Aishagain
-
Posts
1,448 -
Joined
-
Last visited
Content Type
Forums
Blogs
Knowledge Base
Posts posted by Aishagain
-
-
Thing is, the server version rolled to Main is not what was on LeTigre and BlueSteel last week, and there was no indication that Magnum was rolled at all last week. There are no release notes in the viewer either this time, they simply show what was rolled to Main last week.
@Cindy Evanier: I'm too old and ignorant to know what FIFY means
-
I guess the answer was "NO".
-
1
-
-
The GSP tells us that there will be rolling restarts on Main Server and the Release Candidate channels, yet less than two hours before this work begins there is nothing about them in this thread once again.
Please may we know what is to be rolled?
-
Quite so, Nal, and I wasn't implying that server work had been done (or at least I hadn't intended to give that impression). My third paragraph was perhaps misleading in that regard. The server-side script performance issue is indeed being given a lot of attention by LL now, and I was surprised and pleased to see Rider Linden on our homestead, having a look at our set-up.
I was relieved to be told it all "looked pretty reasonable" and that the low script-run % was not something he expected. He did, however, vouchsafe that they had "some potential fixes" in the pipeline, so I guess we should wait and see what happens. I, for one, am greatly encouraged by this.
-
1
-
-
10 hours ago, Miller Thor said:
Last but not least, this monster project has gone off without a hitch and the increase in performance is noticeable if you are familiar with the data of the Statisticbar. Why the project manager April Linden, however, feels bad because of the failures is a mystery to me. The downtime was indisputably necessary to be able to carry out the project in a secure network environment and to ensure the security of all data. A huge praise and thanks to all involved in the project for this great achievement they have done in the last few days.
Miller, I think I slightly disagree with part of your response to the recent work and the above comments. April's comment reflected the fact that the failures were widespread and (some) had been going on for a considerable time. I would think THAT was the reason for her comment and to me at least, it was justifiable.
That the work was done, and done with considerable feedback in the forum from LL staff (Grumpity in particular) was commendable and I join with you in praising the staff.
Looking at my statistics panel on my homestead I have to say I see no great change, especially in the area of script performance (rezzing was not an issue before and is not now), but at a mainland sim I do security for, I see a substantial improvement in rezzing though no great change in script performance. I think the work has improved the performance on regions that had, in recent times, become highly stressed and unpleasant - Wirtz was a classic example. This may have been the aim and as such was successful. Perhaps on more lightly loaded regions, such as Woods of Heaven, my home, the improvement is more subtle.
So the result to my eyes is a bit of a "curate's egg", that is, good in parts. I also notice that the ten-day restart cycle is being assiduously followed now with resect to my homestead on Main Server at least. This had NOT been the case until the last two or three weeks.
One thing I think we agree on is that the communication by April was very comprehensive and for my part makes me feel a whole lot better about Linden Lab and their attitude to their customers (yes, customers, not simply users). I pay for my access and that makes me a customer.
-
Miland
Your best solution is to contact SL support and ask them to synchronize your inventory between Aditi (the Beta Grid)and Agni (the Main Grid). That, other than waiting, is your best option, I believe.
-
1
-
-
The issues you list there are all signs of either a modem/router-cache issue (turn it off fully for at least 2 minutes) or a poor internet connection (that's down to your ISP), not SL failures. I know of no-one else with such severe issues.
So while I sympathize, I suspect your problems begin at home not in SL.
One thing I have noticed for the past day or so, and that is spells of bad chatlag, usually lasting less than 30 seconds, but awkward to say the least. That I suspect is down to network issue, but since my partner and I (we have different ISPs but we are both in the UK) suffer it at different times, I don't *think* it's a server issue.
-
1
-
-
Just as a final word, THANKS to all the LL guys (a non-gender specific term, note) for the hard work and long hours catching all those pesky hamsters and getting their wheels running again!
I'd been wandering around thinking that I hadn't noticed much improvement when I realised that I'd set my draw distance over twice the normal while taking some photos, and things were rezzing really quite well (!) At my usual settings things are just great, so thankyou. Gliding for 30 mins and a dozen or so border crossings, loads of TPs today...NOT ONE disconnect!!
-
1
-
-
Thankyou Mazidox,
I got release notes via my viewer just after complaining so it was a total "mea culpa" for me.
-
Well, it seems my browser omitted to show me Maxidox's post of 9 hours ago! Sorry, I spoke way too soon! Weird though!
