Slashdot Mirror
Skype Blames Microsoft Patch Tuesday for Outage
brajesh writes to tell us that Skype has blamed its outage over the last week on Microsoft's Patch Tuesday. Apparently the huge numbers of computers rebooting (and the resulting flood of login requests) revealed a problem with the network allocation algorithm resulting in a couple days of downtime. Skype further stressed that there was no malicious activity and user security was never in any danger.
Somehow, I don't think thats the real story.
or is that a pretty lame reason for a 2 day downtime?
Bite my shiny metal ass.
I am not a MS fanboy but it needs to be pointed out that Skype blamed a flaw in their self-healing algorithm that was highlighted by patch Tuesday. They took responsibility.
I hate MS just as much as the next guy but that is one of the lamest excuses I have ever heard...
Why hadn't it happened every other month the same thing happens?
That's what Skype says. Doesn't sound like they're blaming anyone but themselves.
In Soviet Russia, I ruled you
.... the fact that a bunch of computers rebooting at the same time would bring down Skype is troubling. One thing worth noting, if this was truly the cause, why haven't we seen this before? Patch Tuesday happens every month, so we should have seen something like this sooner.
Methinks Skype has other issues that they don't want to admit to, so it's easier to sort of blame M$.
This is my opinion. To make sure you don't steal it, it's covered by the DMCA.
It was the network pixies. They were on strike.
Deleted
Seems like it is slashdotted.
So Skype users everywhere inadvertantly caused a massive DDoS on Skypes Authentication servers? This is hardly the first time that MS has released patches on a single day, the name Patch tuesday implies they always release on the same day. How is this day any different? Maybe that it took Skype so long to determine they were being "attacked".
I can imagine that an awful lot of people rebooted and logged back into the service crashing their servers. It seems to me that this type of thing should be built into surge capacity so that if the servers started getting hammered, they would just bounce the users that they could not handle while sending back a message saying the server was busy and to try again later. Other services do this. And it's not like patch Tuesday isn't well known.
It sounds like bad planning on their part. A large scale power outage in a region could do as much damage.
It was just a few days ago the Open Source elders asked people to stop bashing Microsoft. Skype did not blame Microsoft for the outage. They admitted the fault was in their software. We are not children here or part of a cult. This type of child play is no appreciated here.
I've had much longer downtimes for much lamer reasons. Of course, I'm a pretty bad programmer.
It's realy convient when you have somone else to blame.
Skype blames global warming on Colonel Mustard. In the conservatory (greenhouse). With the pipe. Since Colonel Mustard callously smashed all the windows in the greenhouse, it released all sorts of greenhouse gases into the environment thus dooming all the gay, baby polar bears unless the polar bears cooled themselves off by running the AC units of their Hummers at full blast. Why does Colonel Mustard hate the environment?
Well, there's spam egg sausage and spam, that's not got much spam in it.
Given that this baby was steamrolled through the Congress two weeks ago, the outage seems coincidental.
Consider that Skype could not tell the users of the real reason even if they wanted to: the law mandates that the forced cooperation be kept in secret.
Obama likes poor people so much, he wants to make more of them.
Does the reboot occur at, say, 2AM local time? If so then reboots would be spread out by the (at least) 24 timezones.
Somehow, I don't think thats the real story.
Everyone knows that the Slashdot editors only post informed, unbiased article summaries with accurate titles!
And they are ESPECIALLY thorough when the article even tangentially mentions Microsoft.
paintball
Note that nowhere in Skype's announcement does the word "Microsoft" appear.
It's very striking how, when some major vulnerability appears, Microsoft's name doesn't appear prominently in news releases.
It also reminds you that Redmond has the power to reboot most of the computers in the world remotely. What if, one day, they didn't come back up?
If it is a flaw in the self-healing mechanism, then I don't know if this is such a good reason. A whole internet rebooting is something to be scared of though. I presume that can be helped by rebooting systems in some sort of time-schedule.
