I was at Cambridge during the late 90's-early Noughties, and I seem to recall a number of stern warnings to students about bandwidth usage from both College and University computing authorities. One of them even included a plea to use European or British mirrors as much as possible.
That time period coincides with the decision by JANET to charge institutions per gig for use of their transatlantic link, I don't know if they still do. The University were probably passing that bill on to the individual colleges. That's why they included a plea to use mirrors on this side of the pond. Makes perfect sense to me, and the mirrors were usually a heck of a lot faster.
There isn't really a comparison here - the Enabling Law gave dictatoral powers, this act is far from it and would still require a successful vote in the elected Parliment for any law to pass.
To pass a law, yes. To change it afterwards (eg 28 days detention without trial to 90), no. Even to change itself to give more powers. It really is that scary.
What is in question is this new proposed act, that allows any cabinet member to alter any piece of legislation by conducting a single vote with the minimum of debate or discussion.
*No*, that's the *status quo* (almost). The new Act will allow a cabinet member to alter any piece of legistlation *without recourse to parliament*. Ie, without a vote! Read it. Listen to the screams of those who have been attending the backwater committee stages that have been cooking this up. This is an unprecedented move to bypass parliament altogether to punish it for standing in the way of the government's 'reforms', hidden under the cloak of 'deregulation'. Only 'controversial' changes would have to be voted on in parliament, with the ministers themselves deciding what is 'controversial'! If this passes, Jim Murphy's name may well go down in history as the man who killed democracy in the UK.
If you think this sounds like hyperbole, just check it out yourself.
Here in the UK, MS has been running ads with people wearing dinosaur heads making comments like:
"I'm either here for the 11:00 meeting on the 12th or the 12:00 meeting on the 11th"
- Microsoft Office has evolved. Have you?
The thing I don't understand is that all the "problems" the ads show haven't actually existed since around Office '97.
Exactly, because it's Office '97 that new Office (what's it even called now?) is competing against. If you look at some of those adverts, it even has a dinosaur saying "We've got Office 97, is that good enough?" and the other replies "not nearly!". People have been saying for a while that MS's biggest competitor are their own old products, well now we see MS 'fessing up to that. Googling around you find bloggers and commentators annoyed and insulted by the ads. I don't think they're a great idea.
Have a look at zoneminder. It may or may not meet your needs. I've only used it with two rather crappy IP cameras at once. I'm currently using it with an IP camera at the wrong end of an ADSL link and the monitoring server remote. It works for me, and I've barely touched it. Lots of development, lots of features, lots of (professional) users.
"It supports capture, analysis, recording, and monitoring of video data coming from one or more video or network cameras attached to a Linux system. ZoneMinder also support web and semi-automatic control of Pan/Tilt/Zoom cameras using a variety of protocols. It is suitable for use as a home video security system and for commercial or professional video security and surveillance. It can also be integrated into a home automation system via X.10 or other protocols."
Not affiliated in any way, but I *am* about to cut a cheque as a donation because this very morning the alarm monitoring centre called me saying that the (professionally installed, approved) alarm had triggered. I can see from zoneminder that nothing's happened, so I'm not going to panic. Otherwise I'd be racing a 100 miles to check up.
...sorry to followup my own post. More on 'smart cooking' here, including PDF codes you can download for various recipies, print out & swipe at your smart oven for 'perfect cooking'.
I despair sometimes.
Re:Wrong target market.
on
RFID Cookware
·
· Score: 1
The people who would cook by plugging a card into a pot would not be using a pot in the first place. They'd be using a microwave.
And in the UK we already have microwave ovens that can read the cooking instructions and automagically cook your food as intended by the producer.
And, anyway, why do we need lossless for portable players? So far, they don't hook up to SPDIF...
Mine does. Iriver IHP-120 (later the H120). SPDIF in *and* out. Line in & out, ogg, FM radio, excellent sound quality (better than any iPod available at the time - I haven't seen later generations compared). Iriver didn't carry the SPDIF on to the later generations - I guess they'd over-egged the IHP-120, and didn't find it worth keeping. I predict it'll reappear on HDD MP3 players in about a year's time as they struggle to add features to distinguish themselves from the competition. I'm just happy I bought one when I did.
Unless you want to do something useful. Like watch more than OTA or non-digital programming.
In the UK (and lots of countries with DVB - digital terrestrial broadcasting) you can simply use a DVB card and hey presto - dozens of digital channels on your Myth box without any blaster nonsense.
Does anyone else get that thing where you're about to fall asleep, but you feel as if you're freefalling; and then you have a kneejerk reaction and jerk awake violently?
