Sun v440s have hardware mirroring on internal disks, although it's only supported on 2 of the 4 (i.e. you can mirror any 2 disks using the hardware RAID, after that you need to use software RAID). In any event, I've not seen any issues in using software mirroring on any server (Sun, HP or IBM). The extra CPU load is minimal.
Additionally, having the boot disks internally (or on a small enclosure) simplifies the bootup of the system which is crucial for problem resolution; this is especially important in SANs; if your system won't boot is it because of a problem in the server hardware, HBA, fibre cable, SAN switch or the SAN itself? With SCSI disks (internal or otherwise) there are far fewer things that can be wrong and they are far easier to debug.
I'm not aware of any limitation in booting from external disks on any Sun systems (it's mandatory on 4800 systems and higher, up to 25k); all it really needs is a disk path the PROM can understand.
One of Sun's mottos is Reliability, Availability, Servicability. You've covered one there and Availability generally follows on from that.
As for servicability, hot-plug on most components is what really sets server-clas hardware apart from desktops. Even if you're not 24/7, being able to remove a failed hard disk at any time means you don't have to schedule an outage which may introduce more errors (hard disks tend to fail more often when they're stopped/started). On a decent server, you can replace a hard disk in 2 minutes flat, although the resync operation may take a few hours.
Welcome to planned vs unplanned downtime; in many cases, a 10 hour outage can still give you a 100% availability if you planned that outage. What they're probably quoting is 0.0000001 unplanned downtime.
By the time I left my first job in March 2001, I had been logged into my Sun desktop since October 2000. This wasn't even a standard Solaris desktop, I was running WindowMaker + the KDE panel with zero problems over that time. This wasn't even an idle system; I used that desktop extensively, including running xmms (probably including dev builds as I was maintaining the Solaris XMMS port at that time), CDE mailer and netscape all the time.
In summary, that's a 5 month login session when NT boxes often need rebooted every 5 days (or hours in extreme cases). I'm sure others have managed longer.
I stopped using floppies after I'd had a home network for a while; it was easier to use FTP or Samba to move files around. What killed off my floppy drive usage was when I had problems in Windows. Every time I opened explorer, it hung for about 5 seconds. After much digging, I tracked it down to the floppy drive. I tried various options, including disabling the drive, but eventually I just removed it and never missed it; all the files I cared about were on the network and I burned copies to CD-R if I needed to move them.
Last time I had to use a floppy was to install network drivers on a PC.
I think I know exactly the bug, because I kept getting it. Basically, every time I tried to sync my iPod (USB device...) XP crashed and corrupted the registry requiring a reinstall. Eventually I decided to use the serial docking device. However, transferring files from/to a USB CF card reader did the same, so I narrowed it down to the USB device. My workaround was to install a spare USB card to handle it; this works fine.
Worth pointing out that everything worked great from a VMWare session running under linux, so I was certain it wasn't (a) hardware or (b) the type of transfer or (c) some combination; I blamed the VIA drivers for my motherboard, but neither the stock XP drivers or the MB drivers which came with the MB did any different.
Mebbe I'll load up SP2 and see what happens; my system's been playing up for a while and a fresh install might clear it up:)
Compatibility with 64-bit apps written 10 years ago.
Bull. 64-bit Solaris started with Solaris 7 which must have been about 98/99; maximum age on a production 64-bit Solaris app is 6 years.
That said, Solaris has reasonably good binary compatibility with apps from SunOS 4 and any 32-bit app written to comply with the ABI specs of previous Solaris releases.
Hot-swappable CPUs
Hot-swappable if (a) you can find documentation to confirm that you can hot-swap system boards and (b) the system will let you DR the board out. The latter is a kicker, as I've seen 12Ks which wouldn't DR out a board because of a caged thread on one of the CPUs; I don't think we got an answer as to whether it was a kernel thread or part of Sun Cluster that was locking the board.
I can dynamically split single Solaris instance, running on 128 processors, to N instances each running on 128/N processors.
The smallest unit on a SunFire system with domains is a system board, i.e. 4 CPUs + all RAM on that board. Solaris 10 introduces zones which take that limitation away; time will tell how well that runs.