-
Greetings
I have a question. I see there were rolls to the main RC channels today and rolls to cake and snack are ongoing. However there is nothing in the forum regarding WHAT the new software actually is or does "internal changes"...logging improvements"? please can we be told? I saw Wirtz on LeTigre is on 19.05.01.526856, but there are no release notes.
-
Indeed I found Gliding a pleasure for the first time in a while.
Stand by your beds though, seconds out...round 2 just started!
-
1
-
-
Not at all Marianne, perfectly reasonable question!
By the bye I see they're at it again tonight, gluttons for punishment! Get the coffee tanker ready!
-
2 hours ago, Marianne McCann said:
Any news this week, or are these on hold due to other network... work?
To be fair to LL techs, I think they are due some R&R after pulling an all-nighter last night to get SL functional today! I'd be VERY surprised if there are any rolls this week, TBH I'd not be surprised if some of the "preventive" restarts were forgotten too!
-
'S looking pretty major. I can't ping the Tucson server that connects to SL so it may not be SL itself but their network of ISPs. I looked at online numbers just now and it is falling slowly so whatever it is is taking a while to fix. Looks like time t read a book or watch one of the crud DVDs I got recently!
-
Someone in LL needs to update this thread. LeTigre and Magnum RC channels did NOT receive the above mentioned update today, the roll was stopped after BlueSteel was completed (source LL support).
No reason was given and frankly I didn't expect one, but I DID expect the GSP and this thread to be corrected. I became suspicious when the roll was flagged as complete after one hour. It is far too short a time for all RC channels to be rolled and the region I run security for, Wirtz on LeTigre, had not rolled. It is now 4 hours since the roll was stopped.
-
2
-
-
4 hours ago, Qie Niangao said:
I just wish they'd tell us what they actually did.
If they're trying to fix a race condition by tweaking timing constants, it's doomed.
2 hours ago, MBeatrix said:Yes, Qie, I believe that's what they have been doing. But I do understand that properly changing the method would have serious implications, not to mention the time it would take.
Nevertheless, I'm amazed that after so many years they admit they don't know why sometimes it gets so bad. I can only guess that they never gave it the importance it has for many, many users.
All this has an much too familiar ring to it. I've been around SL in my alter ego since early 2008 and I was aware of it from about 2006. When Philip Rosedale and his team developed SL out of the LindenWorld project, it was then understood that the chassis on which SL was built was a complicated and pasted-together one, but those who had pasted it together were still around to "bang on things". In the years following Philip's departure SL passed through a phase of utter incompetence, during which time much knowledge and understanding of the nuts and bolts of SL was lost as experienced developers were "let go". I will not comment on that time, it is too ridiculous to describe adequately.
When sanity began to return to Linden Lab, the new developers were only able to patch and Band-Aid issues and new features as they were introduced, to the point now where fixing one leak causes the programme to spring another. I do not see any sign of this retrospective bodging growing into a real understanding of our virtual world.
The case of TPs and region-to-region crossings is a case in point. They are considered by many to be much the same, but the truth is that they have little in common above some basic processes. Region crossings require parameters such as vector, velocity and azimuth to be transferred from simulator-to-simulator and sometimes server-to-server. TPs do not require that data.
One final comment. The reason why we are not flooded with "me too" comments on such threads as this is that most users are too numbed by the lack of contact with the Lindens in past years to bother with complaining for a matter such as this. Many were also rebuffed by a few who believe that we should simply be grateful for what we have and not make trouble. It is only those of us that can "be bothered" that post our thoughts suppositions and occasional frustrations here. It does NOT mean that it isn't really happening!
-
1
-
3
-
-
You clearly read my posts but you seem to show little or no understanding of what I write. You display the same wilful misunderstanding of Sharie, Qie and Beatrix so I am led to the conclusion that while you DO understand them you are purposefully adopting a combative and opposing viewpoint. To some that might constitute trolling.
Y'all have a nice day now.
-
1
-
1
-
-
With regard to the apparent changes at Woods of Heaven:
1) The onset was sudden. It was more than a year ago but I cannot be more precise. The situation does not get significantly worse in the days following restarts, nor does it improve immediately after a restart.
2) The script number had not changed, the performance drop was after a restart. Nothing else had changed, and we were still on Main Server channel.
3) So far as I can tell sleep time tracked and still does track spare time. Script time has not noticeably increased since.