I'll look into those Chaparral and Dot Hill controllers. Anything significantly more robust than a PERC 3/DC might well be worth investigating in.
I don't think that either make host controllers - I've only seen them in enclosures (ie external RAID units). We have them in Fortra enclosures. They fly. I've benchmarked a software RAID0 of two RAID5 arrays of 6 disks, each array connected on its own scsi channel at 200MB/s sustained - and this was at least two years ago. Having googled, I see that Dot Hill have bought Chaparral! Sun's storedge 3000 series is OEMed from them, but Dot Hill also OEM stuff from Infortrend, so who actually makes what is a bit confusing.
As to the so called Strawman. The experiment I described would actually simulate the incident the original poster described.
Yabbut I wasn't replying to the OP! Nevermind!
I do have some experience doing Lab setups prior to deploying enterprise system (I hate that buzzword too). I have also had to simulate failures that happened for real in the field.
I'm sorry to say that my worst failure (pulling a drive from a degraded array) was due to a mislabeled enclosure and failing to double-check by lighting the drive-id LED.
I live in Jamaica and the way to get something critical delivered in less than 2 days guaranteed is to send an employee to the vendors office in Miami to pick up the item and then board a plane with it
Well we're in Aberystwyth, mid-west Wales, on the west coast. Suppliers often say "OK, we'll overnight that to you and you'll have it by 9am", and I have to gently point out that couriers seem to get mugged somewhere in the mountains on the way, since nothing has *ever* got to us before 11am, other than the regular mail (go Royal Mail!) and some Sun engineers who've actually tried the journey before.
3Ware SATA controllers handle drive re-ordering just fine. Even on degraded arrays.
Good to know. I've never used IDE nor SATA RAID (well, not counting the Uli in the shuttle on the corner of my desk which I haven't managed to settle on a distro for yet) but I was surprised with the suggestion that the controllers depended on drive order to determine array membership.
On a related note... stay away from Seagate 7200.8 300GB drives.
SATA? I'll ask our admins what we've got in our SATA enclosures. We've got some 300GB SCSI, but I'm not sure what makes. I used to like seagate a lot for SCSI. IBM's ultrastars were my bugbears (no, not the same problem as the deskstars!).
No thanks, that strawman you're plaguing isn't really fighting back.
1. Pull a disk from a RAID 5 on your "enterprise controller".
...but first I'll stop here and address the snideness. OK, you got me, I don't really like 'enterprise' as a tech adjective either, but it was useful to differentiate from what passes as RAID in the home market.
2. Wait for the machine to recognize this failure. 3. Initiate some activity on the array. Something with lots of reads and writes. 4. Randomly reshuffle drives.
If you can do that without any data loss or downtime. Consistently.
I'm sorry, I seem to be in the wrong thread. I can't remember claiming to be able to do any of that. HINT - I simply stated that other-than-IDE-SATA (since you didn't like "enterprise") RAID controllers write array config to disk so that disk shuffling won't leave you with a confused controller. I didn't mention downtime.
I don't think we're disagreeing here, and the MegaRAID (AMI, LSI, some PERC) & Adaptec (some PERC) controllers aren't my favourite either, but I've had a disk pulled from a degraded RAID5 array during disk activity (I couldn't swear it was writing, but then I'm not in the business of aiming for mobile goalposts) and had it recovered. Then again, I've lost entire partitions on the same controllers without disk-pulling tomfoolery.
Chaparral make (made? haven't bought one for a while) good controllers, I've been fairly happy with those. The oob management on the controllers isn't the best (although that could be my system manufacturer's fault), but they're better at recovering from disasters than the Intel, Adaptec & LSI ones which I've used. Dot Hill's stuff is good (as seen in some of the Surestore arrays).
Not that I was replying to the original post, but yes, better than you did, it seems.
The RAID already had a failed disk which means that if it was a RAID 5 it would be operating as RAID 0. The disk shuffle was started in that scenario with the machine running. Pull a disk from an active RAID 0 and game over....until you put the disks back in (in whatever order) and depending on your controller, you may be hot to trot, or you may have to reset it. You certainly don't scratch your head for a week if you have a halfway decent controller, which was my point.
The post I was replying to didn't seem to know that controllers write array configurations to disk, and that shuffling the disks isn't the end of the world.
say you have 6 drives, with 3 arrays. how in hell is your controller going to know what you were intending to happen if you remove the drives and randomly insert them in different positions?
Erm, like every enterprise level (SCSI, FC) RAID controller I've used, it would read the config off the disks and carry on happily. Some will tell you that the disk configuration is different from the NVRAM configuration stored on the controller, and ask you which to trust, but shuffling the drives isn't a big problem. Relying on the controller alone to remember the array configuration is bonkers - what if your controller dies?
