As a Briton living in the USA (resident alien), I do not get to vote. Strangely, a rather large chunk of my pay is docked for these things that look suspiciously like "taxes". Let's see... "Federal Income Tax", "Medicare Tax", "CA State Income Tax", "Social Security Tax"... oh, and they're under a heading, "Statutory". Hrm.
Please, tell me more of your "No taxation without respresentation", this is an intriguing concept. I wish to drink of your kool-aid.
I didn't read it because I didn't reload the thread after getting back from work issues and eventually finding^Wresetting the password for my account and then going looking for other updates in the meantime. My mistake. The points I raised remain valid -- the paid support is claimed to not have gotten them far, but the people providing the support are confused over who this could be. A paying customer would, after contacting Support, indeed have been told about the performance issues and the options available.
And no, after a grand total of 4 comments in three and a half years, I'm clearly not that interested in my Slashdot rating. I know because of your accusations that someone had rated me informative. How nice. How sad that you're unable to conceive of someone else disagreeing with you, so that any other assessment must automatically be faked.
For clarity and the record, no I've not used sock puppets. I had to think for a moment to figure out what you meant. Thanks for reminding me of why I rarely bother with Slashdot.
There's two responses to this. One is that if you have 90,000+ zones you're presumably making some money from this so can afford to pay for support from ISC -- how far did this get you? Or were you using the free software and not contributing anything back and then going on a rant about the sucky software that you use for nothing and then make money from?
The other response is to point to the ISC feedback in the CERT Advisory:
"ISC is also making beta releases, BIND 9.5.1b1 and 9.4.3b2 available for download and testing. These beta releases provide the same improved resiliency as the patches but with better performance for servers with query volumes at or above 10,000 queries per second. They are however betas, not fully tested production releases. The patches,(P1 versions), are fully tested today and released for production use."
Of course, a paying customer would presumably know this after seeking their paid support and being told the options available.
IPv4's IP checksum covers the TTL so needs to be recalculated at every router. IPv6 was designed to be streamlined and more efficiently processed by routers; no checksum recalculations, improved option handling, etc.
There are many other improvements too. Buy/borrow the book reviewed above for more information.;-)
SATA scares me a little, simply because Native Command Queuing doesn't allow the OS to impose any kind of ordering constraints on the commands, the way that Tagged Command Queuing does.
So, as I see it, in a power-failure, all that hard work done on ensuring that a modern file-system is always consistent just goes out of the window: meta-data updates happening before the data itself is updated. In fact, that's likely to happen quite a bit with many writes to separate files, since the metadata's slightly more likely to be grouped together.
However, I don't know how this works out in practice, as opposed to theory from reading what the features provide. Do any OS/FS combinations actually support enabling even TCQ in a safe manner, instead of just a potentially unstable performance boost? If so, recommendations on an FS choice gratefully received (please, specific to this issue, not the usual/. FS holy wars).
Disable NCQ and get a major performance hit? Or pick a platform (SCSI, ATA4) which supports TCQ and an OS/FS which then uses this intelligently for meta-data handling?
Sure, there's UPS power and decent SAN storage arrays take battery-backup to decent quality levels, but with each network link being a point of failure the next time some monkey is let loose on "unrelated" cabling issues, and the general flakiness of data-center UPS kit not quite managing to perform as advertised, having SATA with NCQ in a reliable environment leaves me apprehensive and I really don't want to be responsible for a Supported System relying upon SATA/NCQ in a separate box from the box which has the file-system drivers in it.
A Windows build of the openssl suite is essential, if for nothing else than for the checksum tools. Given the tendency of the network drivers to corrupt data, getting an intact copy of a large file restored to a machine when reinstalling is hit and miss. Try:
openssl sha1 filename
This is one of the very first pieces of software to get put on my wife's machine when I reinstall, just so that I can finish the reinstall.
Slashdot Mirror
As a Briton living in the USA (resident alien), I do not get to vote. Strangely, a rather large chunk of my pay is docked for these things that look suspiciously like "taxes". Let's see ... "Federal Income Tax", "Medicare Tax", "CA State Income Tax", "Social Security Tax" ... oh, and they're under a heading, "Statutory". Hrm.
Please, tell me more of your "No taxation without respresentation", this is an intriguing concept. I wish to drink of your kool-aid.
I didn't read it because I didn't reload the thread after getting back from work issues and eventually finding^Wresetting the password for my account and then going looking for other updates in the meantime. My mistake. The points I raised remain valid -- the paid support is claimed to not have gotten them far, but the people providing the support are confused over who this could be. A paying customer would, after contacting Support, indeed have been told about the performance issues and the options available.
And no, after a grand total of 4 comments in three and a half years, I'm clearly not that interested in my Slashdot rating. I know because of your accusations that someone had rated me informative. How nice. How sad that you're unable to conceive of someone else disagreeing with you, so that any other assessment must automatically be faked.
For clarity and the record, no I've not used sock puppets. I had to think for a moment to figure out what you meant. Thanks for reminding me of why I rarely bother with Slashdot.
There's two responses to this. One is that if you have 90,000+ zones you're presumably making some money from this so can afford to pay for support from ISC -- how far did this get you? Or were you using the free software and not contributing anything back and then going on a rant about the sucky software that you use for nothing and then make money from?
The other response is to point to the ISC feedback in the CERT Advisory:
"ISC is also making beta releases, BIND 9.5.1b1 and 9.4.3b2 available for download and testing. These beta releases provide the same improved resiliency as the patches but with better performance for servers with query volumes at or above 10,000 queries per second. They are however betas, not fully tested production releases. The patches,(P1 versions), are fully tested today and released for production use."
Of course, a paying customer would presumably know this after seeking their paid support and being told the options available.
Yes, latency improvements come with IPv6.
;-)
IPv4's IP checksum covers the TTL so needs to be recalculated at every router. IPv6 was designed to be streamlined and more efficiently processed by routers; no checksum recalculations, improved option handling, etc.
There are many other improvements too. Buy/borrow the book reviewed above for more information.
SATA scares me a little, simply because Native Command Queuing doesn't allow the OS to impose any kind of ordering constraints on the commands, the way that Tagged Command Queuing does.
/. FS holy wars).
So, as I see it, in a power-failure, all that hard work done on ensuring that a modern file-system is always consistent just goes out of the window: meta-data updates happening before the data itself is updated. In fact, that's likely to happen quite a bit with many writes to separate files, since the metadata's slightly more likely to be grouped together.
However, I don't know how this works out in practice, as opposed to theory from reading what the features provide. Do any OS/FS combinations actually support enabling even TCQ in a safe manner, instead of just a potentially unstable performance boost? If so, recommendations on an FS choice gratefully received (please, specific to this issue, not the usual
Disable NCQ and get a major performance hit? Or pick a platform (SCSI, ATA4) which supports TCQ and an OS/FS which then uses this intelligently for meta-data handling?
Sure, there's UPS power and decent SAN storage arrays take battery-backup to decent quality levels, but with each network link being a point of failure the next time some monkey is let loose on "unrelated" cabling issues, and the general flakiness of data-center UPS kit not quite managing to perform as advertised, having SATA with NCQ in a reliable environment leaves me apprehensive and I really don't want to be responsible for a Supported System relying upon SATA/NCQ in a separate box from the box which has the file-system drivers in it.
Or am I smoking crack?