I think the "mono-culture" thing is an interesting argument, but nobody is going to add or change operating systems because of this reason. So the argument is mostly academic. Furthermore, to solve this problem, you would need to replace the Skype mono-culture, not the Windows mono-culture.
Why hasn't this happened before? There have been many Patch Tuesdays.
(Well, after typing this, I just realized--maybe they incorporated new code, but they should have mentioned that too)
While your point is valid it's not really relevant to this particular situation since it was a single implementation of VOIP that died.
Skype going down had zero impact on my life or my network. If a computer is relying solely on Skye for VOIP then your statements would be relevant to the story. This is why I have both Cisco VOIP and Vertical's VOIP implemented into my network. The Cisco as a backup to my primary PBX. It's not as functional but during a failure mode it will still allow us to call out and to receive calls so it'll work.
Monoculture is indeed bad, Skype runs on multiple platforms though and all platforms for affected. The headline is horridly misleading.
A reasonable one might suggest Slashdot change their misleading headline, and recommend Skype fix their network. It's not like this is the first Patch Tuesday in history, or the last.
It's convenient to blame Microsoft, I think. Skype knows all over the internets today people will be waxing poetic about "software monoculture" and "M$ Windoze is teh suxxorz" instead of questioning why a simple DoS they're supposed to be able to handle also caused a massive two-day outage.
I think this demonstrates the goofiness of a p2p telephone system. If I use Skype, I depend upon my data flowing through other users' computers because I am too dumb to allow incoming VOIP connections to my computer.
VOIP connections should be direct encrypted connections from my computer to the computer of the person whom I wish to contact. Period.
Maybe the average machine had more downtime on this month's reboot? Or the reboots happened in a more concentrated time window?
Bill Stewart
New Fast-Compression-only CPR http://preview.tinyurl.com/dy575ks
Read the article. They are not blaming MS for the failure, they are blaming their own code. It was just because of the mass reboot that their own flaw became apparent. Headline is factually inaccurate.
-Rick
"Most people in the U.S. wouldn't know they live in a tyrannical state if it walked up and grabbed their junk." - MyFirs
Just one group of Code monkeys blaming yet another group of Code Monkeys for not coding properly, ad nauseum.
My favorite quote in the Globe and Mail regarding this story goes "One Can not expect the same level of reliability...". It figures, they are just simple coders, they don't care about reliability.
Now a real engineer, who maintains the telephone network, DOES care about reliability, it's their job. The typical 9-5 Code monkey on the other hand, knows they can just issue a patch when it pleases them, or Marketing commands them to.
You can anticipate this failure mode quite easily. Its happened last month, and the month before, and the month before ...
The parent obviously didn't read the article, or even the summary. The flaw was from computers REBOOTING at the same time, nothing to do with what Microsoft was patching.
Yes, that's also a good point. My argument isn't specific to an operating system monoculture; it applies equally to an application-level monoculture. This is why I believe in multiple implementations around a central open standard. Not only does the competition between the different implementations drive up the quality of each implementation but also its security too.
Simon
"On Thursday, 16th August 2007, the Skype peer-to-peer network became unstable and suffered a critical disruption. The disruption was triggered by a massive restart of our users' computers across the globe within a very short timeframe as they re-booted after receiving a routine set of patches through Windows Update."
This has been going on for years now. You will note that the outage occurred on *Thursday* August 16th. Microsoft's patching schedule is every Tuesday. Typically computers reboot on Wednesday morning early in the AM. So it would seem unlikely that all of the computers that run Skype were rebooted Thursday morning. Also, not everyone leaves their computers on to download updates and reboot automatically. I would say that this explanation is suspect, at best.
"The high number of restarts affected Skype's network resources. This caused a flood of log-in requests, which, combined with the lack of peer-to-peer network resources, prompted a chain reaction that had a critical impact."