That's odd... short of saving at 16bits/channel (which maybe Photoshop can't do - I don't touch that thing), I don't think there are any quality settings with PNG per se; it is a lossless format, after all.
Now, having said that, the Slashdot interface clearly leaves something to desire, even when using Lynx. Why is there no true modular command line interface? I would think of something along the lines of
Slashdot doesn't include the comments in RSS (probably a good thing), but you can get your stories nice and easy. I just tried it after reading your message. I'd have preferred vi keybindings, but you can change them in a rc file. There's a bunch of extension scripts (including some Slashdot ones). JFYI.
Keyboard shortcuts randomly stop working. Command-W being the most obvious, as you go to close a tab or window...and nooooothing happens.
I get that a lot with FF on linux - it seems to be a lot to do with what's under the pointer (I'm on a laptop with a touchpad, so the pointer tends to be left anywhere on the screen) if there's a flash ad/image I have to move the pointer from over it to make it work. Totally reproducible. Try it.
for a streaming server, you may want to consider a streaming server like VideoLAN's VLS.:-)
You'd think so, wouldn't you, but not for a while now! Videolan themselves recommend VLC for streaming, rather than VLS, which hasn't been much developed recently. The streaming stuff has been folded into VLC (plus lots more!).
Coo. I've seen ssh session cope with me restarting local network connections, but I didn't think they'd survive cycling the remote sshd process. I suppose it might depend on what runlevels you have sshd running in.
Don't listen to me - that'll teach me to post late at night. I missed the context of an runlevel change - they'll survive restarting the network service, but I'm unsure about ssh... let me try...
Oh yes, it does work. I can init 3 & init 5 and my ssh session doesn't blink.
Slashdot Mirror
I can't believe that nobody gave up a "Funny" mod for a clever HTML blink-tag joke.
Who is this, and what did you people do with the real slashdot?
Let's face it, the kids don't even remember the millenium ticking over, let alone usenet and HTML 3 jokes.
This place was a lot better when it was org!slashdot
I was at Cambridge during the late 90's-early Noughties, and I seem to recall a number of stern warnings to students about bandwidth usage from both College and University computing authorities. One of them even included a plea to use European or British mirrors as much as possible.
That time period coincides with the decision by JANET to charge institutions per gig for use of their transatlantic link, I don't know if they still do. The University were probably passing that bill on to the individual colleges. That's why they included a plea to use mirrors on this side of the pond. Makes perfect sense to me, and the mirrors were usually a heck of a lot faster.
There isn't really a comparison here - the Enabling Law gave dictatoral powers, this act is far from it and would still require a successful vote in the elected Parliment for any law to pass.
To pass a law, yes. To change it afterwards (eg 28 days detention without trial to 90), no. Even to change itself to give more powers. It really is that scary.
What is in question is this new proposed act, that allows any cabinet member to alter any piece of legislation by conducting a single vote with the minimum of debate or discussion.
*No*, that's the *status quo* (almost). The new Act will allow a cabinet member to alter any piece of legistlation *without recourse to parliament*. Ie, without a vote! Read it. Listen to the screams of those who have been attending the backwater committee stages that have been cooking this up. This is an unprecedented move to bypass parliament altogether to punish it for standing in the way of the government's 'reforms', hidden under the cloak of 'deregulation'. Only 'controversial' changes would have to be voted on in parliament, with the ministers themselves deciding what is 'controversial'! If this passes, Jim Murphy's name may well go down in history as the man who killed democracy in the UK.
If you think this sounds like hyperbole, just check it out yourself.
Its not like I'll be ebaying on an 1 1/2" screen... Am I missing something?
Maybe. The UK press have been running ads for 'ebay on your mobile' for a couple of weeks now. Google says: Monitor Ebay on Your Mobile
Also qicture
Here in the UK, MS has been running ads with people wearing dinosaur heads making comments like:
"I'm either here for the 11:00 meeting on the 12th or the 12:00 meeting on the 11th"
- Microsoft Office has evolved. Have you?
The thing I don't understand is that all the "problems" the ads show haven't actually existed since around Office '97.
Exactly, because it's Office '97 that new Office (what's it even called now?) is competing against. If you look at some of those adverts, it even has a dinosaur saying "We've got Office 97, is that good enough?" and the other replies "not nearly!". People have been saying for a while that MS's biggest competitor are their own old products, well now we see MS 'fessing up to that. Googling around you find bloggers and commentators annoyed and insulted by the ads. I don't think they're a great idea.
