Fight the Lag! - with metrics

[drunkendwarf]

Why are people posting traces for

Arm doesn't own their own box so all infrastructure involved is not fixable by anybody reading this forum

In the past, when lag issues surfaced, it has been requested by the staff. While it may not be fixable by anyone reading this forum, there are staffers on this forum that can take the information to people that can fix it...and the more information you present those people with, the better.

Presenting these people with full traces from 10+ different origins certainly can't hurt, and may be infinitely more useful than "Shit's runnin' slow."

Hosting company should be sorting that out because it's their job. Back when Ginka was a PC in somebody's house that made more sense.

?
Why do you give a shit? Staff asked us to. We do it. I realize this time it's looking like an issue on the box itself, but in the past that has not been the case.

[Case]

Why are people posting traces for

Arm doesn't own their own box so all infrastructure involved is not fixable by anybody reading this forum

In the past, when lag issues surfaced, it has been requested by the staff. While it may not be fixable by anyone reading this forum, there are staffers on this forum that can take the information to people that can fix it...and the more information you present those people with, the better.

Presenting these people with full traces from 10+ different origins certainly can't hurt, and may be infinitely more useful than "Shit's runnin' slow."

Hosting company should be sorting that out because it's their job. Back when Ginka was a PC in somebody's house that made more sense.

?
Why do you give a shit? Staff asked us to. We do it. I realize this time it's looking like an issue on the box itself, but in the past that has not been the case.

Have fun wasting time I guess? Not sure why I'm a bad person for questioning something.

[Harmless]

heh, yeah, I could have told ya'll your tracerts were meaningless.

ping armageddon.org

in a cmd window and see tons of beautiful pings.

and the fact that the website loaded meant this was not a network problem.

[RogueGunslinger]

I said as much a while ago. All tracerts are going to tell is that the problem is not on our end. Which is pretty obvious by now.

[Nehoc]

^ +1 on ping, it's great, a for tracert being meaningless, not for me to decide, lag is gone now for me anyways

[Nehoc]

hello dawn, hello heavy lag

i recall my first heavy lag of the night came at around dawn too, maybe increased NPC walking around and shop opening activity, couple with increased player activity?

[Necro]

playtime interspersed with heavy lag .. hard to pin down, but when pinged in a normal state above 50ms to return, 200 ms during spikes. not going to be able to play long enough to wait for it to happen again and tracert, hope this helps

[X-D]

Is it safe to assume that we'll be logged out eventually? I really can't wait any longer and I'm fairly certain my PC would die if I just continued to be where I happen to be.

Actually staff does not log people out...unless asked, usually through wish.

Best you can hope for is that people are mature enough to not take advantage of LD PCs...Or, if you are in an unsafe area, maybe somebody will move you to a safe area...I've already done so with a couple PCs I found in the wilderness...I'm just an old softy that way.

[musashi]

Is it safe to assume that we'll be logged out eventually? I really can't wait any longer and I'm fairly certain my PC would die if I just continued to be where I happen to be.

Actually staff does not log people out...unless asked, usually through wish.

Best you can hope for is that people are mature enough to not take advantage of LD PCs...Or, if you are in an unsafe area, maybe somebody will move you to a safe area...I've already done so with a couple PCs I found in the wilderness...I'm just an old softy that way.

It's true. Once when I went link dead due to power outage X-D sapped subdued and drug my PC back to safety, and put my weapons in my inventory.

I pay it forward now when I find PC's in similar circumstances.

[tiptoe]

Is it safe to assume that we'll be logged out eventually? I really can't wait any longer and I'm fairly certain my PC would die if I just continued to be where I happen to be.

Actually staff does not log people out...unless asked, usually through wish.

Best you can hope for is that people are mature enough to not take advantage of LD PCs...Or, if you are in an unsafe area, maybe somebody will move you to a safe area...I've already done so with a couple PCs I found in the wilderness...I'm just an old softy that way.

It's true. Once when I went link dead due to power outage X-D sapped subdued and drug my PC back to safety, and put my weapons in my inventory.

I pay it forward now when I find PC's in similar circumstances.

Thankfully bad situations were avoided this time around (for me, at least). But it does seem like there should be some sort of protocol in situations like this...game-wide logout with a reboot, just to purge players who couldn't wait 3 hours to find somewhere safe to quit. Granted, I don't think it's usually quite as bad as it was.

Personally, I handle it the same way and will either drag someone to a safe location or just wish up that they're link dead. Good to know I'm not the only one

[Karieith]

Staff might be just as laggy and just as unable to issue commands with wish ups during global lag like the type we're facing now.

But I highly approve of sapping/subduing/wishing up/being nice to linkdead characters when it's not huge global lag.

