I'm using a PVR-250 right now with great success on a MSI KM2M Combo-L (Via KM266 chipset). Try the latest drivers, they've had a lot of bugs fixed in the last few months which have made the work work a LOT more reliably.
I had a similar Oops with the 5328 Nvidia driver. However, the patches at minion.de didn't compile against 2.4.23 until I commented out the "typedef void irqreturn_t;" on line 409.
But even after that, X fails to start up. Back to the last release which were working OK on this Fedora Core 1 install and 2.4.23 kernel.
While NiMH batteries (and NiCad and to a much lesser extent, Alkaline batteries) will self discharge over the course of a a few weeks to a few months depending on the battery, you can greatly slow down the rate by storing the batteries in the freezer. I keep all my charged up batteries in the freezer sealed in a zip-lock baggie where they will store for a long time without losing much of their charge.
As long as you stick with the pre-forked MPM of Apache2, you really shouldn't have any problems with add-on modules like mod_perl and mod_php. Problems only arise when using one of the threaded MPMs.
IMO, the best reason to use Apache 2.0 is that with mod_deflate, you can now easily add content encoding compression to an entire website to save bandwidth. Previously with Apache 1.3, you could add in mod_gzip, but mod_gzip wouldn't compress SSL content without some very ugly config hacks including mod_proxy with a substantial performance benefit. 2.0 eliminates this issue.
I've seen bandwith drop on websites drop from 20-80% depending on how much content is non-compressible (like graphics).
The latest 2.4.23-pre* has had substantial amount VM updates from Andrea Arcangeli's branch integrated.
You should give that a shot as it has been reported to substantially improve kernel behavior when under VM pressure like when you're doing intensive IO as described in your post.
BTW, out of the tasks you listed most of them are IO bound, not CPU bound. Notable exceptions are bzip2 unless you have a REALLY fast CPU.
...filtering to provide you with a tool which automatically learns who you normally receive emails from and not mark them as SPAM? Bayes appears to be much smarter and adaptable than custom rulesets.
I know that at least SpamAssassin and bogofilter will specifically learn message headers appropriately as you train your filter.
Once you train enough of your spammy eBay notices as HAM, you should not have to worry about having them get marked as SPAM anymore.
I assume you're using Sendmail which has a horrible method of handling short bursts of mail.
I switched to Postfix so that I could get it's much better rate limiting rules. Postfix will limit the number of incoming deliveries to a limit you set. This means you can directly control the load put on your server due to mail processing. By default it limits concurrency to incoming messages to 2 which works out well for a dedicated single CPU mail server.
Much more effective than Sendmail which sends off message for delivery as fast as it can before it notices that the load has shot up. Before you know it you've got 100 spamc processes running and your machine is deep into swap!!!
However, keep in mind that unless the bus speed of your CPU runs at the same 166 the memory does, you may even see a drop in performance!
You often get better performance when the memory and cpu are running the same clock speed as then you don't end up with the CPU waiting on odd-cycles for the memory to pump data.
You should run tests to verify, but hey, at least this is only two settings in stead of 289 or whatever.
A validator [w3.org] would be a better choice. It's a proper syntax checker, not just a linter.
What I would love to have is have something like the http header viewer built in so I don't have to pipe in pages to the validator manually.
Re:I know this book is about software RAID ...
on
Managing RAID on Linux
·
· Score: 2, Informative
Actually, 3ware IDE RAID cards are much cheaper than that. About $120 for a 2-channel card good for raid 1/0 and $245 for a 4-channel, $365 for a 8-channel and $520 for a 12-channel. I pulled these prices from hypermicro.com and no, I'm not affiliated with them, just a satisfied customer.
If you're looking to do any more than 4 channels, I'd take a serious look at the SATA cards and drives simply to reduce cabling hassles.
SpamAssassin 2.50 (due "Real Soon Now") will also offer Bayes filtering so that it will add learning capabilities and individualization to the filtering rules it already has.
I haven't tried it out myself yet, but it promises to greatly reduce the number of false positives and negatives. Initial reports from people running CVS code concur.
