Why do I keep getting "degraded performance - log in somewhere else for the next thirty minutes" messages.

[Chic Aeon]

This keeps happening. Last time my scripts weren't working and things weren't loading. This time (I only stayed a minute then logged off ) my animations weren't working. It is "me" as the sim is carefully designed for no lag and minimal scripts and I get 128 fps on the ground.  

 

I have NEVER seen this message before the last week or so.  Not boding well. 

[Profaitchikenz Haiku]

Just a guess, in the viewer stats check two things, the physics memory details (if there's a growing amount of RAM usage you're heading for the "unable to rezz object that has caused problems in this region" message), and then lower down, look at the timings and see what the spare time and script times are. There might be a hint there that your region needs a restart to clear a problem.

[Qie Niangao]

Are we talking about Umber? Because it has oodles of Spare Time and (unsurprisingly) rock-solid 45 FPS performance. And only 82MB physics memory allocated, so there's just nothing I can see wrong with this region. So if this message is arriving while on Umber, I'd first want to confirm that it's really a message from the server and not something from a HUD or conceivably the viewer—although it's sure not obvious why those would suggest going to another region.

[Edit: Ah, I now see "It is 'me'…" where I'd misread "Is it 'me'…?" so yeah, you already know it's not the region. I'm getting network delay vibes, but that's a pain to diagnose, so maybe first try with a different viewer? And/or with all attachments removed? just hoping it's not some weird network thing.]

Edited October 23, 2022 by Qie Niangao

[Vicious Hollow]

Curious about this also, as I've been getting it off and on for a few months now. It doesn't happen when I 'just' log in, but rather, after a cache clear, and more specifically, an inventory cache clear. When I do the "clears" I log in at a skybox we have, not at our home. I thought, since the skybox is at our shop, might be script or even lag related due to the number of vendors and displays we have out, but, when trying a full cache clear, then logging in at home, I still got the message. 

 

To make matters even more strange, when doing a full clear, it doesn't happen every time. For example, yesterday morning, I do a texture/inventory clear, relog to our skybox, got the message, tp'ed back home after my inventory loaded, relogged again and everything was fine. Then after some traveling, I followed the same steps of texture/inventory clearing, relogged at the skybox, and got no message. Then this morning, no matter what combo I tried (clearing texture cache, relogging, then clearing inventory cache vs clearing them both and relogging) it happened. I also got the message when tp'ing back home after getting the message at the skybox.

 

As mentioned by the OP when this happens, I also can't move and no animations work, plus I change clothing until I tp back home. However when I got the message this morning, when tp'ing back home, I expected not to be able to change clothing or move, but could. Also, a side note, when I do these clears, I remove all attachments and at our skybox (in Badger) I get well over 200 fps.

[Ardy Lay]

I only see this happen when I am testing residential routers.  Some just run out of resources trying to track flows.  If they cannot add a new a flow because their NAT table is full of flows that haven't either been explicitly closed or empirically expired due to inactivity, then the client fails.  When Second Life Viewer experiences this sort of failure, it raises this message.

It's quite possible that somewhere in the path is a device whose capacity is being exceeded.  Not all capacity limits in routers equate to interface congestion or "buffer bloat".

If it's your router you can find out quickly by giving it an old-fashioned restart.  If it's a Carrier Grade Network Address Translator running out of flow-tracking table space, then you can see an improvement by converting all of your flows through it into a single pair of flows by using a VPN whose connection passes through it.  Have a look at the IP address your access provider is issuing to your own router's WAN interface (or your computer if you run nude on the Internet).  Is it in the network 100.64.0.0 /16?  That's a definite indication of CGN.  There are other address spaces used on the "inside" of a CGN too, but only this one is designated for Carrier Grade NAT.  Using others thusly just confuses residential routers that think they are smart.

Edited October 25, 2022 by Ardy Lay

[animats]

4 hours ago, Ardy Lay said:

I only see this happen when I am testing residential routers.  Some just run out of resources trying to track flows.

Hm. I hadn't thought about that resource before. Your viewer is talking to every region you can see, even partially. Each has a different IP address. And routers may still have resources tied up for recently visited regions. Are there routers that can't handle that?

[Love Zhaoying]

Does this happen for your Alts too? Wasn't it you who also had the issue with some touch events for only one login?

[Chic Aeon]

Hi and sorry for not getting back here.  Yesterday was a blur. 

2 hours ago, Love Zhaoying said:

Does this happen for your Alts too? Wasn't it you who also had the issue with some touch events for only one login?

I haven't tried with an alt when the problem is happening although I haven't had it happen during an online time with an alt and they have been on a fair amount.  This DID happen seemingly before the issue with touch (and most touch things are working fine still for me, just the payment kiosk that I know of). 

On 10/23/2022 at 10:26 AM, Qie Niangao said:

Are we talking about Umber?

The FIRST time this happened a couple of weeks ago it was both STRAND (Artscape Homestead) and Umber (shop). That lead me to believe it was fairly widespread. It also lasted quite awhile but by morning all was well.  Then it happened a week ago or so later.