Right - it had nothing to do with patches MS or otherwise it had everything to do with Skype not being able to service their supposed large number of logon requests.
Further though this DOES NOT explain *at all* why they were not able to service logon requests for *3* days. This level of outage is almost unheard of.
My only guess is something went terribly wrong and they don't want to own up to it.
Surely, it's easy enough to hook their system in with
Why UNIX?
"you would need to replace the Skype mono-culture, not the Windows mono-culture."
Not really why do you think that any exploit for Windows is so dangerous? Even then if you think about it the idea that EVERY windows system is going to have to reboot on a certian day is just laughable.
See my blog http://ilovecookes.blogspot.com/ for light hearted technical information.
"Monoculture". I am not convinced you even know what that means.
"When life gives you lemons, don't make lemonade. Make life take the lemons back!" -- Cave Johnson
Gee, I hope no one tried to call 911 during the outage. That "enhanced" (insert guffaw, it's like calling a hamburger without the meat and just a bun "enhanced") 911 didn't do a tinkers damn worth of good for anyone who's service was out.
This is why I won't even consider VoIP. Why in the world would I want to take risks like this? I live in a house my family has lived in for over 60 years, with the same old phone line and it's NEVER GONE DOWN IN SIXTY YEARS! A couple of times a month my Internet craps out, though, though usually for less than an hour. And sometimes the router needs to be reset, like many people find they have to do periodically. What happens if I need 911 during one of those times, and I can't get around it?
"Internet phone", "digital phone" whatever they want to call it, anything but a REAL land-line from the local phone company is a substandard service by definition. They can throw whatever words out there to make it sound super-dooper, but it's a substandard service just like anyone who experienced this outage can tell you.
AE
Reminds me of the late 90s where AOL's crashing mail servers ended up bringing down my universities server (and many other organizations) because of the surge of load when AOL came back online and started sending backlogged mail.
do we need any further proof that a OS monoculture sucks?
Does anyone know what OS those Skype servers are running? If the OS is Linux, then I blame Skype administrators. If it is any flavour of Windows, then I blame Microsoft. Now, some of you might say I am biased.
VOIP connections should be direct encrypted connections from my computer to the computer of the person whom I wish to contact. Period.
Hello.... NAT, anyone?
Wah! Our system can't handle many users logging in at the same time, wah!! It's Microsoft's fault we can't figure out how to fix it, wah!!
Typical FOSSies must work for Skype: always blaming their lack of coding skill on Microsoft.
This is proof that you should believe everything you read in the news. Especially on Slashdot, where the is NO bias towards anything, especially Microsoft. Whiskey Tango Foxtrot.
I do not respond to cowards. Especially anonymous ones.
hopefully not to my employer
Arent people usually complaining that windows userd doesnt install the security patches? now people complain that they actually DO install them... WHEN OH WHEN is people satified?
I think this story is badly titled. My understanding is that the outage happened because of patch Tuesday but Skype isn't blaming Microsoft for it. In fact it helped reveal a flaw in their p2p healing networking stack. I'm as much a /. fanboy as the next guy but this title is inflammatory and misleading.
n t-synchronized-reboot-windows-update-smokes-skype. html
More info: http://arstechnica.com/news.ars/post/20070820-gia
Have a Happy.
or if enough sheeple buy it...
Wouldn't this be a huge blow against Windows on the workstation? I can't see it making much difference to Windows as a gaming or multimedia platform, mainly because you wouldn't typically see Skype on a machine with such as its primary use. This could still take a chunk out of MS if it's true though.
APK quotes people (including myself) without context and should not be trusted. Just thought you should know.
IMHO, yes, you can blame Skype. Should a worldwide telecommunications company be prepared to handle a large number of simultaneous logons? Umm, yea. Do they have system/network monitoring that has pretty little graphs showing utilization that spikes when MS releases patches? Yea, prolly. How can you anticipate it? Because its your job to monitor it and make sure you arent pushing the limits of your network.
