Slashdot Mirror
Can Maintenance Make Data Centers Less Reliable?
miller60 writes "Is preventive maintenance on data center equipment not really that preventive after all? With human error cited as a leading cause of downtime, a vigorous maintenance schedule can actually make a data center less reliable, according to some industry experts.'The most common threat to reliability is excessive maintenance,' said Steve Fairfax of 'science risk' consultant MTechnology. 'We get the perception that lots of testing improves component reliability. It does not.' In some cases, poorly documented maintenance can lead to conflicts with automated systems, he warned. Other speakers at the recent 7x24 Exchange conference urged data center operators to focus on understanding their own facilities, and then evaluating which maintenance programs are essential, including offerings from equipment vendors."
I suppose that I'd agree. Back in the early 90s, I inherited from a friend a fear of rebooting, turning off, or performing maintenance on a computer. Half the time he opened the case, the computer would become unbootable or never turn back on. Luckily, as a talented engineer, he could usually fix whatever the problem was, but it was a huge pain in the ass. Of course, back then, commodity computer hardware was hugely unreliable, with vast gaps in quality between price ranges, and we were working with pretty cheap stuff. Still, to this day, I dread the thought of turning off a computer that has been working reliably. You never know when some piece of crap component is nearing the end of its life, and the stress of a power cycle could what pushes it over the edge into oblivion (or highly unreliably behavior). I used to be fond of constantly messing with everything, fixing it until it broke, but his influence moderated that impulse in me, to the point where I usually freak out when anyone suggests unnecessarily rebooting a computer. Surely, there's something to say for preventative maintenance, and I'd rather be caught with an unbootable PC during regularly scheduled maintenance than suddenly experiencing catastrophic failure randomly, but there's something to be said for just leaving the shit alone and not messing with it. Every time you touch that computer, there's a slight chance that you'll accidentally delete a critical file directory, pull out a cable, or knock loose a power connector. The fewer the times you come into contact with the thing, the better. If I could build a force field around every PC, I probably would.
The purpose was partly to stop qualifying being its own arms race, with cars in completely different specification than for the race, and partly to reduce costs and the number of travelling staff. At the same time, "T Cars" --- a third car, available as a spare --- were banned, so that if a driver destroys a car in practice the team either have to rebuild it or not race. They're allowed to travel with a spare monocoque, but it cannot be built-up and it does not get pit space.
There were endless howlings from the teams, claiming that without a complete strip-down after qualifying, with a large crew working overnight to check everything on the car, reliability would go through the floor and races would finish with only a handful of stragglers fighting a durability battle (our US viewers may find this ironic in light of a certain US Grand Prix, of course).
The same argument was advanced, mutatis mutandis, over limitations on engines and gearboxes, limitations on the number of gear clusters available, limitations on certain forms of telemetry and a wide variety of "the cars can't just be left to run themselves, you know" interventions.
In fact, reliability is now far greater than ten years ago. It's not uncommon for there to be no mechanical retirements, certainly not from the longer-standing teams, and the days of engines imploding on the track are long gone. A front-running driver will probably only have one, if even that, mechanical DNF per season. The teams deliver a functioning car when the pit lane opens at 1pm Saturday, and that car then runs twenty or thirty laps in qualifying and sixty or seventy in the race, a total of perhaps 250 miles, without much maintenance work beyond tyres, fluids and batteries (section 34.1 on page 18 of the sporting regulations).
So again, we see that "preventative maintenance" turns out to really be "provocative maintenance", and leaving working machines alone is the best medicine for them.
i can't agree. i used to but now i cannot afford to.
we recently experienced 2 catastrophes (datacentre-wide downtimes, you know things that NEVER happen) and the results were unbelievable. GRUBs failed to load OSes, machines were without a bootloader (due to emergency disk hotswaps), some machines simply didn't turn on, services didn't autostart, a few virtual servers autostarted on multiple hosts (instead of just one), fsck on some of our volumes took hours to finish, 30% of supermicro IPMI cards were unresponsive, etc. it revealed that almost nobody had followed procedures properly.
after that, every single service we have is built in a clustered manner with nodes spread across multiple datacentres. I now restart machines and pull cables at regular intervals to test bgp/ospf, clustering, recoveries, to check filesystems, etc. i am now also ABLE TO SLEEP.