The biggest problem with it is that integrating server-side filtering with the client is somewhat difficult. For example at my company, our mail server runs Linux, while most people use Outlook/pop3 as their client. Figuring out a method to categorize emails so that the server can learn from them is not easy. The best method I've thought of so far for these users to have them log into a webmail interface which uses IMAP and move spams which were not caught into a SPAM folder. A bit of extra work for the end user, but it should pay off as SpamAssassin learns and catches higher percentages of SPAM with fewer false positives in the end leading to less work for the end user.
Re:Shania was lip syncing, I'm sure of it.
on
Superbowl XXXVII
·
· Score: 1
It was totally obvious, especially after hearing No Doubt sing after her.
This scheme had nothing to do with online betting. It had to do with the operators running the software which handles the money, odds and wagers behind the scenes at the vast majority of racetracks across the country.
There online horse racing available, but these systems simply tie a web interface into the Autotote system. The Autotote software is where the real vulnerability lies, not the online horse racing sites.
You obviously know nothing about the horse racing industry. While there may be some shady characters out there, most people in the scene are just your average blokes who are hoping to win a couple bets while at the racetrack. Those are the guys who eventually end up losing because of people who cheat the system.
If you're only buying one or two items, they charge a fixed shipping rate per item. It's not quite a fixed shipping rate.
For instance, I put a round IDE cable which listed as $6 shipping and a WAP11 wireless access point which was listed as $10 shipping.
But after calculating the total shipping, it was a bit less at $15.01. Adding a second WAP11 brought the total up to $17 something.
Anyway, the thing to note is that I put the same 2 items into my cart at www.mwave.com and the shipping was only $9.49. And mwave's price on the WAP11 was $5-6 cheaper saving about $10 overall.
Given that the level of service from both mwave and newegg is about the same (I've actually found that mwave tends to get orders shipped sooner) I will usually order from mwave, but always check with newegg as well.
Slashdot Mirror
I'm using a PVR-250 right now with great success on a MSI KM2M Combo-L (Via KM266 chipset). Try the latest drivers, they've had a lot of bugs fixed in the last few months which have made the work work a LOT more reliably.
Oops, mean to say thanks for shitty drivers. At least the old ones work for me.
Yeah, thanks for shitty that don't work for a LOT of people. And good luck figuring out why without the source.
I had a similar Oops with the 5328 Nvidia driver. However, the patches at minion.de didn't compile against 2.4.23 until I commented out the "typedef void irqreturn_t;" on line 409.
But even after that, X fails to start up. Back to the last release which were working OK on this Fedora Core 1 install and 2.4.23 kernel.
Mod parent up, threads are sorted by thread and time in Thunderbird!
Is to try out 2.6.0-test11 which is the latest at this point in time.
While NiMH batteries (and NiCad and to a much lesser extent, Alkaline batteries) will self discharge over the course of a a few weeks to a few months depending on the battery, you can greatly slow down the rate by storing the batteries in the freezer. I keep all my charged up batteries in the freezer sealed in a zip-lock baggie where they will store for a long time without losing much of their charge.
As long as you stick with the pre-forked MPM of Apache2, you really shouldn't have any problems with add-on modules like mod_perl and mod_php. Problems only arise when using one of the threaded MPMs.
IMO, the best reason to use Apache 2.0 is that with mod_deflate, you can now easily add content encoding compression to an entire website to save bandwidth. Previously with Apache 1.3, you could add in mod_gzip, but mod_gzip wouldn't compress SSL content without some very ugly config hacks including mod_proxy with a substantial performance benefit. 2.0 eliminates this issue.
I've seen bandwith drop on websites drop from 20-80% depending on how much content is non-compressible (like graphics).
This would work well on web logs who have lots of users (IE, slashdot), but won't work well for web lots who have few readers.
The latest 2.4.23-pre* has had substantial amount VM updates from Andrea Arcangeli's branch integrated.
You should give that a shot as it has been reported to substantially improve kernel behavior when under VM pressure like when you're doing intensive IO as described in your post.
BTW, out of the tasks you listed most of them are IO bound, not CPU bound. Notable exceptions are bzip2 unless you have a REALLY fast CPU.