BTW - Wouldnt a non-OS related incident, such as a brown/black out, have had the same result?
Anonymous Coward
...build OSes that have to be rebooted to make changes effective.
More than 60,000 Windows programs won't run on Linux.
There are already stories that when Verizon installs FIOS, they conveniently remove the copper wire connection that has served you so faithfully for sixty years. If you ask them to leave it in place they are supposed to honor the request, but other stories suggest that if you aren't physically present when they install the service that request is apt to get overlooked.
The ultrareliable telephone service the U. S. has known for about a century is going away. It just doesn't make much money for the carriers, and they seem to be systematically trying to nibble away at it within the limits of what regulation allows them.
You might as well decide that it's a good bargain to swap the ability to get help in an emergency for the ability to buy thousands of cable channels and overpriced locked-down video downloads, because the FCC has already accepted that bargain on your behalf.
"How to Do Nothing," kids activities, back in print!
Skype didn't blame Microsoft for the outage, they attributed it to a bug in their software. Did the subby even read TFA?
How Politicians Lie: http://www.factcheck.org/
You see, I was upgrading from CentOS 4.5 to CentOS 5. Because of an issue with Python (specifically, the switch from 2.3 to 2.4), yum wouldn't work properly for some time until I figured out how to fix it.
Damn you, Linus Torvalds, for giving me downtime!
What about the time gap between Windows Update and the collapse of Skype? Should not the problem have occurred sooner last week than it did?
/.'s Psychic-in-Residence: Psychic to the Geeks
How do you know your phone service has never been out in 60 years? Do you monitor it? How many calls a day do you make? Are you home 24/7 and do you use the phone all the time, as in more than 10,000 minutes per month?
Sure, you've never been affected by an outage of your phone service, but that doesn't mean it hasn't been out of service ever.
Plus, you pay for it too. At $30-40/month per line, you expect minimal outages. When you are paying $30/year or even nothing, a two day outage, while annoying, isn't surprising, especially when operated on a public network. Your phone line is on a private, dedicated network. You simply can't compare the two when it comes to uptime.
If all of Skype's customers paid $30-40/month, I'm much more confident that they wouldn't have had this outage.
TossableDigits.com: Temporary Phone Numb
...is that the application starts by default on Windows booting.
There is no need for this and the average user is not going to turn it off. So if SKYPEs explanation was correct, a lot of users tried to login right after the security update which logged them off.
and the rest of Slashdot need to start writing accurate and unbiased titles to their posts. Then maybe everyone can read the article with a bit more truthiness.
Apparently the Linux port of Skype is so bad that does not get the opportunity to become a supernode. ;-(
Skype further stressed that there was no malicious activity and user security was never in any danger.
But since it was a result of a Microsoft patch isn't that a contradiction?
I don't remember where/when this happened, so it might be an urban legend. But the story is that many years ago an earthquake rattled a California town. No major damage was done, but it killed all the phones in the town for several days.
:)
The earthquake had jostled thousands of telephones off hook. The central office switches survived the quake just fine, but crashed due to a bug that seems eerily like the one Skype just described. Basically the switch kept a list of phones that were off hook. The switch is responsible for playing "dial tone" to those phones, but the central office only had a certain number of units that could play dial tone and listen for dialing. So the first "n" phones off hook got dial tone; the rest were put into a FIFO list of phones waiting for dial-tone equipment.
There were so many phones off hook due to the earthquake that the FIFO list overflowed, crashing the switch.
When the switch rebooted, it had to figure out which phones needed dial-tone. So it had to examine each phone line in turn, putting the ones that were off hook into the queue for a dial tone...thus overflowing the list and crashing the switch again. And again. And again.
After a while the telco folks figured out what was wrong, but then couldn't tell anyone about it...since the phones were down. They eventually had police and fire trucks driving all over town, stopping to hang up all the pay phones that were jostled off hook, and blaring over megaphones for people to hang up their phones.