"I" do not freeze (I remember that freezing from long ago). The first time I could do many things even on the sims in question but for example my WINDOWS (Trompe Loeil building) had closed "by themselves" and clicking on them didn't turn them back to glass.   Their were other similar things but no specifics that I can remember. 

 

The only other odd thing that has been happening at STRAND (I have mostly just been on my building pad at UMBER) was that a few plant textures have reverted to  non-alpha.  I finally figured out what was going on there and its is a very easy fix. This was happening before I ever saw the degraded message. It is in a small box  on a dark background, not a blue "script" box. 

I will try the router power cycle. This doesn't last long and is less frustrating than the scripts don't work on the kiosk thing. They are putting fiber in around me (not to me yet) and that might possibly have some baring. I did hear from other folks that they have been having major sim issues. I didn't ask what exactly.  Both Umber and Strand are very low script and low lag and kept that way purposefully so I don't think it is a overuse of resource issue.  

 

That's all I can think of.  Thanks for all your input. 

[Love Zhaoying]

3 hours ago, Chic Aeon said:

6 hours ago, Love Zhaoying said:

Does this happen for your Alts too? Wasn't it you who also had the issue with some touch events for only one login?

I haven't tried with an alt when the problem is happening although I haven't had it happen during an online time with an alt and they have been on a fair amount.  This DID happen seemingly before the issue with touch (and most touch things are working fine still for me, just the payment kiosk that I know of). 

I had the crazy thought that maybe "your" user has response time issues due to more inventory but your alt's wouldn't! 

Sorry for dumb questions!

[Love Zhaoying]

3 hours ago, Chic Aeon said:

"I" do not freeze (I remember that freezing from long ago).

Only teensy-bit off-topic: I had only started freezing extremely recently, moved my cache to SSD and it stopped. Yays.

But on-topic: Are you using SSD?

[Ardy Lay]

11 hours ago, animats said:

Hm. I hadn't thought about that resource before. Your viewer is talking to every region you can see, even partially. Each has a different IP address. And routers may still have resources tied up for recently visited regions. Are there routers that can't handle that?

There are residential routers that crap out at 256 flows, but not many of them are still working.  They are 20 years old now.

There some that blow up at 16386 or so.

And there are some that survive fine all the way up to 65534 or so.

Most run out between 16K and 64K flows.

I am not sure how many flows-per-region a Second Life Viewer requires but it certainly isn't ONE.  It's dozens, plus CDN and other service components.

It's quite possible for the number of active flows a residential router can support to decline with up-time.  I have a CGN platform that has to be restarted at about 45-day intervals, most likely because of memory leaks or some other software issue.  I am migrating subscribers off of it so I can retire it.  It still has many global address-and socket tupples to assign but it cannot add them to the tables due to HEAP allocation failures, prompting a reboot.

I would not expect this symptom to vary much from one SL account to another on a given connection, but I don't have a monster-Inventory to test with.

[Chic Aeon]

9 hours ago, Love Zhaoying said:

I had the crazy thought that maybe "your" user has response time issues due to more inventory but your alt's wouldn't! 

Sorry for dumb questions!

It doesn't seem to have anything to do with that -- not that kind of message since it is about the REGION being faulty. My inventory is only 60K after 15 years. I keep LOTS of backups of things (sometimes that does NO good at all) but I through away a LOT of stuff too

9 hours ago, Love Zhaoying said:

Only teensy-bit off-topic: I had only started freezing extremely recently, moved my cache to SSD and it stopped. Yays.

But on-topic: Are you using SSD?

Yes all on a SSD and has been that way on this machine for a year and a half. I haven't knowingly changed anything. Windows updates sometimes make a mess of various programs but this really doesn't "feel" like that sort of thing either.  Oftentimes when I have these bizarre issues (not just SL) the problems go away by themselves eventually so I am hoping that they will again.  Just all pretty odd. 

 

[Mazidox Linden]

Disclaimer: I don't know why, and it's not a cureall for this issue let alone other issues,  but sometimes when you get the degraded performance message, it's related to one of the landmarks in your favorites no longer existing. If you clear out your favorite landmarks it may resolve the issue.

 

(And yes, this is definitely on the Long List Of Things To Fix, but that's not the same as "don't file bug reports/support tickets/forum threads about it", we use all of that information to figure out what current pain points are.)

[Ardy Lay]

1 hour ago, Mazidox Linden said:

Disclaimer: I don't know why, and it's not a cureall for this issue let alone other issues,  but sometimes when you get the degraded performance message, it's related to one of the landmarks in your favorites no longer existing. If you clear out your favorite landmarks it may resolve the issue.

 

(And yes, this is definitely on the Long List Of Things To Fix, but that's not the same as "don't file bug reports/support tickets/forum threads about it", we use all of that information to figure out what current pain points are.)

That's weird.

[Mazidox Linden]

Yes, yes it is

[Monty Linden]

That message arises when the UDP stream coming from the viewer is particularly well-scrambled.  Look for problems on the local network:  multiple network paths, unhappy router or cable modem, kids up to no good.  If it occurs again, given an exact time and place and I might find a clue here.  (And the viewer logs may have something interesting to say.)