Post your best uptime

A

archivalbackup

Gawd
Joined
Oct 12, 2007
Messages
643
I thought it would be fun as a tangent to the networking gallery thread to see some extreme uptimes for devices under your control.

Here are a few of mine:

Storage Network Switch: (Cisco 3750G)
862688107_R3SdF-L.jpg

This is the master, other switches have been added to the stack over time w/ no downtime.

SAN: (Equallogic iSCSI)
862688109_6tqm2-XL.jpg


Sadly, I had to take the SAN down for a software upgrade (and to add a new node), so this is my current uptime.
862693053_Ph7HD-XL.jpg

I am going to have to take it down again soon for another software update in the near future.

I have a few more good ones to dig up. 3 years on a switch seems pretty good to me :D Can anyone break the 5 year mark?
 
I have Cisco gear that has been up for 3+ years but before 4-5 years I end up upgrading the IOS or replacing the router. My record....back when things didn't change so much.... is a Novell NetWare 3.11 server that was up just past 5 years without a reboot.
 
Yeah we had some Netware stuff that was over 5 years of uptime. We were really scared to turn it off when we had to move them. Sorry no screenies. :(
 
A few more:

Home Wireless AP (last shutdown was the last major power loss that went longer than my UPS battery)
862706898_5LAra-L.jpg


Home Core Switch:
862710619_rFtNR-L.jpg
 
I know I've had Windows Servers run for over a year, I believe just short of 2 years was the highest I've seen. That was for a tiny client, server didn't do much, never got on the internet. Most of my clients though, I do maintain Windows Updates and reboots of the servers just for best practice.

As for routers, at home I could go as long as I wanted..my problem is, I often try to new routers and new *nix distros...so due to my dorking around, ..well..that kills uptime.

I do have one client, my largest one, network of around 70 workstations/lappys, and 6x servers....they moved locations 2.5 years ago, since the move..and building my server room there, they've had HP ProCurve series gigabit switches which I have not powercycled since the install. So those switches are going on 2.5 years..and counting. I know I have other clients with switches that I've not cycled in a long long time..even lower grade Linksys/Cisco SRW switches, etc.

Once you go with business grade hardware...uptime should get long.
5 years for a server though...that's quite notable. :cool:
 
Valnar said:
is a Novell NetWare 3.11 server that was up just past 5 years without a reboot.
Click to expand...

We had an old novell box at a past job whos uptime was about 6 years, they decide to reboot for the hell of it. The box never came back up lol
 
k1pp3r said:
We had an old novell box at a past job whos uptime was about 6 years, they decide to reboot for the hell of it. The box never came back up lol
Click to expand...

That's exactly why we were scared to shutdown and move ours. After 6 years its tough to even find parts if it won't come back up.
 
Yeah I've been through that before...the client that I mentioned in the above post, with the new office...2.5 years ago. When I moved them, naturally I had to shut down their servers at the old office. The servers had been rebooted..but most of them had never been fully powered off. Their primary healthcare database server, a Dell PE 1600SC or something..big honking dual Xeon box with RAID 1 'n RAID 10...or maybe data was RAID 5...I forget...anyways...I go to fire it back up at the new location..and freaking disk failure on the data volume.

2..yeah...2 of the drives decided to misbehave. :eek: Managed to massage one back online in the RAID controller..and the data volume snapped back to life. Other one didn't want to come back at all. Replaced them both a day later...and that server ran fine til we retired her last fall moving that box into ESXi world.

Any physical hiccups that may happen...usually surface when you power down and things cool off. That expansion/contraction thing.
 
YeOldeStonecat said:
Any physical hiccups that may happen...usually surface when you power down and things cool off. That expansion/contraction thing.
Click to expand...

Very true. We have a bunch of the last batch of Gateways that went out before they declared bankruptcy and sold to MPC (professional side) and every time we have windows updates that reboot the machines, a few die. Poor hardware right there.

The thermal stress of changing temps can really kill machines quickly; this is why I get upset about our backup A/C in our server room that can't maintain temps. Even if we stay under 85 degrees, the thermal shock of going from 65 to 85 in 20 minutes is a killer on hardware.
 
My tomato 1.27 WRT54GL:
uptimem.png

Last reboot was to install a UPS for the network equipment.

There was a time when I had that little router up for well over 100 days. Only reason I ever have to reboot is for firmware updates.
 
-bash-3.1$ uptime
14:32:09 up 954 days, 22:16, 1 user, load average: 0.02, 0.03, 0.00


Linux server at work - it doesn't do a whole lot, but it's always there when we need it!
 
My Server 2008 playbox has been up for exactly 1 year, 11 months today. Unfortunately we have been rewiring the office lately so our firewall and router got cut down in their prime :p
 
my best uptime on my 2821 is what i have currently stupid electric company is less reliable than anything i can think of LOL. my server on its battery backup has seen like 2 years uptime no reboots before, but then i upgraded so that went out the window
R2821-Edge uptime is 2 weeks, 19 hours, 41 minutes
 
our print server......6375hrs ~ 265 days..

I usually keep screenies of 'the best' I come across..
have one with an XP machine that has 1010 profiles in Documents and Settings...
have another with 65k + files in C:\Windows\Temp...

