Best Linux Server uptime?

[Fallen Kell]

# uptime
1:51pm up 1043 day(s), 5:07, 4 users, load average: 0.11, 0.07, 0.08

First one I looked at. I am sure we have plenty more that are longer.

Edit there we go :

# uptime
1:52pm up 1589 day(s), 1:28, 1 user, load average: 0.05, 0.03, 0.03

 

[Fayd]

Just wondering what the best server Uptime youve ever had with a Linux box (or really any server if you want to include other numbers as well). I have a Minecraft server thats running up on almost 90 days (its at 87 days and change right now) with no issues. I was thinking if my little crappy Minecraft box can pull off 3 months and running, i'm sure somebody with a linux box in production somewhere on nice hardware must have some awesome uptimes. Let's hear em!

Mine:

Cent OS 5.4 - Primary Function: Public Minecraft server for me and some friends. Uptime 87 days 16 hours 25 minutes.

3 months?

the first time i seriously booted into vista for use, it stayed up that long. >.>

 

[Paperlantern]

3 months?

the first time i seriously booted into vista for use, it stayed up that long. >.>

Wow that must have been one seriously underpatched box by the end of its run there.

I for one thought around 100 days (which was about what it hit before a power failure finally took it back to 0... its running at 6 days now) was pretty good for a server made out of a desktop having no UPS protection at all. I think you misinterpreted my posting as bragging. I simply thought if a measly little desktop server could hit 100 days, then there may be some VERY impressive numbers in actual production, and decided to ask what people were attaining in production environments. Someone posted 1500+ days. I do think that is amazing myself. Or did you have a Windows ME machine that did that the first time you used it?

 

[Evadman]

I had a Windows ME box that was acting as a Tribes game server up for just over 6 months way back when. I had to power it down to put it on a UPS; go figure.

 

[Crusty]

I had a Windows ME box that was acting as a Tribes game server up for just over 6 months way back when. I had to power it down to put it on a UPS; go figure.

Have you contacted Guinness yet? That has to be a record of some kind.

 

[PCTC2]

The longest I had was probably around 400 days, with the average around 180 days.

Right now, my longest uptime is
04:46:24 up 130 days, 12:54, 1 user, load average: 0.00, 0.00, 0.00
due to maintenance and hardware issues. Ubuntu 9.04

My main production server is at
21:45:35 up 52 days, 3:18, 2 users, load average: 3.58, 3.12, 2.94
because of a physical move. Before that, it was at 220 days. CentOS 5.5

 

[bobdole369]

Fallen has me beat:

19:34:23 up 823 days, 07:24, 5 users, load average: 0.32, 0.21, 0.45

CentOS 5.3 - dedicated in a datacenter in Chicago - email, spam filter, corporate presence website.

There hasn't been a need to bring it down in that time - the machine runs by itself inside a hypervisor that manages patches. Its only kernel patches that would require a reboot, and there hasn't been a reason for that in nearly 3 years.

 

[whiteonline]

None of your boxes show any real load. Here is one for y'all.

xxxxxx@xxxxxx # uname -a
SunOS xxxxxx 5.8 Generic_117350-36 sun4u sparc SUNW,Sun-Fire-480R
xxxxxx@xxxxxx # uptime
8:57am up 1816 day(s), 13:13, 3 users, load average: 19.48, 21.66, 21.86

Server is on a very secure network, so patching is not as much of a concern.

 

[you2]

My freebsd box ran for around 5 years without a reboot (I think it was 4.5). Normally reboots are due to kernel patches for my current system. The field machines we have are typically up for 200+ days; again kernel patches are the primary events that cause a reboot.

 

[OVerLoRDI]

Longest I have currently is 45 days, which for a bitcoin miner is pretty good. Usually the cards need to be downclocked, get too hot, or need some kind of attention, but this machine has been the only machine I setup, hit go and never had an issue with.

 

[kamikazekyle]

1.25 load? What are you running to slam it ?

Heh, I managed some SGI boxes that have hit loads in the 40's to 60's with 8 processors before. Granted, a one of those octo-core box has less processing power than my modern i5 at home, but still

The longest uptime I've ever had was on a NetApp I used to manage at over 490 days, I think. The only thing that really brought it down was a ESH failure that caused a loss of several RAID groups, which then panic'ed the filer and rebooted it. Nice little story about how nobody listened to my doomsaying that goes with that.

 

[xSauronx]

I had a Windows ME box that was acting as a Tribes game server up for just over 6 months way back when. I had to power it down to put it on a UPS; go figure.

the hospital i work at has a few windows 2000 boxes that are 2 years+, but they run almost nothing. i think they act as a gateway or forwarder for some fetal monitoring systems

 

[Fayd]

Wow that must have been one seriously underpatched box by the end of its run there.

I for one thought around 100 days (which was about what it hit before a power failure finally took it back to 0... its running at 6 days now) was pretty good for a server made out of a desktop having no UPS protection at all. I think you misinterpreted my posting as bragging. I simply thought if a measly little desktop server could hit 100 days, then there may be some VERY impressive numbers in actual production, and decided to ask what people were attaining in production environments. Someone posted 1500+ days. I do think that is amazing myself. Or did you have a Windows ME machine that did that the first time you used it?

every computer i have is always on an ups, and vista x64 with stable drivers was pretty stable, even from the get-go.

while i power down and install updates on an as-needed basis now, i wasn't quite so religious about it then. but it had no ill effect. I've as of yet never had a security breach on any of my systems.

ups is required here, as the power here is pretty bad, dropping below 100v fairly often throughout the day as the AC of the neighbor houses kick on. or at least it was at the time. nowadays it's a fairly constant 114v.

