and it is stable and running relatively well, and have noticed that SL is now showing serious packet loss, at time high server ping and bandwidth as well. This is also accompanied by a massive use of virtual memory which seems to be a prelude to these crashes and can only e fixed by relogging our computer and dumping our bloated cache which is apparently a byproduct of SLs massive packet loss.

At this point this problem is becoming a game stopper and if not fixed will leave me no choice but to sell the rest of my land and leave SL as so many others have been forced to do over the last few weeks.  This problem did not begin till today's restart and appears closely associated with the work LL was doing on the servers this evening, and since it is impacting both main servers and RC servers would seem to be a problem in the core hardware.

We trust LL to get these restarts right and to carefully test them before installing them and this case is no exception, but since the code has until now worked well  on the RC servers from which it was taken, we must conclude that the problem is not in the code but in the hardware preparation that accompanied its installation. Given the fragile nature of the LL economy and the exodus of people and businesses leaving sl due to technical problems, one would think that LL would have been more careful in their preparation, but apparently they were not.  Now we can only hope that either the RC updates fix this problem or LL decides to correct the obvious mistakes they made last ning and this situation does not continue for days or even weeks as some of these problems have.  One more thing prior to about 6 slt this evening things were running smoothly but suddenly they fell apart in a series of serious crashes and since then have settled into the two hour crash cycle mentioned above.  To say that this in intolerable would be to do it an injustice, but I would say that it needs to be fixed soon or LL will loose whatever small amount of trust its player still have in its ability to maintain a stable playing environment.  

 

 

 

 

Link to comment
Share on other sites

I have been having crash issues since a restart last week, I did hit up live chat support as I couldn't access the land I had recently purchased.

For the last 12 months I have had a relatively trouble free time in second life, no matter where I went.
I was using WindowsXP from the start of 2012, and when given a chance to use a trial version of Windows8, I started using that in about June, 2012.
At first, I could not get into second life, and found that I had to run SL in XP mode. Life was good again for the rest of the year still using the Windows8 trial version, with the Phoenix Viewer.
At the begining of 2013, after having settling into a home in on the protected waterfront  property on the top of the Nautilus Continent, and building and renting a property of Villas on the waters of the Corsica Continent, three weeks into January, the trial ran out and I decided to take the full version of Windows8.
I found troubles straight away, not being able to log in, but soon sorted that out with updating the driver updates needed to run with the new program. 2 days after that, while at our buisness property on Corsica, a restart was warned for the region, so I went back to my home for the duration.
On return to the region in Corsica, I wasn't even able to rezz fully before my feet were frozen on the ground, and soon after the viewer crashed. I tried ALL of the viewers avasilable, with no luck in any. I could get to the regions each side of that region, but as soon as I crossed the sim line into the region with my property and objects, I was either frozen and viewer crashed, or dragged across that region, and the next 2 regions until I crashed out.
I did ask for in live chat for help, and was told to check my drivers, try the SL Viewer (which crashed instantly) and they did do another restart of that region on my request, as it seemed to be the trigger that started me not being able to access my property.
Thank you SL Support, but that didn't work, and I am still able to get everywhere except THAT region where I have buildings rented and land owned.I do hope the tenants don't have any requests of things needing fixing, as I cannot get there to look after them.
I wiped Windows8 from my HD, and clean installed Windows7 from start, as others say it runs well in SecondLife. Still, I have the same problems, even running in compatability mode for all choices to try and fix it. Can anybody help?


I am running on a GIGABYTE 880G-UD3H motherboard,
CPU: AMD Phenom II X4 955 Processor (3214.6 MHz)
Memory: 16381 MB
Graphics Card: AMD Radeon HD 6450
Windows Graphic Driver Version: 8.17.0010.1172

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

I have done some figuring out myself with my friend through sailing,
It seems to not be an issue with anything on my PC, as I run perfectly fine EVERYWHERE, except Regions on the RC Magnum.

All other regions, *Le Tigre, *Second Life and *Blue Steel are perfectly fine.
There must be something either NOT in RC Magnum, or maybe it has something in it that the others don't which is causing the fault.

Is there any way to include, or remove the Odd Factor in RC Magnum, to make it work like the other 3 Regions do?
Just a thought to head for where the problem lays.

Link to comment
Share on other sites

Some odd tings noticed yesterday sailing in Blake.  Alas it wasnt feasible to get hard data but one oddity was the handling of a sim border crash.   One time a nice pop-up appeared telling me (is this a step towards being able to PREVENT it? lol).the sim had lost touch with my avi (I forget the exact wording).

One innovation is that after the border crash and while "sailing to inifinity" the scritps on the boat coninued working (text HUD display updating as the boat moved, sails luffing appropriately etc) but the boat just could not be steered - totally non responsive to any keys (well, arrow keys or function keys).   I dont think Ive seen that before.

 

Link to comment
Share on other sites

Maestro and Andrew,

Unfortunately today's roll did not fix the bot excessive bandwidth/resend problem. It still shows the same pattern as before: downloaded data reaches 100 kB/sec within 2 to 3 minutes of landing in a Magnum sim. :(

Notice that is 100 kilobytes/sec, not 100 kilobits/sec. The data transferred reaches over 10GB daily. Pretty major if you ask me.

This chart shows 6 minutes of time, with download data on the left and upload data on the right. Time is shown along the bottom. The first minute or so shows the bot's normal usage while sitting in a main server channel region. The first spike you see is when I TP'ed the bot into a Magnum region. The interesting part to me is that the upload data (maybe some type of request?) continues to rise for a full minute even after the download data (responses?) saturates at 100kB/sec. The upload data eventually levels off too, at around 8 to 10 kB/sec.

