The Planet servers down/

we got 2 unmeterd and 2 total control boxs, all working fine, check the logs, no any down time record, only 1 performed a reboot yesterday

First thought: they've deliberately disabled data logging. My servers are back up but the graphs are still 'stuck' on that current reading

99.5% uptime - about 3 1/2 hours downtime per month
99.9% uptime - about 45 mins downtime per month (this outage will exceed that)
100% uptime - well...

We are talking thousands of servers, a couple of hundred employees.
They are huuuuge. http://www.theplanet.com/datacenter.html

I have a server at Servermatrix the discount part of theplanet and this is the first serious downtime I´ve had in 15 months.
Combined I think I have 30 minutes max downtime until now

Ain't it a kick in the balls to wake up to this shit i'm down right now Time to get some backup servers in at least 3 different places.

install MRTG

Agreed, I've been on servermatrix for around that period and the only other downtime I can remember was about 15 mins long.

This must be a serious fuckup by someone or something. Their redundancy has obviously not worked.

Both my servermatrix servers were unreachable for about 71 minutes.

looking at it now, 99.5% only seems to apply to xtreme unmetered
all others are 99.9%(going off google cache)
im sure they'll be all set at 80% soon, which might be why the site is currently down and their network is back online

My point was more that it MAY have been a deliberate decision to disable data logging during the downtime period. (Why would you need to otherwise?) The servers are back up but they've forgotten to re-enable it. I could be wrong, of course.

For utterly unreliable, huge downtime, incompetant service, visit

www.reliablehosting.com

By far the worst hosting company I have ever dealt with.

Combine all the hosting companies like Webair, Jupiter, NatNet, Isprime and it won't match ThePlanet's size

oh, im sure it is. i just think its interesting seeing my realtime stats. a few drops to zero and about 1/2 the normal and dropping. i wonder if they freeze your threwput at 1mpbs if it'll log that as bw used towards your limit. wouldnt that be a bitch

was there an explaination as to what happened?

It happens when there is a little down time also happened for me when I had no down time might be their system that is a little crazy.
If you need the extra GBs badly write a ticket about it and I am sure they will add a couple extra GB this month.

Not yet but I heard to would post an explanation here forums.servermatrix.com but nothing yet.

At approx. 4:15 AM CST, a pair of redundant Powerware 500KVA UPS units failed - creating a power failure in Section B of our D2 datacenter. Emergency teams were deployed within minutes and power was restored, however intermittent power outages have continued to occur until 6:45 AM CST. Powerware, JT Packard, and electricians are currently onsite with over 100 Planet technicians working to resolve the issue. We do NOT anticipate any futher outages. A formal RFO will be released once the team debriefs. We apologize for all the issues this has caused you.

We will be performing a sweep of the data center as soon as this issue is completely resolved to reboot and ensure connectivity of all Active servers.

Thank you for your patience in this matter.

I think they're just trying to pull the wool over our eyes. None of my 3 servers have been restarted, and they were completely (not intermittently) unreachable for about 71 minutes.

82.36.139.111 - - [31/Mar/2005:20:13:50 +1000] "GET /showgal.php?srcdom=&link=new&th=tn1585.jpg HTTP/1.1" 302 5 http://www.****/tgp.html 28275 0
84.216.55.32 - - [31/Mar/2005:21:23:27 +1000] "GET /tgp.html HTTP/1.1" 200 24096 - 113 1

These are two sequential lines from my raw server logs clearly showing the gap.

only 71 minutes ?

mine was down for about 5 hours

Are you sure it's not just the RTG graphs getting stuck at 0? See my earlier post about this...

Did your server(s) get rebooted?

i am 100% sure and my server got rebooted twice. however its up again, but very very slow

umm umm power issues that's rough

from 4am to almost 9am one of my servers flatlined
ugh!!!

See my screenshot above. My server was down for just over an hour, not 5 as the graph seems to suggest.

BTW, they've taken https://orbit.theplanet.com/ (the support site) off the air.

good info
thanks
also to Keyzer

Oh Please.

The Planet's graphs are public. The adult hosts' graphs aren't. ThePlanet is doing approx 13.5 Gbits aggregate. If you really think NatNet and ISPrime (forgettign Webair and Jupiter) aren't doing >20 gbits combined, you're deluding yourself. I don't know how big Webair and Jupiter is, but if I had to guess, the aggregate total of all 4 (Webair, Jupiter, NatNet, ISPrime) is 30-40Gbits.

Also, 13.5/4 = 3.375 Gbits. Whatever you "think" the hosts are doing, if you think that they're smaller than 3.375 gbits you're on crack. For example, NatNet hosts ConsumptionJunction. That alone is probably >3gbits.

Further:
http://www.isprime.com/index.html
Read the top article. ISPrime has recently upgraded their network to have 35gbits of capacity (and you may or may not see a new announcement when their upgrades are complete and bring them to 45gbits of capacity fairly soon), Use your best judgement to figure out why that would be. Using >50% of your network capacity is risking outages, using <25% is wasting money. The answer lies somewhere between those lines.

(that being said, you can look at these graphs and see that theplanet is running >50% on many (most?) of their uplinks)
http://www.theplanet.com/datacenter.html

About halfway down they have "Network Infrastructure"

As a fun example:
http://www.theplanet.com/image.html?...rovider=Level3

927Mbits at peak on a Gig-E (which in real-world usage maxes out at about 950-960Mbits).

On top of all of that, *WHO CARES* who is bigger? Do you really think that it's better to be with a "huge" host that is down, than a small host that is up?

Also, ThePlanet's main business is in low end hosting, many of the adult hosts have plans that *start* at >$100/mo. Why be surprised when a low-end host has problems?