Stable device driver model. Drivers from Solaris 2.6 will work fine in Solaris 10
Yes, they do, but if you maximise the xterm/dtterm window with a high enough resolution (1152x864 is sufficient, not sure of 1024x768), vi will complain on startup; actually, it just won't start.
Addition to the above; when choosing your language/compiler etc, balance the costs of extra hardware to run the slower language versus the costs of programmers to program/debug the language. The balance these days is that hardware is comparitively cheap compared to programmers, so an easy to write language with quick deploy time is cheaper, even if it takes more hardware.
Read the article; what is actually said is that current products that use broadband (e.g. VOIP providers) will probably not be viable in 5 years time. The/. article is badly worded and misleading, IMO.
The users often are the problem; give a user 10 steps to perform to possibly view some naughty pictures of a celebrity and chances are, a significant proportion of them will do so and infect their computer in the process. Heck, some of them would probably run it as root/admin if you asked them to...
Partially, yes; Windows is still suffering a legacy of idiotic design ideas dated from the early 90s when PCs were standalone boxes with no network. At that point, security wasn't really an issue and the internet was a nicer place. Changing some of these parts would probably require substantial rewrite, break a lot of apps and piss off its customers.
IE popups are a standard part of web browsers and Netscape did exactly the same (until Mozilla popup blockers); most other GUI browsers should allow popups; they are required in a number of places (try installing BMC Patrol). You cannot blame MSIE for that as it was (for a change) following web standards.
Despite all the problems in Windows, you don't need to have an insecure system to be exploited. People are stupid and will do all sorts of silly things on the promise of nude celebrity pictures. Besides, even linux has had security flaws which would allow a worm to spread; the lack of these being widespread is as much to do with the low penetration of linux rather than any inherent security by design.
Back in the NT days, I heard a rule of thumb; "always avoid even numbered service packs". Having not been responsible for any number of win2k systems for many years, I can't say how true that is now, but I do remember that NT sp2 & sp4 introduced just as many problems as they fixed and sp6 was withdrawn after a few days because of issues (Lotus Notes being a major one, IIRC).
In summary, I'm waiting a few weeks between sp2 coming out and installing it on my PCs just in case.
No, the pirates have a blatant disregard of the infrastructure by keeping running insecure, unpatched software. Microsoft should not be held responsible for pirates who illegally run unlicensed software.
The fact that Windows is everywhere is why it's such a tempting target; a hit rate of 1% on virus infection of Windows PCs is a good number, so it's worth going after. If linux had a good market share, it would be running the spam zombies.
Yes, linux can be more secure than Windows, but the fact is that over 90% of these zombie PCs could have prevented infection by simply having (a) their firewalls enabled and/or (b) having intelligent users. By default, most linux distros don't come with firewalls enabled either (at least, the last time I checked; I think it's becoming more common for firewalling to be enabled though, as with XP SP2) and as for (b), well, we'll always have stupid users.
Feh, I've given up trying to figure out moderators; I'll get up to +5 for a fairly simple post and get modded down just as quickly for something else. Overall, I tend to hover at the higher end of the scale, but I'm past caring.
Slashdot Mirror
Additionally, having the boot disks internally (or on a small enclosure) simplifies the bootup of the system which is crucial for problem resolution; this is especially important in SANs; if your system won't boot is it because of a problem in the server hardware, HBA, fibre cable, SAN switch or the SAN itself? With SCSI disks (internal or otherwise) there are far fewer things that can be wrong and they are far easier to debug.
I'm not aware of any limitation in booting from external disks on any Sun systems (it's mandatory on 4800 systems and higher, up to 25k); all it really needs is a disk path the PROM can understand.
Even the USA would have trouble spinning this as calling him an "enemy combatant".
As for servicability, hot-plug on most components is what really sets server-clas hardware apart from desktops. Even if you're not 24/7, being able to remove a failed hard disk at any time means you don't have to schedule an outage which may introduce more errors (hard disks tend to fail more often when they're stopped/started). On a decent server, you can replace a hard disk in 2 minutes flat, although the resync operation may take a few hours.
Lies, damned lies and availability stats...