 

[silverpig]

My router is now at 375 days or so...

 

[TwiceOver]

My router is currently at 924 days. Last time I took it down was to put it on battery backup. But that is being retired in about 30 days so...

 

[Knavish]

Back 10 years ago I had an AMD K6-II machine in the 400 MHz range running as a http and email server for > 450 days. (Mandrake Linux)

I don't know if I'd do that these days -- some security patch would probably require a reboot within a year.

 

[Knavish]

I had a Windows ME box that was acting as a Tribes game server up for just over 6 months way back when. I had to power it down to put it on a UPS; go figure.

You win.

 

[Red Squirrel]

I had hit 1 year on my hosted web/mail/dns/UO/etc server, then an app I deployed to it had a memory leak and used up all the ram, causing it to crash as the last available memory had been taken over the course of a week or so.

Fixed the memory leak and it ran for a good 500 days or so after that. Found a better deal so switched my stuff over to another server which is the only reason it did not stay up longer. I'm sitting at 155 days now on the new server.

My home servers tend to be up less long as I'm always screwing around with stuff like the electrical panel adding new circuits and stuff and I'm too chicken to do it live. UPSes only go so far.

Every single time that I've had a Linux server go down was as a result of something I did. I can't say the same for Windows.

 

[Fallen Kell]

Unfortunately I don't think any of our high load systems have an uptime past 450 days at the moment. Our main data center had its power and cooling upgraded then, at which time only the absolute essential systems were all shutdown. However, during that upgrade, the water cooling system blew a pipe and started flooding the datacenter floor room, which cause the UPS and backup generators to go into fault protection mode, and since that was what was powering the few systems we had kept online, they crashed hard. Our other datacenter had an issue when while the UPS battery system was being taken down to replace old batteries, someone thought it would be a good time to run a generator test (since everyone was already there), not realizing that to test the generator, you need to put it under load, which cuts the street power and bypasses through the UPS battery system, and since the batteries were offline.... The systems I have left are ones in a few location based data closets, or really small server rooms. Most of these are now simply DNS/NIS/LDAP/Bootp systems which won't show a lot of load.

Before the main data center went down, I know that we had 4 or 5 systems which had been up since 1997.

 

[lord_emperor]

15:25:03 up 97 days, 21:46, 1 user, load average: 0.01, 0.04, 0.00

It's just my personal server, no UPS so this number mostly keeps track of the last power outage at my place.

 

[ForumMaster]

i've seen red hat servers here with uptime of upwards of 1200 days. there was one with close to 2000.

they finally were all reset when there was a controlled data center power upgrade.

 

[VinDSL]

Just wondering what the best server Uptime youve ever had with a Linux box (or really any server if you want to include other numbers as well).

Here's my production server (@ JaguarPC):

Current Time: Sunday, 07-Aug-2011 08:17:51 EDT
Restart Time: Saturday, 03-Jul-2010 21:20:50 EDT
Parent Server Generation: 5217
[B][COLOR="Red"]Server uptime: 399 days 10 hours 57 minutes[/COLOR][/B]
Total accesses: 17039586 - Total Traffic: 37.8 GB
CPU Usage: u17.28 s3.49 cu535.72 cs0 - .00161% 
CPU load .494 requests/sec - 1175 B/second - 2381 B/request
7 requests currently being processed, 8 idle workers

 

[Soulkeeper]

613 day uptime on my original Eee 4G netbook
this easily beats my previous uptime record on an athlonXP 2000+ of around 500 days I believe (power outtage got it)
the laptop battery has saved the Eee atleast twice now from small power outtages

I'm gonna let it go till it fries !
uptime is now it's ONLY purpose
wish I had upgraded the kernel to support turning the lcd backlight off, but ohh well

 

[Red Squirrel]

Wow that's pretty good for a netbook! Being a "workstation" it's much more vulnerable to being rebooted given it's being used directly.

I don't think I've ever gone more than a week with any of my workstations just because I'll usually end up having to reboot for a reason or the other. Updates, or whatever.

 

[Paperlantern]

Wow that's pretty good for a netbook! Being a "workstation" it's much more vulnerable to being rebooted given it's being used directly.

I don't think I've ever gone more than a week with any of my workstations just because I'll usually end up having to reboot for a reason or the other. Updates, or whatever.

Yeah, my workstation at the office is a Win7 machine that I never turn off unless for updates, I think its at about 3 days uptime right now, typically it never gets over about 15 days before an update reboots it.

My production minecraft server though, I've since migrated it to a virtual machine where the host has a UPS and it currently is sitting:

08:46:29 up 54 days, 13:28, 1 user, load average: 0.00, 0.00, 0.00

It hasnt been used in probably 3 weeks of that time though i'd say. I have a non production test minecraft server that the guys and I are currently using to try out some mods before we add them to the actual server. We have been on the test box for, well lets see...

08:50:57 up 13 days, 19:08, 3 users, load average: 0.00, 0.03, 0.00

Although I think ive rebooted it during the testing period for one reason or another anyway. Test box is my old production server that i talked about in my original post, so it may have been caught by a power failure too, who knows.