Have a look at zoneminder. It may or may not meet your needs. I've only used it with two rather crappy IP cameras at once. I'm currently using it with an IP camera at the wrong end of an ADSL link and the monitoring server remote. It works for me, and I've barely touched it. Lots of development, lots of features, lots of (professional) users.
"It supports capture, analysis, recording, and monitoring of video data coming from one or more video or network cameras attached to a Linux system. ZoneMinder also support web and semi-automatic control of Pan/Tilt/Zoom cameras using a variety of protocols. It is suitable for use as a home video security system and for commercial or professional video security and surveillance. It can also be integrated into a home automation system via X.10 or other protocols."
Not affiliated in any way, but I *am* about to cut a cheque as a donation because this very morning the alarm monitoring centre called me saying that the (professionally installed, approved) alarm had triggered. I can see from zoneminder that nothing's happened, so I'm not going to panic. Otherwise I'd be racing a 100 miles to check up.
...sorry to followup my own post. More on 'smart cooking' here, including PDF codes you can download for various recipies, print out & swipe at your smart oven for 'perfect cooking'.
I despair sometimes.
The people who would cook by plugging a card into a pot would not be using a pot in the first place. They'd be using a microwave.
And in the UK we already have microwave ovens that can read the cooking instructions and automagically cook your food as intended by the producer.
And, anyway, why do we need lossless for portable players? So far, they don't hook up to SPDIF...
Mine does. Iriver IHP-120 (later the H120). SPDIF in *and* out. Line in & out, ogg, FM radio, excellent sound quality (better than any iPod available at the time - I haven't seen later generations compared). Iriver didn't carry the SPDIF on to the later generations - I guess they'd over-egged the IHP-120, and didn't find it worth keeping. I predict it'll reappear on HDD MP3 players in about a year's time as they struggle to add features to distinguish themselves from the competition. I'm just happy I bought one when I did.
Next came 78's. These were cast in a mold and made of the miracle plastic bakelite.
Shellac, mainly.
Unless you want to do something useful. Like watch more than OTA or non-digital programming.
In the UK (and lots of countries with DVB - digital terrestrial broadcasting) you can simply use a DVB card and hey presto - dozens of digital channels on your Myth box without any blaster nonsense.
Does anyone else get that thing where you're about to fall asleep, but you feel as if you're freefalling; and then you have a kneejerk reaction and jerk awake violently?
Myoclonic jerks.
I'll look into those Chaparral and Dot Hill controllers. Anything significantly more robust than a PERC 3/DC might well be worth investigating in.
I don't think that either make host controllers - I've only seen them in enclosures (ie external RAID units). We have them in Fortra enclosures. They fly. I've benchmarked a software RAID0 of two RAID5 arrays of 6 disks, each array connected on its own scsi channel at 200MB/s sustained - and this was at least two years ago. Having googled, I see that Dot Hill have bought Chaparral! Sun's storedge 3000 series is OEMed from them, but Dot Hill also OEM stuff from Infortrend, so who actually makes what is a bit confusing.
As to the so called Strawman. The experiment I described would actually simulate the incident the original poster described.
Yabbut I wasn't replying to the OP! Nevermind!
I do have some experience doing Lab setups prior to deploying enterprise system (I hate that buzzword too). I have also had to simulate failures that happened for real in the field.
I'm sorry to say that my worst failure (pulling a drive from a degraded array) was due to a mislabeled enclosure and failing to double-check by lighting the drive-id LED.
I live in Jamaica and the way to get something critical delivered in less than 2 days guaranteed is to send an employee to the vendors office in Miami to pick up the item and then board a plane with it
Well we're in Aberystwyth, mid-west Wales, on the west coast. Suppliers often say "OK, we'll overnight that to you and you'll have it by 9am", and I have to gently point out that couriers seem to get mugged somewhere in the mountains on the way, since nothing has *ever* got to us before 11am, other than the regular mail (go Royal Mail!) and some Sun engineers who've actually tried the journey before.
3Ware SATA controllers handle drive re-ordering just fine. Even on degraded arrays.
... stay away from Seagate 7200.8 300GB drives.
Good to know. I've never used IDE nor SATA RAID (well, not counting the Uli in the shuttle on the corner of my desk which I haven't managed to settle on a distro for yet) but I was surprised with the suggestion that the controllers depended on drive order to determine array membership.
On a related note
SATA? I'll ask our admins what we've got in our SATA enclosures. We've got some 300GB SCSI, but I'm not sure what makes. I used to like seagate a lot for SCSI. IBM's ultrastars were my bugbears (no, not the same problem as the deskstars!).
No thanks, that strawman you're plaguing isn't really fighting back.