In summary, that's a 5 month login session when NT boxes often need rebooted every 5 days (or hours in extreme cases). I'm sure others have managed longer.
Last time I had to use a floppy was to install network drivers on a PC.
Worth pointing out that everything worked great from a VMWare session running under linux, so I was certain it wasn't (a) hardware or (b) the type of transfer or (c) some combination; I blamed the VIA drivers for my motherboard, but neither the stock XP drivers or the MB drivers which came with the MB did any different.
Mebbe I'll load up SP2 and see what happens; my system's been playing up for a while and a fresh install might clear it up :)
Also, how about other uses, like readers for the blind or visually impaired?
Isn't that the NIC that caused a product recall?
That said, Solaris has reasonably good binary compatibility with apps from SunOS 4 and any 32-bit app written to comply with the ABI specs of previous Solaris releases.
Hot-swappable if (a) you can find documentation to confirm that you can hot-swap system boards and (b) the system will let you DR the board out. The latter is a kicker, as I've seen 12Ks which wouldn't DR out a board because of a caged thread on one of the CPUs; I don't think we got an answer as to whether it was a kernel thread or part of Sun Cluster that was locking the board. The smallest unit on a SunFire system with domains is a system board, i.e. 4 CPUs + all RAM on that board. Solaris 10 introduces zones which take that limitation away; time will tell how well that runs. I'm pretty certain that isn't supported.Yes, they do, but if you maximise the xterm/dtterm window with a high enough resolution (1152x864 is sufficient, not sure of 1024x768), vi will complain on startup; actually, it just won't start.
Possibly in the short term; in the longer term, people (i.e. historians) will want to know as much as possible about life pre-nuclear war.
Addition to the above; when choosing your language/compiler etc, balance the costs of extra hardware to run the slower language versus the costs of programmers to program/debug the language. The balance these days is that hardware is comparitively cheap compared to programmers, so an easy to write language with quick deploy time is cheaper, even if it takes more hardware.
Could be worse, could be "David Hasselhoff uber alles". He is rather popular in Germany, after all...
We've had cameras in cities in the UK for years with little/no outcry and (to my knowledge) no abuses being reported.
For those that are unaware, Sun's HME interface is known as "Happy Meal Ethernet".
Read the article; what is actually said is that current products that use broadband (e.g. VOIP providers) will probably not be viable in 5 years time. The /. article is badly worded and misleading, IMO.
Hrm, nice, you've broken the speed of light with ICMP packets; round trip time to Mars would be a number of seconds/minutes....
The users often are the problem; give a user 10 steps to perform to possibly view some naughty pictures of a celebrity and chances are, a significant proportion of them will do so and infect their computer in the process. Heck, some of them would probably run it as root/admin if you asked them to...
IE popups are a standard part of web browsers and Netscape did exactly the same (until Mozilla popup blockers); most other GUI browsers should allow popups; they are required in a number of places (try installing BMC Patrol). You cannot blame MSIE for that as it was (for a change) following web standards.
Despite all the problems in Windows, you don't need to have an insecure system to be exploited. People are stupid and will do all sorts of silly things on the promise of nude celebrity pictures. Besides, even linux has had security flaws which would allow a worm to spread; the lack of these being widespread is as much to do with the low penetration of linux rather than any inherent security by design.
In summary, I'm waiting a few weeks between sp2 coming out and installing it on my PCs just in case.
No, the pirates have a blatant disregard of the infrastructure by keeping running insecure, unpatched software. Microsoft should not be held responsible for pirates who illegally run unlicensed software.
Yes, linux can be more secure than Windows, but the fact is that over 90% of these zombie PCs could have prevented infection by simply having (a) their firewalls enabled and/or (b) having intelligent users. By default, most linux distros don't come with firewalls enabled either (at least, the last time I checked; I think it's becoming more common for firewalling to be enabled though, as with XP SP2) and as for (b), well, we'll always have stupid users.
Feh, I've given up trying to figure out moderators; I'll get up to +5 for a fairly simple post and get modded down just as quickly for something else. Overall, I tend to hover at the higher end of the scale, but I'm past caring.
If you make it WHATWGWBD you're probably right.