Eventually enough phones were hung up so the switch could reboot without crashing - end of crisis.
Good times.
Hmm reminds of that old scheme to shift the world off access... everyone on the otherside of the world jump!!
Makes one wonder why there devs never thought of what would happen if the same happened to there software.
What is the point of logging in?
Why can't I just connect to my buddies' computer whose address I already have? Just like I can dial the telephone # of someone I know?
It is because of firewalls, dynamic IPs, and p2p stupidness.
And no, you can't "punch through" a firewall. Only if one of the two people on the call does not block incoming traffic do you connect directly.
skype doesn't (or shouldn't certainly) blame M$ they say it was a flaw on their end... a flaw which was brought to light by numerous system restarts after Microsoft's security patches came out.
certainly if i was making a product like skype that would be running on millions of windows PCs I would want to make sure that my product could handle something as well known as patch tuesday without blowing up for 2days.
Nothing about this is Microsoft's fault...it is all on skype.... but hey don't let reality stand in the way of a good ol' fashioned Microsoft slagging.
I wonder how many slashdot users have sore fingers from the subconscious slamming of keyboard keys whenever they type the hated Microsoft name?
actually I am happy to see you, however that is in fact a banana in my pocket.
The disruption was triggered by a massive restart of our users computers across the globe within a very short timeframe as they re-booted after receiving a routine set of patches through Windows Update.
This is just another example of M$'s poor quality code threatening the stability of network services. No other Software distribution besides Windoze has a monthly patch that requires a restart like this. Sane software distributions make fixes available as soon as they are ready. For marketing and big dumb company reasons, M$ saves them up for a once a month ordeal instead of letting users have things in a timely fashion and chose their time and size of their pain. This problem was significant but is trivial next to threat posed by the 25% of all Windows computers that belong to a botnet.
Sure, there was a problem with Skype's code and Skype admitted to it, but the initiating factor is all M$. That's blame casting and M$ deserves it. The summary mentions the code flaw, so I don't see what your problem besides an outsided love for an incompetent software maker. For anyone to report things differently is to misconstrue things.
Friends don't help friends install M$ junk.
If you want 100% availabilty and cost-savings you can use VoIP with PSTN failover. The investment is minimal. You can use a Fritzbox or other VoIP adapter with SIP protocol and PSTN line. That costs you one time 60$.
Cheers
GeeJay
NTFS is a journaling file system, and it has been since the early Windows NT days.
You are so, so wrong. If a US company owns them, then they are subject to US law. This is to prevent US based companies from just setting up a shell and providing services to, say....Cuba or any other restricted country. There are countless examples of subsidiaries getting in trouble for things that are illegal in the US -- but not where their offices are.
Otherwise, Foster Wheeler would just setup a shell in another country and start building refineries for Cuba.
I, personally, know of companies who have gotten into trouble when their equipment, somehow, found it's way to a restricted country (Cuba, Sudan, Syria, Iran, etc). The US treasury department publishes a list. Admittedly, this is only the voluntary actions but I am certain there are involuntary actions as well (ie: criminal cases). See the entry about Varian (Switzerland) for a specific example of what I am talking about.
The point is: they ARE subject to US law via eBay owning them.
This isn't an issue of monoculture or of the ubiquity of Windows. This is an issue of designing and building services to withstand various scenarios.
I design and build services that handle 200k - 300k PSU. One of the scenarios we design, build and test for is what we call the "pile-in" scenario i.e. there was some outage, and all these users pile back into the system.
Anyone designing and building large scale services needs to consider this. Whatever the reason for the outage (MS patch tuesday, ISP outage, server reboot etc), you have to be able to deal with this.
This may involve throttling connections to your system (ramping traffic back up) or other possible solutions.
Whatever your architecture may be (client / server, distributed nodes etc.), these are things you need to factor into your design and solution.