2003 Print Server
 
Well, can't really compare with what you guys have got there. I usually don't get past a year or so before I need to do a major software upgrade or some other maintenance.

This is the best I can find that I can get at remotely:
Code: 
3Com SuperStack 3 Switch 3226

System Name        : fes0
Location           : Upstairs Wiring Closet

Time since reset   : 382 Days 8 Hours 8 Minutes 52 Seconds

Not on battery backup though, so really not bad :p
 
We had 2 cisco 3550's at the office that were up for almost 2.5 years before we took them down for a re-wiring job
 
Bleh, good for bragging rights but thats about it.

When I see long uptimes, I see unpatched security risks and old UPS batteries.

Here's mine at home :D

Code: 
ATL1107-EDGE-RTR#sh ver
Cisco IOS Software, 1841 Software (C1841-ADVIPSERVICESK9-M), Version 12.4(25c), RELEASE SOFTWARE (fc2)
Technical Support: http://www.cisco.com/techsupport
Copyright (c) 1986-2010 by Cisco Systems, Inc.
Compiled Thu 11-Feb-10 22:52 by prod_rel_team

ATL1107-EDGE-RTR uptime is 1 week, 5 days, 4 hours, 7 minutes
 
Vito_Corleone said:
I've posted this here before:

uptime.jpg


It's of a 2948 before it was decommed.
Click to expand...

Just wow. Bout time for it to be taken down eh?
I update/upgrade my equipment as well as all my clients equipment often, it always hurts to reset that uptime count. Moderate uptimes IMO usually mean that the admin cares enough to take stuff down for updates and maintenance. My record is 1.5 years on an xp box being used as a file server, which I reset recently to update.
 
Lightworker said:
Bleh, good for bragging rights but thats about it.

When I see long uptimes, I see unpatched security risks and old UPS batteries.

Here's mine at home :D
Click to expand...

UPS batteries can be hot-swapped, devices have redundant power supplies, and high end switching & routing platforms support In Service Software Upgrades

When devices are physically isolated, protected by other means, or not using services that have patches it is not a big issue as long as you have good change control. IMHO you don't patch things just because you can; you research the patch and see what it contains, if you actually need it, and test it in a lab environment. It depends on the device of course, an end user PC is much different than a core switch.
 
^I agree with almost all of that, with the exception of UPS batteries, not all are hot swappable :)

But anyway, there's a lot of equipment out in the field that doesn't have redundant everything--it's not optimal, but it's a fact of life. At least Cisco is offering twin psu's and stackable power on their new X series switches, finally bringing power redundancy down to affordable levels for non-chassis access switches...that will help a bit
 
Lightworker said:
At least Cisco is offering twin psu's and stackable power on their new X series switches, finally bringing power redundancy down to affordable levels for non-chassis access switches...that will help a bit
Click to expand...

Ok not all UPS are hot-swap, but it would be an advisable feature to make sure you are getting.

Cisco has had redundant power options on switches for a long time, they just required an extra piece of hardware. At least back through the 2950 and Fast Hub 400 era they have a 20-something plug to connect to an RPS unit.
http://www.cisco.com/en/US/products/hw/routers/ps2883/products_data_sheet09186a0080091b4d.html

http://www.cisco.com/en/US/products/ps7130/index.html
 
Yeah I have a few of those RPS units lying around here, we don't use them at all. They don't provide for any sort of graceful switchover back to the primary power supply. To go back from the DC to AC, it requires a switch reload. Plus, trying to use those with high density PoE switches is a total crap shoot :)
 
Lightworker said:
Yeah I have a few of those RPS units lying around here, we don't use them at all. They don't provide for any sort of graceful switchover back to the primary power supply. To go back from the DC to AC, it requires a switch reload. Plus, trying to use those with high density PoE switches is a total crap shoot :)
Click to expand...

Depends on the Switch-Model. The newer ones (3750&3750E's) have graceful switchover with the RPS 2300.

Man I love 3750's =); got any RPS units to sell? I have been wanting one to play around with.
 
Unless the docs are wrong, the graceful switchover is only on E-series switches, 3750E and 3560E's. Anyways,I've got an RPS 2300 with a 300w psu and a ton of cables sitting in my lab at work collecting dust, but it's not mine to sell :(
 
While not my best uptime, right now I'm at a nursing home..poking around the servers, just happened to remember this thread and check out the stats. On their old server, which is an ML 350 G3....I can't wait to retire this box, but it's Server 2000 and Exchange 2000 and SQL 2000...yes this one server was the only server for this network of 50+. Since I've taken over their network I've been adding some DL360 and DL380 servers to split up the roles, and right now this Server 2000 box is really just their file and print share server...with Exchange running.. but no mailboxes left.

net statistics server
Statistics since 5/6/2009 10:13pm

So she's gone for over a year without wonkiness....I think I'll see if I can get anutter year out of her...hope to retire her next winter.
 
I found this one on my network.

Catalyst 4506
Code: 
GV4006-IDF1 uptime is 3 years, 25 weeks, 4 days, 10 hours, 24 minutes
System returned to ROM by power-on
System restarted at 21:59:52 EST Mon Nov 27 2006
System image file is "bootflash:"
 
You must log in or register to reply here.
Back
Top