It seems that Seattle is considering a tax on expresso.
Yeah, because coffee drinkers should pay extra for preschool!
...filtering to provide you with a tool which automatically learns who you normally receive emails from and not mark them as SPAM? Bayes appears to be much smarter and adaptable than custom rulesets.
I know that at least SpamAssassin and bogofilter will specifically learn message headers appropriately as you train your filter.
Once you train enough of your spammy eBay notices as HAM, you should not have to worry about having them get marked as SPAM anymore.
I assume you're using Sendmail which has a horrible method of handling short bursts of mail.
I switched to Postfix so that I could get it's much better rate limiting rules. Postfix will limit the number of incoming deliveries to a limit you set. This means you can directly control the load put on your server due to mail processing. By default it limits concurrency to incoming messages to 2 which works out well for a dedicated single CPU mail server.
Much more effective than Sendmail which sends off message for delivery as fast as it can before it notices that the load has shot up. Before you know it you've got 100 spamc processes running and your machine is deep into swap!!!
Eco Golf has been making corn based golf tees and balls for ages.
http://www.ecogolf.com/
Do you have DMA on as well?
However, keep in mind that unless the bus speed of your CPU runs at the same 166 the memory does, you may even see a drop in performance!
You often get better performance when the memory and cpu are running the same clock speed as then you don't end up with the CPU waiting on odd-cycles for the memory to pump data.
You should run tests to verify, but hey, at least this is only two settings in stead of 289 or whatever.
Actually, 3ware IDE RAID cards are much cheaper than that. About $120 for a 2-channel card good for raid 1/0 and $245 for a 4-channel, $365 for a 8-channel and $520 for a 12-channel. I pulled these prices from hypermicro.com and no, I'm not affiliated with them, just a satisfied customer.
If you're looking to do any more than 4 channels, I'd take a serious look at the SATA cards and drives simply to reduce cabling hassles.
SpamAssassin 2.50 (due "Real Soon Now") will also offer Bayes filtering so that it will add learning capabilities and individualization to the filtering rules it already has.
I haven't tried it out myself yet, but it promises to greatly reduce the number of false positives and negatives. Initial reports from people running CVS code concur.
The biggest problem with it is that integrating server-side filtering with the client is somewhat difficult. For example at my company, our mail server runs Linux, while most people use Outlook/pop3 as their client. Figuring out a method to categorize emails so that the server can learn from them is not easy. The best method I've thought of so far for these users to have them log into a webmail interface which uses IMAP and move spams which were not caught into a SPAM folder. A bit of extra work for the end user, but it should pay off as SpamAssassin learns and catches higher percentages of SPAM with fewer false positives in the end leading to less work for the end user.
It was totally obvious, especially after hearing No Doubt sing after her.
No-one sings that perfectly live.
If I could consistently (more than 50%) pick the winner of any race I'd be a wealthy man!
To pick 4 winners in a row, either damn lucky or I took a peek into the future!
This scheme had nothing to do with online betting. It had to do with the operators running the software which handles the money, odds and wagers behind the scenes at the vast majority of racetracks across the country.
There online horse racing available, but these systems simply tie a web interface into the Autotote system. The Autotote software is where the real vulnerability lies, not the online horse racing sites.
You obviously know nothing about the horse racing industry. While there may be some shady characters out there, most people in the scene are just your average blokes who are hoping to win a couple bets while at the racetrack. Those are the guys who eventually end up losing because of people who cheat the system.
If you're only buying one or two items, they charge a fixed shipping rate per item.
It's not quite a fixed shipping rate.
For instance, I put a round IDE cable which listed as $6 shipping and a WAP11 wireless access point which was listed as $10 shipping.
But after calculating the total shipping, it was a bit less at $15.01. Adding a second WAP11 brought the total up to $17 something.
Anyway, the thing to note is that I put the same 2 items into my cart at www.mwave.com and the shipping was only $9.49. And mwave's price on the WAP11 was $5-6 cheaper saving about $10 overall.
Given that the level of service from both mwave and newegg is about the same (I've actually found that mwave tends to get orders shipped sooner) I will usually order from mwave, but always check with newegg as well.