Site Issues Notification Thread

I just got word from the hosting company that they did see the leap second issue on one of the servers, and they've apparently fixed it. I'll keep a close eye on the site for the next few days, but please let me know if you continue to see issues.

9 AM NY Time, July 4: it's taking a little less than a minute to get from this current page to the previous page and, prior, to get from my Settings to here. It used to be nearly instantaneous.

(Only one computer is online btw.)

distaff

I'm sending you a PM directly. This does not seem related to the leap second issue that the hosting company had found.

I think I have a plausable theory: But i cannot confirm or do much as my VPN crashed/hung so I am stuck with a slow link out of PR

@hfot2: Can you traceroute 75.126.163.232 (use the terminal window on the mac)

@nadreg: Can you traceroute (loose source route) from indiamike server to 66.50.9.9

PM me if you like, if my hunch is correct, there is an interface fritz problem...maybe !

Right now and last five minutes (first noticed 2.34am IST) --- site barely alive.

Yup. Same here. Didn't seem to last very long, the way we're now posting on it, but, whatever glitch is there it doesn't seem to be gone. Tho' we may of course now be getting sensitive to whatever temporary or not-so glitches may always and normally occur.

(The thread of course caused me to read up some on this leap second issue, that's not uninteresting. How was the dreaded Y2K bug largely averted, as far as I know? By everyone being so wise as to take adequate and timely measures? I can hardly believe that, I thought it mostly just blew over unnoticed.) --> Anyway maybe better to not get into a discussion of it here, as said it may only cause Nadreg to think Aargh there's another report on something amiss, he'll no doubt receive just a notification that there is a new post on this thread, but not seeing its contents until he looks into it. Must indeed be a botheration. The thought just naturally bubbled up.

Moreover, I'm not sure but from what I read the last leap second before it can't have been inserted that long ago, i.e., before the web got as active as it is now (or was it?) It is new to me and I've never heard of it crashing the internet or general computing before, but then it is quite possible I was busy picking my nose at the time. Maybe we could start a thread on it in the Chat or Off-Topic forums or so.

Well I'm glad to learn it's not just our connection that's slow.

distaff

Funny but I can't get to IM directly it just times out. I'm currently connected using a web proxy. It started about 4:30 pm PST yesterday. Happened when I started collecting stats using Ping Plotter, oh noes I couldn't be satisfied with trace route. Did I get my IP blocked? If there was a sudden spike in ICMP traffic from all the trace routes folks were doing and I hit it a little harder than most the heuristics in a intrusion protection system could have ganked my IP. Mebbe the internets just doesn't like me.

From what I've been reading a thing with this leap second could be that systems start to call for operations that are "inactive" to become "awake" (now with this one second's lag), so in effect freezing up the whole lot.

It is decidedly over my head though. We're also a few days ahead again now; I wonder if most servers wouldn't have fixed it, and so we're looking at more regular interference now.

The hosting company is big enough to sustain your ICMPs, jumbo packets and then some. Ping Plotter tells you nothing.

Guys, while most of this means nothing to me, it seems to mostly pertain to Linux kernels etc. and other systems derived from it, sustaining much of the web, right. (I guess Java was heavily affected?)

It will have little to do with your personal home set-up.

Correlation doesn't equal causation but I haven't been able to access IM since I messed around with P.P. yesterday.

Max MTU on the internet is 1500, Jumbo frames run up to 9000.

You must be using windoze

A majority of users would.

(As in you may not consider it "cool," but it is kind of irrelevant. You'll just want to accommodate them.)

C:\Users\gskell>ping -l 1501 128.95.221.100

Pinging 128.95.221.100 with 1501 bytes of data:
Request timed out.

Windoze has a neat way of auto sizing packets if there is a restriction. But I can't ping nothing with a MTU of 1501.I'll just put in a work order to have all the routers on the internet allow jumbo frames.

ETA:
Reloaded my DSL modem and got a new DCHP address it changed from IP 71.X.X.50 to 71.X.X.38 got me back in with my native IP. Teh Internets like whateva... I still think I shot myself in the packet header using the High Altitude Pea Shooter AKA ping plotter. Them Data Centers in Dallas sure ain't forgotten about the Alamo.