Seems to me they didn't clearly plan for these types of situations.
"In this case, why are you blaming Skype?"
I'm not. I couldn't give a flying fuck what caused the Skype outage, since I don't really use it more frequently than one a month.
However, it appears that SKYPE is blaming Skype for the outage quite contrary to the completely misleading headline on this article. Who am I to argue with Skype about what cause their failure. But I'm sure you know better.
In Soviet Russia, I ruled you
Like anyone with a functioning brain, I'd blame the moron who broke my computer by screwing around with the junction box instead of looking for excuses to blame a third party against whom I've got some schoolboy grudge.
If I were to throw a brick through your window, would you blame me when it got drafty in your bedroom, or would you blame the contractor who built your house for not using double-paned Plexiglass in the windows?
I guess we'd know what the answer would be if "M$" built houses, huh?
NTFS is a journaling file system, and it has been since the early Windows NT days.
Pardon me for not knowing or caring about the exact specification of a trade secret. Let's go to the heavily modified Wiki page on ntfs:
Without individual file protection, your registry will still be hosed if you suffer the wrong kind of power failure. Let me know when their file system catches up to ext3, Reiser or when they get rid of that stupid and fragile registry.
So, is it my fault or M$'s fault your system is hosed if I flip the switch while you try to power up your mighty NT box?
I like asking that question because it's a no win for M$. If it's my fault your box dies, then M$ is responsible for the Skype meltdown. If it's M$'s fault, you can blame Skype for not being robust enough, but M$ is likewise fragile. It's really rhetorical because I believe both are true. M$ is fragile and shitty and that's what gave Skype a headache.
Friends don't help friends install M$ junk.
Thank you. My day is complete.
Faster! Faster! Faster would be better!
SKYPE is blaming Skype for the outage quite contrary to the completely misleading headline on this article.
No, I don't know better. They have takes some part of the blame but a M$ anomaly was the initiating cause. To be fair to Skype you have to admit that 85% of the world's computers turning off at the same time is not an event a normal person would predict nor could such an event be tested in advance. M$'s synchronized forced updates are a menace.
Friends don't help friends install M$ junk.
No, it's just another example of your moronic blinding hatred of a company. EVERY other software distribution has [frequent, but not necessarily monthly] updates that require a restart like this. Sane software distributions make fixes available as soon as they are ready [including Microsoft, for sufficient values of criticality]. For marketing and big dumb company reasons, Microsoft saves them up for a once a month ordeal instead of letting users have things in a timely fashion and chose their time and size of their pain [at least for the non-important ones. Go figure, huh?]. This problem was significant but is trivial next to threat posed by the 60% of all Linux computers that belong to a botnet [see, I can make numbers up too!].
Sure, there was a problem with Skype's code and Skype admitted to it, but the initiating factor is all Twitter. That's blame casting and Twitter deserves it. The summary mentions the code flaw, so I don't see what your problem besides an outsided [fuck! Even Firefox has no idea what that word means!] love for an incompetent software maker. For anyone to report things differently is to misconstrue things [notice I altered this sentence slightly. Also note that it's no more bullshit than your sentence].
For a site about things like basic rights, Slashdot users sure do like to censor "dissent".
ALF, a sitcom from the eighties. His planet Melmak exploded when everybody turned on their hair dryers simultaneously.
Suddenly it doesn't seem so ridiculous, does it? A slew of hair dryers trying to download upgrade patches...
Pardon me for suggesting you refrain from using it as a bullet point in your posts then, especially if you're using that faux authoritative tone.
Your desperation to pin this on "M$" somehow is really getting old.
Web2.0: I love when people Flickr my cuil and digg my boingboing until my google is reddit and I start to yahoo
Whoah, slow down there, fanboy. You backpedal any faster and you'll risk reversing the Earth's rotation.