4) Script numbers have only increased slightly since then, as have script events.
5) I have not attempted to restore function by removing old scripts - since the change in performance of scripts was sudden with no other changes, I see no value in doing that.
and 6) There are no other significant performance reductions in the region. The few scripted objects that we have seem unaffected. There have been and remain significant changes to scripted event operations concomitant wit the deterioration in % script run, though these are not massive.
I raised the matter with Live Chat at the time it first occurred and I was told a restart would cure it. When it did not I returned to Live chat who offered no further remedies other than I should reduce my script number (which was not particularly high anyway. I did not nor could I do this.
Upon raising a JIRA, it was summarily closed with the comment that I should contact support (?). Frankly I could not see the value in doing that.
Now I reiterate that all this was gleaned from the in-viewer Statistics panel and like other residents I think it is perfectly valid to rely on this metric to describe performance changes. It was designed by LL developers for precisely this function and is the best tool for the purpose.
-
2
-
-
This is getting ridiculous. Sharie is complaining and yes, beligerently, since this problem is seriously impacting her regions. It is NOT the work of a "jaded Armchair critic" but of a determinedly loyal (to SL) resident and creator.
As far as "empirical measurement" is concerned, if we cannot rely on the tools specifically created by LL to measure SL performance, what hope do we have of convincing LL that something we create is any more reliable?
As for not being an Apologist, you have disagreed strenuously with every point that Sharie, myself or Beatrix has made recently, yet you have offered no true path to a potential proof or disproof of the issue. If that doesn't make you an apologist I don't know what would. I had enough carping criticism from Theresa to last me a lifetime. This issue exists, it is real and it is being ignored by LL systems people.
I am sick and tired of trying to convince folk like you, Coffee, Solar and Theresa, it is not possible. There are none so blind as WILL NOT see.
For the last time: I am NOT saying categorically that LL are dissembling about putting more regions on servers, I am saying that something WAS done and I would like to know WHAT IT WAS, because for some folk in some locations it has hamstrung their operations.
-
1
-
1
-
-
23 hours ago, CoffeeDujour said:
None of what you're saying is actionable.
The only numbers being thrown about are from the statistics floater, which could be wrong. A bug in stat reporting is a much simpler possibility than underhanded LL are doing the sneaky on the sneak because evil.
Instead of pages and pages of unsubstantiated moaning and speculation, you could just write a an open benchmark that tests multiple aspects of the scripting system.
Today my script did X things in Y seconds, yesterday it was much faster.
Here is the script. Here is the output. Here is the SLM page so others can try it. Here is the JIRA page to report your experiences.
Nope. This is NOT a valid response. If it was a glitch in the stats reporting (which in itself is highly unlikely) we would not see some regions with >95% script run. No one has accused LL of being underhand, just a bit less than transparent about a change.
My guess that someone WILL produce (or maybe has indeed already produced) such a script and if my experience is valid LL will simply say that it is not measuring parameters correctly.
As to the comment that our comments are "unsubstantiated moaning and speculation" I suggest you re read some of the previous posts in this thread.
Something HAS changed and really all most of us are asking is WHAT? @Sharie Criss has a very valid case to be answered.
-
1
-
-
Fingers crossed for the RC roll!
-
Well, let's hope it isn't something fundamental in the EEP code, 'cos it's coming back to the RC's on Wednesday, "with regressions fixed". I do hope this doesn't cause the Band-aid to fall off.
-
21 minutes ago, Qie Niangao said:
__________________
[* ETA: @Aishagain , you'll remember this one from @Torric Rodas's comment in this thread and from this jira but I realize it's slightly different from Vallone, in that the London sims had Spare time but Vallone's Sleep time was counted as Script time.]Yes Qie, and Torric got the same short shrift from LL as I did. They just do NOT want to know about this issue.
-
2
-
Facebook
Instagram
Twitter
YouTube
Flickr
No news of this week's rolls yet
in Second Life Server
Posted · Edited by Aishagain
Yes, well apart from showing myself up as an old fart
, it seems that LL's recent bout of open communication has exhausted them. Main Server and RC channel rolls are complete and not so much as a whisper on this thread. That's disappointing, especially since the server version rolled to Main on Tuesday was not what was rolled to LeTigre and BlueSteel last week and there was no news of any roll to Magnum last week. No release notes via the viewer either - it still shows the last roll to Main, two weeks ago.