Usage04.jpg

Second Life 3.3.4 (264214) Aug 31 2012 06:30:36 (Second Life Release) Release Notes

You are at 253,191.0, 256,260.0, 25.2 in Bay City - Falconmoon located at sim8018.agni.lindenlab.com (216.82.37.85:12035) Second Life RC Magnum 13.01.28.269574 Error fetching server release notes URL.

CPU: Intel® Core i7 CPU         950  @ 3.07GHz (3064.54 MHz) Memory: 12287 MB OS Version: Microsoft Windows 7 64-bit Service Pack 1 (Build 7601) Graphics Card Vendor: NVIDIA Corporation Graphics Card: GeForce GTX 260/PCIe/SSE2

Windows Graphics Driver Version: 9.18.0013.1090 OpenGL Version: 3.3.0

libcurl Version: libcurl/7.21.1 OpenSSL/0.9.8q zlib/1.2.5 c-ares/1.7.1 J2C Decoder Version: KDU v6.4.1 Audio Driver Version: FMOD version 3.750000 Qt Webkit Version: 4.7.1 (version number hard-coded) Voice Server Version: Not Connected Built with MSVC version 1600 Packets Lost: 86/269,306 (0.0%)

Link to comment
Share on other sites

My Viewer logs when I pass from one server version to another, and Le Tigre _> MC seems to be associated with failed region crossing. I'm not 100% sure, it might be the region crossing before the failure and I ought to have been logging to the second, but it's often enough to make me suspicious.

Generally, the code is working well. It's not even a failure every time. But I am not seeing the other RCs  appearing so often.

 

Link to comment
Share on other sites

Congratulations! It seems that the problem of showing diagonal regions has been solved in LeTigre:) again;)

My specs:

Second Life 3.4.3 (268262) Dec 17 2012 12:23:54 (Second Life Release)
Release Notes

You are at 287,757.0, 268,294.0, 22.9 in Nautilus - Suniaton located at sim9919.agni.lindenlab.com (216.82.46.61:13006)
Second Life RC LeTigre 13.01.25.269523
Error fetching server release notes URL.

:smileysurprised::):smileyvery-happy:

Link to comment
Share on other sites

Triple, with some details about the problem from Latif Khalifa I believe was able to find the real problem in the server that is sending so much data to libOpenMV bots, and probably too much data to regular viewers as well.

The SL UDP protocol appends ACKs for recieved packets on the end of outgoing packets that have extra room.  Meanwhile some of the UDP packets are "zero coded" which is a simple compression scheme sometimes applied to packets that have large blocks of zeros.  The bug (if my understanding of the code is correct) affects ACKs on the end of zero coded packets -- those ACKs may be lost or read incorrectly, which can make the protocol think that other packets have been lost --> it may resend.

The SL UDP protocol does not require a packet to be zero coded -- it is up to the sender to decide to compress its packets or not.  The receiver should be able to handle them in either case  So, in theory there is a workaround: configure libOpenMV to NOT send any zero coded packets.  This configuration may be beyond most libOpenMV users, but I figure I would mention it in case there are any intrepid residents who want to try it before we can get an update to Magnum.

Link to comment
Share on other sites

Very good!

I can easily park the bot in a non-Magnum region when it is idle, so I'm ok. I am more worried about the users that don't know about this. There is no outward sign of a problem, so anyone on a metered internet connection is in for a (quite large) shock when they get their next bill.

Thanks to you and Latif for looking into it.

 

Link to comment
Share on other sites

Firestorm and  stock viewer users also epierence problems with excessive chat repating , notices,   excessive data.

What about clients which use "optimized" kinds of connection to SL - like LTE users, G3, smartphones, etc.

At home over landline I use Singularity   viewer which doesn't perform zero coding.. But when I use   LTE modem  with same client, when  being in different  place (about 1/3 of my online time) I get that chat repeating and excessive packets. Before  this Magnum RC update   such effects never happen on G4 modem and very rarely while using tablet as G3 modem or using  Lumiya (rare cases of course can be explained by lost packets).

G3\G4 providers often use   special software which manages automatic logon of loked modem to account and creates  virtual network adapter.. So traffic gets compressed and sent  other end of this propietary "tunnel", Same problem I have now with my Android tablet client (Lumiya).

Slight experimentation with primitive  USP send\recieve example code on C++  to send and receive UDP packets   through G4  to external server , I own,   WIreshark had shown that  sent data was modified.

 

Link to comment
Share on other sites


Dora Gustafson wrote:

Congratulations
!
It seems
that the problem of
showing
diagonal
regions
has been solved
in
LeTigre:) again;)

My specs:

Second Life 3.4.3 (268262) Dec 17 2012 12:23:54 (Second Life Release)

Release Notes

 

You are at 287,757.0, 268,294.0, 22.9 in Nautilus - Suniaton located at sim9919.agni.lindenlab.com (216.82.46.61:13006)

Second Life RC LeTigre 13.01.25.269523

Error fetching server release notes URL.

:smileysurprised:
:)
:smileyvery-happy:

To day, after the servers have been rolled out all around Nautilus - Suniaton, I am not so enthusiastic anymore:(

Lost diagonal region_001.jpg

Link to comment
Share on other sites

You are about to reply to a thread that has been inactive for 4094 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...