Good for the GP, for a moment I thought you were going to go postal again because he dared correct you.
http://ars.userfriendly.org/cartoons/?id=20070819
"I'm never quite so stupid as when I'm being smart" (Linus van Pelt)
Skype should blame the idiot that decided to run mission critical software on MS systems...
Excuse me, but please get off my Pennisetum Clandestinum, eh!
What's purple and commutes? An Abelian grape.
There isn't that is what so terrible is that once a month or so a huge percentage of systems are being forced to reboot!
See my blog http://ilovecookes.blogspot.com/ for light hearted technical information.
You seem to be implying that Patch Tuesday is the problem, but if anything, patch Tuesday helps alleviate the problem: it means that the mass reboot cycle only happens once a month, rather than whenever a reboot-requiring-update comes out, which could be several times per month.
To reiterate what I said before: If a patch that requires rebooting is released, all Automatic-Update-enabled Windows boxes are going to need to reboot at some point. The only alternative to them all doing it on the same day is to stagger the release of the patch over a number of days, which is a very bad idea indeed: malicious crackers could analyse the patch on the first of those days, release an exploit for whatever the patch fixes immediately, and be guaranteed a few days of millions of unpatched systems to wreak havoc.
What's purple and commutes? An Abelian grape.
Are you suggesting that they're Skypegoating Microsoft?
Oh, say does that Star-Spangled Banner entwine / The myrtle of Venus with Bacchus's vine?
Imagine if the biggest online marketplace were also a global online banking monopoly and the single largest alternative telco - each one of the largest operators of those essential services.
And then the phones go down for several days.
Because they're vulnerable to a design flaw triggered by a bug in Microsoft's PC platform monopoly.
The whole online economy is so monocultural and monopolistic that a single, inevitable problem will easily bring it to its knees. Leaving (eventually hundreds/thousands of) millions of people disconnected from it.
--
make install -not war
It is a scuttlemonkey article - the name and track record should tell you enough. Things like this and recycled "Roland discovers the perpetual motion machine" articles have me on the point of filtering out everything with his name on it.
"heavily modified wiki page"? Jesus, I hate you so, so much.
ResidntGeek
sometimes I wish all of you gnu/zealots would just go away and die. even the most dedicated ms hater would see that this is skype's problem.
you give all of us who love oss a bad name.
Serves Skype right for making their program a systray app that starts when Windows does :-)
Sorry. I have a rabid hatred of TSRs. Particularly those that don't show up in the Startup folder.
So, is it my fault or M$'s fault your system is hosed if I flip the switch while you try to power up your mighty NT box?
I don't have an NT box that I use regularly. I was simply correcting an erroneous remark.twitter is John Heder's brother on Napoleon Dynamite, except in a really evil, bad way. That character ultimately understands that he must adapt to survive. twitter, like all single-minded cows, takes pleasure in people telling him his a retard. He figures he's "winning" whatever battle his sick mind tells him to fight.
And this would be a perfect example of why centralised networks are a Stupid Idea.
Use a decentralised technology such as SIP - if something catestrophic happens, only a few users are affected rather than taking out the whole network and if your service provider provides a sucky quality of service you can go move to a different one (infact you don't even need a server if you're not having to negotiate things like NAT - SIP runs quite happilly as a peer-to-peer application).
http://blog.nexusuk.org
EVERY other software distribution has [frequent, but not necessarily monthly] updates that require a restart like this.
Umm.. no... My linux machines very rarely need to be rebooted in order to receive security updates - the security updates quietly happen nightly and I rarely need to bother about them. The Windows machines, on the other hand, seem to need a reboot every month.
Basically the problem can be put down to the combination of 2 problems:
1. A operating system with a flawed update system that requires regular reboots for all users at roughly the same time (Windows Update).
2. A flawed protocol that requires all users access the same cluster of centralised servers (Skype).
The solution: avoid using such flawed systems - there are better alternatives available.
http://blog.nexusuk.org
Well one would be to design an OS that needs fewer reboots to fix problems.