1. Pull a disk from a RAID 5 on your "enterprise controller".
2. Wait for the machine to recognize this failure.
3. Initiate some activity on the array. Something with lots of reads and writes.
4. Randomly reshuffle drives.
If you can do that without any data loss or downtime. Consistently.
I'm sorry, I seem to be in the wrong thread. I can't remember claiming to be able to do any of that. HINT - I simply stated that other-than-IDE-SATA (since you didn't like "enterprise") RAID controllers write array config to disk so that disk shuffling won't leave you with a confused controller. I didn't mention downtime.
I don't think we're disagreeing here, and the MegaRAID (AMI, LSI, some PERC) & Adaptec (some PERC) controllers aren't my favourite either, but I've had a disk pulled from a degraded RAID5 array during disk activity (I couldn't swear it was writing, but then I'm not in the business of aiming for mobile goalposts) and had it recovered. Then again, I've lost entire partitions on the same controllers without disk-pulling tomfoolery.
Chaparral make (made? haven't bought one for a while) good controllers, I've been fairly happy with those. The oob management on the controllers isn't the best (although that could be my system manufacturer's fault), but they're better at recovering from disasters than the Intel, Adaptec & LSI ones which I've used. Dot Hill's stuff is good (as seen in some of the Surestore arrays).
Did you read the original post ?
...until you put the disks back in (in whatever order) and depending on your controller, you may be hot to trot, or you may have to reset it. You certainly don't scratch your head for a week if you have a halfway decent controller, which was my point.
Not that I was replying to the original post, but yes, better than you did, it seems.
The RAID already had a failed disk which means that if it was a RAID 5 it would be operating as RAID 0.
The disk shuffle was started in that scenario with the machine running. Pull a disk from an active RAID 0 and game over.
The post I was replying to didn't seem to know that controllers write array configurations to disk, and that shuffling the disks isn't the end of the world.
say you have 6 drives, with 3 arrays. how in hell is your controller going to know what you were intending to happen if you remove the drives and randomly insert them in different positions?
Erm, like every enterprise level (SCSI, FC) RAID controller I've used, it would read the config off the disks and carry on happily. Some will tell you that the disk configuration is different from the NVRAM configuration stored on the controller, and ask you which to trust, but shuffling the drives isn't a big problem. Relying on the controller alone to remember the array configuration is bonkers - what if your controller dies?
That's odd... short of saving at 16bits/channel (which maybe Photoshop can't do - I don't touch that thing), I don't think there are any quality settings with PNG per se; it is a lossless format, after all.
There's a lot you can to optimize PNGs, for example with pngcrush. See Wikipedia's detail on PNG filesizes. For more info on PS's bad handling of PNG, see Photoshop & PNG about halfway down that page.
Now, having said that, the Slashdot interface clearly leaves something to desire, even when using Lynx. Why is there no true modular command line interface? I would think of something along the lines of
How's about a command-line RSS reader?
http://kiza.kcore.de/software/snownews
Slashdot doesn't include the comments in RSS (probably a good thing), but you can get your stories nice and easy. I just tried it after reading your message. I'd have preferred vi keybindings, but you can change them in a rc file. There's a bunch of extension scripts (including some Slashdot ones). JFYI.
Dark alley corners are:
I get that a lot with FF on linux - it seems to be a lot to do with what's under the pointer (I'm on a laptop with a touchpad, so the pointer tends to be left anywhere on the screen) if there's a flash ad/image I have to move the pointer from over it to make it work. Totally reproducible. Try it.
for a streaming server, you may want to consider a streaming server like VideoLAN's VLS. :-)
You'd think so, wouldn't you, but not for a while now! Videolan themselves recommend VLC for streaming, rather than VLS, which hasn't been much developed recently. The streaming stuff has been folded into VLC (plus lots more!).
Your analogy is good except for the advertising involved with music.
And your post is good except this is about Sony's electronic products, not music.
Sheesh. RTFA. +4 Insightful??
Coo. I've seen ssh session cope with me restarting local network connections, but I didn't think they'd survive cycling the remote sshd process. I suppose it might depend on what runlevels you have sshd running in.
Don't listen to me - that'll teach me to post late at night. I missed the context of an runlevel change - they'll survive restarting the network service, but I'm unsure about ssh... let me try...
Oh yes, it does work. I can init 3 & init 5 and my ssh session doesn't blink.
Same difference... you've still shutdown all your network services which to the users means you've had downtime. It's a reboot in all but name.
mmm... I can see that in a few specific cases, like if you have a lot of users who log on over ssh.
ssh sessions survive a linux network reset, unless you've some seriously funky network hardware. Try it!