[tiptoe]

Staff might be just as laggy and just as unable to issue commands with wish ups during global lag like the type we're facing now.

But I highly approve of sapping/subduing/wishing up/being nice to linkdead characters when it's not huge global lag.

Oh yes, of course! I mean more afterwards...when all the lag finally subsides.

[Lutagar]

IT BEGINS

[Barsook]

Can a ping help?

PING 206.72.207.140 (206.72.207.140) 56(84) bytes of data.
64 bytes from 206.72.207.140: icmp_req=1 ttl=53 time=34.0 ms
64 bytes from 206.72.207.140: icmp_req=2 ttl=53 time=36.1 ms
64 bytes from 206.72.207.140: icmp_req=3 ttl=53 time=34.4 ms
64 bytes from 206.72.207.140: icmp_req=4 ttl=53 time=35.0 ms
64 bytes from 206.72.207.140: icmp_req=5 ttl=53 time=34.0 ms

[drunkendwarf]

Yup. Connects in half a second, and then I get no mantis head for many minutes.

[drunkendwarf]

Can a ping help?

PING 206.72.207.140 (206.72.207.140) 56(84) bytes of data.
64 bytes from 206.72.207.140: icmp_req=1 ttl=53 time=34.0 ms
64 bytes from 206.72.207.140: icmp_req=2 ttl=53 time=36.1 ms
64 bytes from 206.72.207.140: icmp_req=3 ttl=53 time=34.4 ms
64 bytes from 206.72.207.140: icmp_req=4 ttl=53 time=35.0 ms
64 bytes from 206.72.207.140: icmp_req=5 ttl=53 time=34.0 ms

Ping jis like traceroute lite - ust measures how long it takes to bounce from your machine to the server. In this case, having established it's not a network issue, ping isn't of much use.

[Barsook]

Yup. Connects in half a second, and then I get no mantis head for many minutes.

Same and here's the traceroute:

traceroute to 206.72.207.140 (206.72.207.140), 30 hops max, 60 byte packets
1  unknown (192.168.1.1)  1.694 ms  2.760 ms  3.890 ms
2  10.169.64.1 (10.169.64.1)  12.145 ms  12.732 ms  12.717 ms
3  ae0-1030.cr02.dytnoh55.swo.rr.com (184.59.244.114)  19.525 ms  23.147 ms  23.187 ms
4  ae0.tr00.clevohek.mwrtn.rr.com (65.189.140.144)  30.953 ms ae1.tr00.clevohek.mwrtn.rr.com (65.189.140.130)  31.378 ms  30.856 ms
5  107.14.19.58 (107.14.19.58)  40.769 ms ae-3-0.cr0.dca20.tbone.rr.com (66.109.6.70)  40.820 ms 107.14.19.58 (107.14.19.58)  50.857 ms
6  ae-1-0.pr0.dca10.tbone.rr.com (66.109.6.165)  43.154 ms  31.789 ms 107.14.19.135 (107.14.19.135)  28.597 ms
7  3-3.r1.dc.hwng.net (69.16.190.113)  29.236 ms  30.281 ms  44.192 ms
8  e4-2.r1.ny.hwng.net (209.197.0.42)  46.798 ms e1-4.r1.ny.hwng.net (209.197.0.34)  50.814 ms e4-2.r1.ny.hwng.net (209.197.0.42)  38.827 ms
9  209.197.17.198 (209.197.17.198)  74.745 ms  74.655 ms  75.170 ms
10  edge-12-teb1.us.as19318.net (64.20.32.78)  37.973 ms  55.270 ms  55.249 ms
11  openvz17.interserver.net (206.72.207.130)  41.508 ms  58.379 ms  41.537 ms
12  206.72.207.140 (206.72.207.140)  35.124 ms !X  51.989 ms !X  35.112 ms !X

[armandhammer]

I admire how many players tough out the lag in order to play.

[Nyr]

We had similar issues 6 months ago and they moved us to a different, newer, faster server with more RAM. 

A few weeks ago the issues came back after we had a strange outage where the server was stuck then just came back up with less RAM than we had, but still more than we are paying for. 

They have responded to the ticket by increasing our I/O priority and increasing our 'lockpages' due to an error they were seeing in the logs.  I still need to get back to them and tell them it hasn't helped.  (Sorry, it's Father's Day weekend, and I'm not able to concentrate on it)

Just reposting this for anyone that has skimmed over it.

[Gunnerblaster]

P.S. - It's working now.



(Hope I didn't just  jinx it)

[RogueGunslinger]

It comes and goes.

[Karieith]

Please behave yourself game, now is not a good time!

Y'know, it got this terrible around this time yesterday too.

[Kismetic]

It's coming.

[Namino]

Here we go.

[BadSkeelz]

Awww, it missed me :3