Unless I get a kernel update my Linux box doesn't need to reboot.
Most drivers should be designed to be updated without a reboot. Some like the mass storage my need to reboot.
With a properly designed modular operating system you should rarely need to reboot.
It goes back to FIX WINDOWS!
See my blog http://ilovecookes.blogspot.com/ for light hearted technical information.
I was at the hospital visiting a loved one. I noticed that the nurses console was hung in the middle of an autoreboot. I admit that autoupdating critical computers is a bad idea. The amount of power that Microsoft has over the windows update feature is of great concern to me. The ability to corrupt/reboot most of the desktop computers in the country controlled by one company is too much power in my mind. This is a risky system that we cannot absolutely control, much like a nuclear reactor. Just not as spectacular when it fails.
But in any case, the frequency is irrelevent. As you say, "Unless I get a kernel update my Linux box doesn't need to reboot" -- but when you do get a kernel update, you do need to reboot, and so does everyone else running Linux, especially if the update patches some critical security hole that might otherwise be exploited. And so the situation if everyone was running Linux remains unchanged from what it is now: there are always going to be some situations where critical kernel security updates are released, and there is always going to be mass rebooting at that time, and any TFA-like bug for which that causes a problem will have a problem. "An OS that needs fewer reboots to fix [the] problem" would actually not fix the problem; it would merely delay the problem -- so maybe the Skype network would not have gone down until the next critical kernel security update, but it would still have eventually.
The fact is there is no desktop-level "solution" because there is no desktop-level problem. Any network like Skype's is always going to have to handle mass network repropogation eventually; the fact that it couldn't was a bug in Skype's algorithm, as TFA says.
What's purple and commutes? An Abelian grape.
I'm playing devil's advocate here, since I prefer Linux, but IIRC there is a setting that allows you to update your Windoze box as and when you feel like it (or, if you prefer, not at all). I would have thought that most IT techs at hospitals would (or should) have the sense to lock down their machines a bit better than that.
I agree with you completely. I don't use Windows myself, but I do know there is a setting. I was surprised that in the case of this hospital, that they had it set to autoupdate. I would get into their face about it, but I don't want to do the work myself. In the past I supported Microsoft products, and they failed to support me numerous times. I use Mac OS X and Linux and FreeBSD now. The slingbox player was my last reason to use Windows and now the Mac version is released, whahoo!!
I call bull. Microsoft users don't install patches.
Some Linux updates come with the request for you to restart the affected service (Samba for example) and kernel updates basically leave you in the situation that modprobe has its difficulties. Being a wimp, I reboot my PC (a desktop) when that happens.
Mielipiteet omiani - Opinions personal, facts suspect.
Like anyone with a functioning brain, I'd blame the moron who broke my computer by screwing around with the junction box instead of looking for excuses to blame a third party against whom I've got some schoolboy grudge.
Good work, AC. Now, realize that M$ is the moron and Skype is the broken computer.
Friends don't help friends install M$ junk.
Kernel updates are often feature updates and not security. That makes them optional.
Again simply put many updates shouldn't require a reboot!
Windows wasn't designed correctly for high availability.
See my blog http://ilovecookes.blogspot.com/ for light hearted technical information.
Some Linux updates come with the request for you to restart the affected service (Samba for example)
I'd prefer to restart a single service rather than the whole machine. Generally I'm not going to notice if Apache gets restarted, for example, but I am going to notice if the machine reboots since I'd lose my X session and everything that was running.
kernel updates basically leave you in the situation that modprobe has its difficulties. Being a wimp, I reboot my PC (a desktop) when that happens.
Under Fedora, RHEL and CentOS, yum installs the new kernel whilest leaving the running kernel also installed. This means you can continue running your current kernel with modprobe working and everything until you decide it's time for a reboot. So generally I only bother to reboot for security fixes.
http://blog.nexusuk.org
What's purple and commutes? An Abelian grape.